Geological development of eastern Indonesia and the northern Australia collisionzone: A review

Peter W. Baillie1, Tom H. Fraser2, Robert Hall3 and Keith Myers4

Keywords: Banda Arc, eastern Indonesia, Timor Trough, Tanimbar Trough, Aru Trough

• Relatively rigid lithosphere of the northern margin ofthe Australian craton.

• Deepwater troughs, including the Java Trench and itsbathymetric extensions the Timor, Tanimbar, Aru andSeram troughs.

• The Banda Arc comprising an Outer Banda Arc(including the islands of Timor, Tanimbar and Seram)and the volcanic arc of the Inner Banda Arc and theWeber Deep.

• The island of New Guinea comprising (a) West PapuaCentral Range fold belt, (b) the West Papua forelandlying south of the fold belt, (c) the Bird’s Headmicrocontinent, and (d) Cenderawasih Bay and the“Bird’s Neck”.

In general terms, the geological evolution of the regionmay be considered in terms of:

• building blocks formed during Proterozoic andPalaeozoic times;

• development of a Mesozoic passive margin; and• Neogene collision of Australia and the Indonesian

Sunda archipelago.

The first two phases occurred when the region was partof Gondwanaland.

While the geological literature on both the northernmargin of Australia (essentially the North West Shelf) andSouth East Asia (including East Indonesia) is voluminous,it is essentially endemic and the overlap of political andgeological boundaries has tended to mask the commongeology between the two regions (Longley et al., 2002).In this paper, we review the geology of the collision zonebetween the Australian craton and the Indonesianarchipelago – one of the most geologically-complex areason Earth.

Much useful information is found in the Journal ofAsian Earth Sciences (formerly the Journal of SoutheastAsian Earth Sciences), the AAPG Bulletin, specialistpublications of The Geological Society together withconference proceedings published by the PetroleumExploration of Australia (PESA), the Australian PetroleumProducing and Exploration Association (APPEA), theIndonesia Petroleum Association (IPA) and the SouthEas t As i a Pe t ro l eum Exp lo ra t i on Assoc i a t i on(SEAPEX).

The seismic database available to us includes severalnon-exclusive surveys acquired by TGS-NOPECGeophysical Company and/or WesternGeco, including EastIndonesia regional surveys EIR98 and EIR99, together withthe GP-ARI and Matahari surveys acquired near theAustralian border (Fig. 1).

Abstract

For much of its history, the region encompassing thenorthern margin of the Australian Plate and easternIndonesia has been part of, or relatively close to acontinental margin.

Three significant periods of Palaeozoic–Mesozoicterrane dispersion occurred, where continentalfragments separated from the margin of Gondwanaland,drifted northward and were subsequently accreted to theEurasian craton. Separation of these blocks in the EarlyDevonian, late Early Permian and Late Triassic–LateJurassic was accompanied by the opening of the Palaeo-Tethys, Meso-Tethys and Ceno-Tethys ocean basins.

By the end of the Jurassic, the northern margin ofthe Australian continent was part of a passive marginfacing an open ocean containing several continentalblocks or microcontinents.

The present geology results from two tectonic forces:(1) northward movement of the Australian Plate drivenby the Southern Ocean spreading centre at 9 cm/year,and (2) rotation of the Pacific Plate which sets up sinistralshear on the northern margin of New Guinea.

This paper summarises the geological evolution as:building blocks formed during Proterozoic andPalaeozoic times; development of a Mesozoic mid-latitudepassive margin on the eastern margin of Gondwanaland;and Neogene collision of Australia and the Indonesianarchipelago.

Introduction

Eastern Indonesia lies within a complex tectonic zone formedas a result of Neogene collision and interaction of theAustralian and Eurasian (Sundaland) plates, and the Carolineand Philippine Sea oceanic plates, and the Pacific Plate. Theregion has been likened to a giant jigsaw puzzle, consistingof a complex of small ocean basins separated by slivers orfragments of sometimes-thickened continental crust, whichseem ultimately destined to form part of a single complexterrane (Milsom, 1991; Metcalfe, 1998).

The active tectonics of the region have long beenrecognised and are clearly expressed in its morphology,volcanicity and seismicity (e.g. Hamilton, 1979; Veevers,2000). The following morphotectonic features arerecognised:

1 TGS-NOPEC Geophysical Company, Level 5, 1100 Hay StreetWest Perth, WA, 6005. Australia. Email: peterb@tgsnopec.com.au

2 Resource System Diagnostics, Jakarta, Indonesia3 SE Asia Research Group, Royal Holloway University of London,

United Kingdom4 WesternGeco, Kuala Lumpur, Malaysia

540 Geological development of eastern Indonesia and northern Australia collision zone

Figure 1. Locality map showing seismic databaseused in this study. Letters A-D indicate positionof seismic lines shown on Figure 7.

Proterozoic and Palaeozoic building blocks

Proterozoic and Lower Palaeozoic rocks form the effectivebasement of the study area and are presently found near themargin of the Australian Craton (Arafura Platform) and theBird’s Head of West Papua. Although the early tectonichistory is poorly defined, the overall north–south structuralgrain of the basement rocks of the Arafura Platform is wellillustrated by satellite gravity data (Fig. 2) where severalkilometres of gently-deformed Lower Palaeozoic andProterozoic section is present.

By the early Palaeozoic, the region was part of the easternmargin of Gondwanaland (Metcalfe, 1998).

The Goulburn Graben (Fig. 3) developed within theProterozoic basement during the early Palaeozoic (Bradshawet al., 1990; McLennan et al., 1990). This northwest-trending graben is asymmetric, with major faultsdeveloped along its northern boundary and a Palaeozoic tolower Mesozoic fill. Similar intracratonic rifts include theBarakan Basin (Barber et al., 2003), Petrel Sub-basin(Bonaparte Basin) and the Fitzroy Trough (Canning Basin;McLennan et al., 1990).

Gondwanaland tectonic development

During Palaeozoic and Mesozoic times, three significant periodsof terrane dispersion occurred, where continental fragmentsseparated from the margin of Gondwanaland, drifted northwardand were subsequently accreted to the Eurasian craton(Hutchison, 1989). Separation of these blocks in the EarlyDevonian, late Early Permian and Late Triassic–Late Jurassicwas accompanied by the opening of the Palaeo-Tethys andMeso-Tethys ocean basins (Fig. 4; Metcalfe, 1998).

The first important tectonic event was the Early Devonianrifting of South China, North China, Indochina, Tarim andQaidam from Gondwanaland and the subsequent formationof the Palaeo-Tethys Ocean (Fig. 4b; Metcalfe, 1998).

The second event was the Carboniferous to Permianrifting of Sibumasu and Qiangtiang as part of the Cimmeriancontinent, now preserved in parts of Central and SoutheastAsia (Fig. 4c; Metcalfe, 1998). On the North West Shelfthis important rifting event gave rise to the Westralian Super-basin(Bradshaw et al., 1988). It relates to the onset of the Sibamasublock separation and resulted in a thick (10 km) continuousfill of mainly Permian and Mesozoic sediments, covering the

Baillie et al. 541

Figure 2. Regional satellite gravity map (source –http://topex.ucsd.edu/marine_grav/mar_grav.html)

entire Westralian Super-basin (Longley et al., 2002). LateCarboniferous uplift, erosion and local folding in the BonaparteBasin have been related to this event (Baillie et al., 1994).

The Late Triassic Carnian to Norian succession of theNorth West Shelf was deposited following a regionallyextensive period of significant tectonism, erosion and upliftalong the edges of the craton (Fitzroy Movement), relatedeither to breakup events along the Gondwanan margin or todocking of continental blocks along the West Papua/PNGsubduction margin (Longley et al., 2002).

The third significant period of terrane dispersal from theGondwanaland margin occurred in the Late Triassic to LateJurassic, associated with the Late Triassic drift of the Lhasablock and the subsequent drift of the West Burma and Woylablocks in the Late Jurassic (Metcalfe, 1998). Longley et al.(2002) recognized the separation of three West Burma sub-blocks: Block I began to rift in the latest Hettangian andextension continued until breakup in the Sinemurian; riftingof Block II subsequently began in the Callovian and wascomplete by the Oxfordian; and Block III separated from aposition outboard of the Bonaparte Basin in the Tithonian.It is possible that a fourth block occupied the position of thepresent Banda Sea.

Late Jurassic continental breakup along the Argo AbyssalPlain, followed by earliest Cretaceous breakup along the rest ofthe margin, brought the first deep-marine conditions into closeproximity to the continental-edge basins (Baillie et al., 1994).

Mesozoic passive margin

Data from the Mesozoic exposures in eastern Indonesiarange from mapping traverses led by BMR/AGSO

personnel during the 1980s to a limited amount ofsubsurface data from seismic and well-bore penetrations.The Upper Palaeozoic and Mesozoic stratigraphy ofsoutheast Indonesia is a continuation of the northernAustralian rift-drift sequence, and was controlled byGondwanaland breakup and the subsequent opening of theIndian Ocean (Fig. 5, modified after Pairault, 2002, originallysourced from Pigram and Panggabean 1981,1983 and Fraseret al., 1993). Three major tectonic stages are recognised,similar in broad character to the North West Shelf:

• A Permian to Late Triassic pre-rift stage, representedby shallow marine and fluvio-deltaic deposition on astable platform.

• A Late Triassic to Early Jurassic break-up stagecharacterised by block faulting, uplift and thermaldoming, and followed by erosion and accumulation ofcontinental red beds.

• Post-rift from the Middle Jurassic when the region startedto cool and subside, leading to the deposition of restrictedmarine sediments until rising sea level in the Cretaceousled to open marine conditions. A thick carbonate successionwas deposited during the Tertiary (New Guinea Limestone).

Fraser et al. (1993) created a stratigraphy for westernWest Papua based on depositional cycles:

• The early Toarcian to late Bajocian InanwatanPolysequence (present in the Agung-1, CS-1, TBF-1 andGunung-1 wells) was deposited in a nearshore setting,progressively passing into fluvial and lacustrineenvironments.

0°S

5°S

10°S

126°E 130°E 135°E

542 Geological development of eastern Indonesia and northern Australia collision zone

Figure 4. Regional palaeogeographic reconstructions for (a) Mid to Late Silurian at 420 Ma, (b) Late Devonian at 360 Ma, (c) Late Permian at 255 Ma and (d) Late Triassic at 220 Ma, showing relative position of East and Southeast Asian terranes and distribution of land and sea (modified after Metcalfe, 1998). NC = North China, SC = South China, T = Tarim, I = Indochina, Q = Qiangtang, L = Lhasa, S = Sibumasu, WB = West Burma, GI = Greater India.

k

Figure 3. Regional tectonic setting (compiled from various sources with some new interpretation).

Baillie et al. 543

CE

NO

ZOIC

PLEIST.

MIO

CE

NE

OLI

GO

-E

OC

EN

E

PLIOCENE

ER

A

EPOCH

LATE

LATE

EA

RLY

MID

DLE

TRIA

SS

IC

ME

SO

ZOIC

PALA

EO

ZOIC

MID.

EAR.

JUR

AS

SIC

CR

ETA

CE

OU

S

EA

RLY

LATE

LATE

MID.

EAR.

EAR.

EAR.

MID.

L.

L.

PERMIAN

CARBONIF.

DEVONIAN

SILURIAN

PALEOEAR.

L.

Pre- Breakup

Stable platform

Breakup

Uplift, block faulting,local volcanism; rapid

sedimentationfollowed by erosion

Post- Breakup

?

Restricted marine

Open marine, lowsedimentation rates

Establishment ofwidespread carbonate

shelf facies

Mid Oligocenesea-level fall

? ?

KETUM FORMATION

AIFAM GROUP

TIPUMA FORMATION

INANWATANPOLYSEQUENCE

ROABIBAPOLYSEQUENCE

KE

MB

ELA

NG

AN

GR

OU

P F

OR

MAT

ION

LOWER NEW GUINEALIMESTONE GROUP= FAUMAI & WARIPI

FORMATIONS

UPPER NEW GUINEALIMESTONE GROUP

= KAIS & SIRGAFORMATIONS

KLASAMAN & KLASAFETFORMATIONS

TECTONICSTAGES STRATIGRAPHY OF WESTERN WEST PAPUA

Metamorphosed turbidites(mostly slate and shale).

Marine mudstone and limestone passing upinto non marine clastic sequence.

Continental red beds (shale and sandstone).

Nearshore marine to progressively non marine:shallow water marine limestone + lacustrine

& anoxic sands, clays and coal.

Transgressive sequence: claystone pass upinto limestone, sandy in places. Little

terrestrial influence.

Open marine fine clastic depositsand deep water clays.

Deep marine argillaceous section. No terrestrialinfluence. Volcanic event at the base.

Faumai Fm: bank and shoal arenaceouslimestone with abundant foraminifers.

Waripi Fm: oolitic and bioclastic foraminiferallimestone, calcareous sandstone and

mudstone in places.

Kais Fm: reefal limestone.

Sirga Fm: transgressive clastic sequence(mudstone, locally sandstone).

SEBYARPOLYSEQUENCE

JASSPOLYSEQUENCE

Major regionaltransgression

Majorerosion:50 m.y.

Major regionaltransgression

Klasaman Fm: sandy mudstone andcalcareous sandstone. Conglomerates

and lignite in upper part.Klasafet Fm: marl, siltstone and a little limestone.

Clastic sedimentationin new basins

Figure 5. Stratigraphy of westernWest Papua (modified after Pairault,2002).

• The early Callovian to early Kimmeridgian RoabibaPolysequence (CS-1 and TBF-1 wells) follows a12 million year period of erosion or non-deposition andcomprises significant sandstones, together withclaystone and limestone deposited in a nearshoreenvironment.

• The mid-Tithonian to early Valanginian SebyarPolysequence represents a major regional transgressionfollowing a major stratigraphic break.

The Coniacian to Maastrichian Jass Polysequence(Agung-1, CS-1, TBF-1, Gunung-1 and Onin South-1 wells)followed early Cretaceous erosion. A major transgressiontook place from the mid-Campanian, leading to the depositionof deepwater argillaceous sediments.

The Triassic rocks of eastern Indonesia have apredominantly marine carbonate aspect. Rich oil-pronesource rocks exist in both Buru (Fraser et al., 1993) andSeram islands. However, in the Bird’s Head, an NNGPMwell, Puragi-1, penetrated a reservoir quality sandstone thatis thought to be Triassic in age.

The Jurassic has been sampled by fieldwork within WestPapua, and also on the islands of Banggai, Sula, Tanimbarand Kai. Numerous wells have penetrated the Jurassic,predominantly as marine clastics. The thickest succession isrecorded within the well TBJ-1, just west of the OninPeninsula (Fig. 1). The Jurassic appears to pinch out towardsthe Bird’s Head and an east–west depositional margin is

considered to exist within the northern Bird’s Head, parallelto the coastline of Berau Bay. Existing northwest–southeaststructural lineaments controlled deposition within the Bird’sHead. It is thought that the Jurassic coastline transgressednorthward through the Jurassic (subject to eustaticvariations). Throughout the eastern Indonesia region and theTimor Sea, reservoir quality sandstones were deposited innearshore marine settings during the early Late Jurassic (seealso Barber et al., 2003), prior to the so-called “Break-upEvent”. It is proposed here that the Plover Sandstone reservoirof the Timor Sea is synchronous with both the reservoir ofthe Abadi discovery (Indonesia Masela PSC; Nagura et al.,2003) and the Tangguh Field, suggesting a similardepositional setting over an enormous area of coastline andnearshore marine environment.

The Late Jurassic and Early Cretaceous are less widelyrecognised because their deposition was adversely affectedby major structural events. The “Break-up Event” resultedin regional inundation of the low-lying coastline and shallowmarine fringe, resulting in widespread deep marine claystonedeposition. Depositional packages are complex and hard tocorrelate, due to subsequent partial erosion and variedstructural attitudes as the rift basins stabilized andsubsequently failed again. A major sediment-starved flooding(deepening) event in Aptian/Albian times resulted in thedeposition of the Darwin Radiolarite in the Timor Sea region.To the authors’ current knowledge, the radiolarite lithofacieshas yet to be recognised within West Papua.

544 Geological development of eastern Indonesia and northern Australia collision zone

The rest of the Late Cretaceous saw continuing deepmarine deposition with occasional shallowing, probablycontrolled by structural tilting linked to post-rift thermal sag.Carbonate rocks are recognised within some wells drilled inthe northern Australian margin.

Key petroleum system attributes of the Mesozoic are:

1. Oil-prone source rocks within (a) marine Late Triassicmarine carbonates, and (b) paralic Middle–Late Jurassicclastics.

2. High quality sandstone reservoirs of Middle–LateJurassic age widespread throughout the north Australianmargin.

3. Thick Cretaceous claystone successions provideeffective seal to reservoir and burial for source.

4. The “Break-up Event” of Late Jurassic age (climax)provides the initial structural setting for most of today’shydrocarbon discoveries.

Neogene Collision

Between 45 and 25 Ma, there was subduction north ofAustralia, remnants of which are seen in the mantle (Halland Spakman, 2002).

Initial collision between the Australian and the Philippineplates began around 25 Ma when transcurrent motion of thePhilippine Plate resulted in the shaving of blocks from thenortherly protrusion of the Australian continent (Pigram andPanggabean, 1984; Hutchison, 1989; Longley, 1997). Theleading edge of the Australia Plate arrived at the Indonesianarc subduction zone and initiated the New Guinea fold belt(Hamilton, 1979; Hall, 1996).

Bird’s Head (Kepala Burung) and the Sorong Fault Zone

The western extremity of the island of New Guinea lying tothe north of 2°S is known as the “Bird’s Head” (KepalaBurung; Fig. 1). The core region comprises Palaeozoic andsouthward-younging continental rocks of the Kemum Block,thought to have formed on a passive continental margin(Pigram and Davies, 1987). The Kemum Block is separatedfrom three principal fault-bounded blocks to the north by theSorong Fault. The Tamrau, Netoni and Tosem blocks comprisedisparate successions of Mesozoic and Cenozoic arc-derivedsedimentary and igneous rocks (Pigram and Davies, 1987;Hall and Wilson, 2000).

There is some uncertainty as to the origin of the Bird’sHead, in particular its location relative to Australia in Jurassictimes and whether it has rotated significantly during theNeogene. There is no compelling evidence for either (a)significant rotation, or (b) an origin from the eastern side ofAustralia (e.g. Struckmeyer et al., 1993).

The origin of Cenderawasih Bay is also uncertain. Thetriangular shape and water depths of more than 2 km suggeststhat it could have opened to the north (cf. sphenochasmconcept of Carey, 1958). However, the geometric argumentsrely on shapes of poorly mapped features and depths in theorder of 2 km are not typical of normal oceanic crust. If theunderlying crust is oceanic, it must be overlain by aconsiderable thickness of sediment. Its age is unknown. Inaddition, satellite gravity data (Fig. 6) indicates a stronggravity high along the Wandamen Peninsula due to very

young metamorphic rocks. Very young ages (<5 Ma) suggestrecent extension and a possible core complex formation;these young metamorphic rocks require further investigation.

A gravity low is present only along the eastern margin ofCenderawasih Bay, therefore it appears unlikely that the entirebay has opened by rotation. Transpressional extension alongthe north-northeast lineament close to the eastern margin maybe responsible for the rhomboidal shape of the deepest partof Cenderawasih Bay, possibly related to the development ofthe Lengguru Fold Belt to the west and south.

The once contiguous microcontinents of eastern Indonesiawere dismembered and dispersed by Sorong fault splays around20 Ma, resulting in the formation of the Bird’s Head, Seram-Buru, Obi-Bacan, Buton-Tukang Besi and Banggai-Sulacomplexes (Pigram and Panggabean, 1984; Hall, 2002). Since12 Ma, the authors consider that the Bird’s Head has not movedany significant distance. A small counter-clockwise rotation tookplace between 4 and 8 Ma, and relatively small strike-slipmovements between 4 and 8 Ma and also 0.5 and 2 Ma (Hall,2002). These events may well have direct relevance to thedevelopment of Cenderawasih Bay.

The Sorong Fault System, the southern boundary of boththe Molucca Sea and the Philippine Sea plates with theAustralian Plate (Hall and Wilson, 2000), is one of the mostimportant structural elements of South East Asia. The SorongFault has been responsible for translating continentalfragments from the northern margin of the Australian Plate(i.e. the Bird’s Head) into the outer edge of Sundaland(Pigram and Panggabean, 1984; Hutchison, 1989). This faultsystem was initiated in the early Miocene, the result ofoblique convergence of Australia and the Philippine Sea,Caroline and Pacific plates, and continues to the present day.

The Banda Arc

The Banda Arc, a west-facing horseshoe-shaped arc, lyingbetween the Timor and Arafura seas and the Birds Headmicro-continent is the locus of the convergence and criticalto understanding the tectonic development of the region.

From the Banda Sea side outwards, the arc comprises aninner volcanic arc, which has been active since the LateMiocene, and an outer non-volcanic arc of islands formedprincipally of allochthonous sedimentary, metamorphic andsome igneous rocks (Darman and Sidi, 2000; Hall, 2002).The Inner Banda Arc includes the Solor and Alor islandgroups, Wetar and Banda, and the Outer Banda Arc includesTimor, Tanimbar, Kai Besar, Seram and Buru (Fig. 1).

The Outer Banda Arc comprises a series of belts of rocks(Bowin et al., 1980; Darman and Sidi, 2000) comprising:

• a discontinuous inner belt of “ophiolite”, which ingeneral is blocky and narrow;

• a belt of low- to high-grade metamorphic rocks;• a fold-and-thrust belt dominated by Permo-Triassic

sedimentary rocks derived from the Australiancontinental margin;

• a fold-and-thrust belt dominated by Late Mesozoic andCenozoic deepwater sediments; and

• an outer belt of uplifted late Neogene basins.

The rocks of the Outer Banda Arc comprise Late Triassicto Early Jurassic shallow water clastics and carbonates

Baillie et al. 545

-400

-400

-400 -800

400

0

400

800

0-400

-800

0

0

400 800

800

800

1200

130°E 132°E 134°E 136°E5°S

3°S

1°S

0 100km

Figure 6. Satellite gravity, Bird’s Head and Cenderawasih Bay (source: www.deos.tudelft.nl/altim/ceo/pim.html), with onshore Bougueranomalies and trends (after Dow et al., 1986) indicated.

unconformably overlain by a Late Jurassic to Palaeogenecondensed sequence of deep-water carbonate and chert– stratigraphies vary only after the Oligocene. Indeed,the virtually identical (“Australian”) Mesozoicstratigraphies of Buton, Buru and Seram (and possiblyEast Sulawesi) suggest that they originally formed partof the Australian craton (Audley-Charles et al., 1979;Milsom, 2000).

Major thrust faults have been mapped north of theTimor Trough to the west of, and on the island of Timor,north of Wetar (the currently seismically active WetarThrust), north of the island of Flores (Flores Thrust) andnorth of Tanimbar. A number of northeast-trending strike-slip faults and fracture zones have been mapped along theBanda Arc deformation zone, extending from Sumba toTanimbar (McCaffrey, 1996).

Earthquakes suggest that the present stage of thecollision of the Australian continental margin with theBanda Arc probably involves more rapid convergence atthe backarc thrusts than at the Timor Trough. The Wetarand Flores backarc zones are considerably more energeticseismically than the subduction thrust fault north of theeastern Java Trench and Timor Trough; all known largethrust earthquakes recorded during the 1990s occurred atthe Flores and Wetar thrusts (McCaffrey, 1996). A changefrom south- to north-directed thrusting after the emergenceof Timor is noteworthy (Price and Audley-Charles, 1987;Hall and Wilson, 2000); subduction reversal may be inprogress, accompanied by the recent disappearance of the

Banda allochthon. Deep seismic studies have reinforcedthe suggestion of subduction polarity reversal (Snyderet al., 1996).

Trough systems

Many previous workers have described a single troughsystem resembling a giant horseshoe extending from theJava Trench through the Timor, Tanimbar and Aru troughsinto the Seram Trough (Fig. 3). This may be fortuitous, butincorrect. The deep-water crustal lineaments andcorresponding gravity lows (Figs 2, 6) observed throughoutthe study area owe their origin to a number of widely differentmechanisms.

The Java Trench (Fig. 3) marks the boundary betweenthe Indian Ocean-Australian plates and Sunda-Indonesianarc lithosphere, and has been the location of subductionthrough most of the Cenozoic. The Java Trench and allpresent-day subduction terminate near the island of Sumba(Fig. 3). Subduction has completed consumption of oceaniccrust outboard of the Tertiary accretionary wedge (greencolour on Fig. 3).

The Timor, Tanimbar, Aru and Seram troughs, generallyless than 50 km across and only 1.0–2.5 km deep, mark theedge of relatively undeformed Australian continental crust– the deformed leading edge of the continental crust isindicated by vertical overprint on Figure 3. Modern highquality seismic data (Fig. 7a-d) permits evaluation of thesecontact zones.

546 Geological development of eastern Indonesia and northern Australia collision zone

Figure 7. Representative seismic lines across collision zone, (a)Seram Trough, seismic line EIR98–201, (b) Aru Trough, lineEIR98–106, (c) Tanimbar Trough, line MH01–11, (d) TimorTrough, line MH01–18; location of lines indicated on Figure 1.

Figure 7a shows a representative line across the SeramTrough, broadly similar to the Timor Trough, albeit withreversed polarity. Gravity data (Fig. 2) indicates a gravityhigh centred on Kai Besar, showing that the Seram Troughis not contiguous with the Aru Trough.

The Aru Trough (Fig. 7b), immediately north of theTanimbar Trough, is totally different in character and showsabundant evidence of strike-slip movements with rapidponding of recent sediment in deep, linear depocentres(Fig. 8). Along the eastern flank of the trough, dredgedsamples include Carboniferous–Permian siliciclastics,Jurassic–Cretaceous black shale, late Albian to late Eoceneopen-marine pelagic sediment, early Miocene shallow-waterplatform carbonate and late Miocene to Holocene pelagicsediment (Cornee et al., 1997).

It is possible that the eastern contact continues thoughthe “Bird’s Neck” and along the eastern (markedly linear)side of Cenderawasih Bay where it is probably terminatedby the Yapen Fault, a splay of the Sorong Fault. West Papuagravity data (Fig. 6) indicates the presence of dislocationacross the “Bird’s Neck”, supporting this conclusion.

Figures 7c and 7d traverse the Tanimbar and Timortroughs respectively. The downwarped Australian plateshows evidence of relatively minor, brittle extensionaldeformation. The contact zone is relatively sharp and displaysrelatively minor compressional deformation. Rocks of theOuter Banda Arc are highly deformed and include a seriesof stacked thrust sheets.

Most previous workers have recognized that the troughsystem is a relatively young and active tectonic feature.New seismic data from the Tanimbar Trough givesindependent supporting evidence for a very young age(Fig. 9). Modern and fossil submarine canyons indicativeof deepwater conditions probably developed duringPleistocene lowstands, when sediment was made availablefrom the Arafura Platform. These are confined to thePliocene and younger section.

Discussion

Subduction and accretion

Almost as contentious as the origin of the Banda Arc, hasbeen the location and timing of post-Jurassic subduction inthe region. From at least early Eocene times until the presentday, oceanic crust has been subducting west and south ofSundaland (Sumatra, Java, Borneo, west Sulawesi) and northof New Guinea (Hall 2002). The current (fast) northwardmovement of the Australian Plate of 9 cm/year was initiatedaround 48 Ma (e.g. Baillie et al., 1994) implying that inexcess of 4,000 km of shortening must be accommodated atthe northern margin.

While conventional subduction of Indian Oceanoceanic crust (oceanic part of the Australian plate)continues in the Java Trench (Masson et al., 1990), allother troughs are not as straightforward. Rocks of theIndonesian archipelago are juxtaposed with continentalrocks of the Australian plate where there is no subductioncurrently taking place. The boundary between the tworegimes is clearly at Sumba where a promontory ofAustralian continental lithosphere collided with Sumbabetween 8 Ma (Keep et al., 2003) and 16 Ma (Rutherfordet al . , 2001), also resulting in the mid-Mioceneunconformity across Timor and subduction of Jurassicoceanic crust in the Banda Sea (Hall, 2002, fig. 24).

GPS measurements indicate that Timor appears to movingnorthward at about the same rate as Australia (Genrich et al.,1994; McCaffrey 1996).

Banda Arc and Trough systems

The Banda Arc appears to have achieved the impossible bybeing in collision simultaneously, and with the same plate,to the north and to the south (Milsom et al., 1996).

The cause of the extreme curvature of the Banda Arc isone of the major unsolved problems of South East Asiangeology (Milsom et al., 1996). As noted by Hall (2002),the curvature itself has suggested derivation as a result ofrotation driven by collision – many other authors have hadmany other ideas. There are four disparate groups oftheories on the origin of the arc:

• Folding of an earlier east-west arc.• The arc achieved its present curvature in the

Cretaceous.• The arc was formed by accretion of Australian

continental material leading to enclosure of the BandaSea oceanic basin.

• The arc inherited its present shape from the irregularJurassic Australian continental margin.

Baillie et al. 547

5.000

6.000

7.000

8.000

9.000

4000.04500.05000.05500.06000.06500.0SP:

5.000

6.000

7.000

8.000

9.000

Figure 8. Detail of Aru Trough seismic line.

Figure 9. Detail of seismic line MH01–18, showing modern and fossil submarine canyons; prominent seismic marker below canyonsis Base Pliocene (green arrow). Note canyons young away from trough axis, suggesting dynamic system with sequential activationrather than a single event.

1.000

2.000

3.000

4.000

5.000

2000.02500.03000.03500.04000.0

- MH01-18 -

4500.05000.0SP:

1.000

2.000

3.000

4.000

5.000

548 Geological development of eastern Indonesia and northern Australia collision zone

The age of the Banda Sea is critical for tectonicreconstructions – assumed ages of formation range fromMesozoic to Neogene (Norvick, 1979; Milsom, 2000; Pigramand Panggabean, 1983). More recently, the age of the BandaSea has been thought to be Late Triassic, related to the openingof Meso-Tethys (Metcalfe, 1998), or Jurassic related toseparation of the West Burma block from northern Australia(Longley et al., 2002; Hall, 2002). It now appears that therehave been two episodes of oceanic crust formation in the region(a) Jurassic oceanic crust generated during the breakup of theGondwanaland margin (“Proto-Banda Sea” of Hall, 2002),which has now been subducted, and (b) Neogene oceanic crustof the present Banda Sea, as originally suggested by Hamilton(1979), and confirmed by dredging and interpretation ofanomalies (Hall, 2002).

The geology of Timor has been used to constrain tectonicmodels for the region. Three main structural models have beenproposed for Timor (Richardson and Blundell, 1996):

1. The Imbricate Model – Timor is interpreted as anaccumulation of chaotic material imbricated against thehanging wall of a subduction trench, the Timor Trough,and essentially forms a large accretionary prism.

2. The Overthrust Model – Timor is interpreted in terms ofAlpine-style thrust sheets.

3. The Rebound Model – the Australian continental marginentered a subduction zone in the vicinity of the WetarStrait. Subsequently, the oceanic lithosphere detachedfrom the continental part due to “slab snapping”, resultingin the uplift of Timor by isostatic rebound on steep faults.

Various combinations of these differing models have alsobeen proposed.

The trough systems of the collision zone are less wellstudied. A recent study of the Seram Trough system (Pairault,2002; Pairault et al., 2003), has shown that evolution iscomplex, with the following sequence of Neogene eventsrecognised:

• A Pliocene deformation event, probably around 5 Ma,which produced the MOKA (Misool-Onin-KumawaAnticlinorium of Fraser et al., 1993).

• A period of erosion in the early Pliocene – the resultingunconformity is now covered by about 1 km of sedimentsnear the culmination of the anticlinorium.

• The unconformity was folded and now dips south towardsthe Seram Trough south of the anticlinal axis.

Pairault et al. (2003) suggested that thrust loading onSeram may have caused the trough to develop and that theSeram Trough is a foredeep and not a subduction trench.Pairault et al. (2002) interpreted the evolution of the SeramTrough in the context of an intraplate shortening model, toindicate that oceanic lithosphere of the Australian Plate startedto roll back towards the southeast at the beginning of the lateMiocene.

Conclusions

For much of its history, the study area has been part of, orrelatively close to a continental margin. The geologicalevolution may be considered in terms of Proterozoic and

Palaeozoic building blocks, development of a Mesozoic mid-latitude passive margin on the eastern margin ofGondwanaland, and the Neogene collision of Australia andthe Indonesian archipelago. Present day regional tectonicelements are shown as Figure 3.

By the early Palaeozoic, the region was part of the easternmargin of Gondwanaland (Metcalfe, 1998).

During Palaeozoic and Mesozoic times, three significantperiods of terrane dispersion occurred where continentalfragments separated from the margin of Gondwanaland, driftednorthward and were subsequently accreted to the Eurasiancraton (Hutchison, 1989).

At around 160 Ma, the northern margin of Australia lay atapproximately 40°S latitude – and remained in mid-latitudesthrough the Cretaceous (Baillie et al., 1994). By the end of theJurassic, this somewhat irregular margin was part of a passivemargin facing an open ocean containing several continentalblocks or microcontinents – much of this oceanic crust hassubsequently been subducted.

The present complex geology of the region results fromtwo disparate tectonic forces: (1) northward movement of theAustralian Plate, and (2) sinistral shear on the northern marginof New Guinea (Sorong fault system), resulting from relativemotion of the Australian and Pacific plates. Initial collisionbetween the Australian and the Philippine plates began around25 Ma when sinistral transcurrent motion of the PhilippinePlate resulted in the shaving of blocks from the northerlyprotrusion of the Australian continent.

Acknowledgements

Figure 2 was compiled by Dino Loschi. We thank NickHoffman and Ron Noble for useful reviews and TGS-NOPECGeophysical Company and WesternGeco for permission topublish proprietary information.

References

AUDLEY-CHARLES, M.G., CARTER, D.J., BARBER, A.J.,NORVICK, M.S. AND TJOKROSAPOETRO, S., 1979,Reinterpretation of the geology of Seram: implicationsfor the Banda Arcs and northern Australia. GeologicalSociety Journal, 136, 547–568.

BAILLIE, P.W., POWELL, C.McA., LI, Z.X. AND RYALL,A.M. 1994. The tectonic framework of WesternAustralia’s Neoproterozoic to Recent sedimentary basins.In: Purcell, P.G. and Purcell, R.R. (Eds), 1994, TheSedimentary Basins of Western Australia, Proceedingsof the Petroleum Exploration Society of AustraliaSymposium, Perth, WA, 1994, 45–61.

BARBER, P., CARTER, P., FRASER, T., BAILLIE, P. ANDMYERS, K., 2003, Paleozoic and Mesozoic petroleumsystems in the Timor and Arafura Sea, eastern Indonesia.Proceedings of the 29th Annual Indonesian PetroleumAssociation Convention, Jakarta, October 14–16 2003,485–500.

BOWIN, C., PURDY, G.M., JOHNSTON, C., SHOR, G.,LAWVER, L., HARTONO, H.M.S. AND JEZEK, P.,1980, Arc-continent collision in Banda Sea region.AAPG Bulletin, 64, 868–915.

BRADSHAW, M.T., YEATES, A.N., BEYNON, R.M.,BRAKEL, A.T., LANGFORD, R.P., TOTTERDELL, J.M.

Baillie et al. 549

AND YEUNG, M., 1988, Palaeogeographic evolution of theNorth West Shelf Region. In: Purcell, P.G. and Purcell, R.R.(Eds), 1988, The North West Shelf Australia, Proceedings ofPetroleum Exploration Society of Australia Symposium,Perth, WA, 1988, 29–54.

BRADSHAW, J., NICOLL, R.S. AND BRADSHAW, M.,1990, The Cambrian to Permo-Triassic Arafura basin,northern Australia. APEA Journal, 30(1), 107–127.

CAREY, S.W., 1958, A tectonic approach to continental drift.Symposium on Continental Drift Proceedings, Hobart,177–355.

CORNEE, J.-J., VILLENEUVE, M. AND 12 OTHERS, 1997,Stratigraphic succession of the Australian margin betweenKai and Aru islands (Arafura Sea, eastern Indonesia)interpreted from Banda Sea II cruise dredge samples.Journal of Asian Earth Sciences, 15, 423–434.

DARMAN, H. AND SIDI, F.H. (Eds), 2000, An Outline of theGeology of Indonesia. Indonesia Association ofGeologists, 192 pages.

DOW, D.B., ROBINSON, G.P. HARTONO, U. ANDRATMAN, N., 1986. Geological Map of Irian Jaya,Indonesia (1:1,000,000 scale). Geological Research andDevelopment Centre, Bandung.

FRASER T.H., BON J., SAMUEL L., RUKMIATI M.G. ANDIMAM HARDJO J., 1993, A new dynamic Mesozoicstratigraphy for the west Irian micro-continent,Indonesia, and its implications: Proceedings of theIndonesian Petroleum Association, 22nd AnnualConvention, 707–761.

GENRICH, J., BOCK, Y., McCAFFREY, R., CALAIS, E.,STEVENS, C. AND OTHERS, 1994, Kinematics of theeastern Indonesian island arc estimated by Globalpositioning system measurements. EOS, 75, 162.

HALL, R., 1996, Reconstructing Cenozoic SE Asia. In: Hall,R. and Blundell, D.J. (Eds), Tectonic Evolution of SEAsia. Geological Society of London Special Publication,126, 153–184.

HALL, R., 2002, Cenozoic geological and plate tectonicevolution of SE Asia and the SW Pacific: computer-basedreconstructions, model and animations. Journal of AsianEarth Sciences, 20, 353–431.

HALL, R. AND SPAKMAN, W., 2002, Subducted slabsbeneath the eastern Indonesia–Tonga region: insightsfrom tomography. Earth and Planetary Science Letters,201, 321–336.

HALL, R. AND WILSON, M.E.J., 2000, Neogene suturesin eastern Indonesia. Journal of Asian Earth Sciences,18, 781–808.

HAMILTON, W., 1979, Tectonics of the Indonesian Region.USGS Professional Paper 1978, 345 pages.

HUTCHINSON, C.S., 1989, Geological Evolution ofSouth-East Asia. Clarendon Press, Oxford.

KEEP, M., LONGLEY, I. AND JONES, R., 2003, Sumba and itseffect on Australia’s northwestern margin. In: Hillis, R.R.and Muller, R.D. (Eds), The Evolution and Dynamics ofthe Australian Plate. Geological Society of AustraliaSpecial Publication, 22 and Geological Society of AmericaSpecial Publication, 372, 309–318.

LONGLEY, I.M., 1997, The tectonic evolution of SE Asia.In: Fraser, A.J., Mathews, S.J. and Murphy, R.W. (Eds),Petroleum Geology of Southeast Asia, GeologicalSociety Special Publication, 126, 311–339.

LONGLEY, I.M., BUESSENSCHUETT, C., CLYDESDALE, L.,CUBITT, C.J., DAVIS, R.C., JOHNSON, M.K.,MARSHALL, N.M., SOMERVILLE, R., SPRY. T.B.AND THOMPSON, N.B., 2002, The North West Shelfof Australia – a Woodside perspective. In: Keep, M. andMoss, S.J. (Eds), 2002, The Sedimentary Basins ofWestern Australia 3: Proceedings of the PetroleumExploration Society of Australia Symposium, Perth, WA,2002, 27–86.

McCAFFREY, R., 1996, Slip partitioning at convergent plateboundaries of SE Asia. In: Hall, R. and Blundell, D. (Eds),Tectonic Evolution of Southeast Asia. Geological SocietySpecial Publication, 106, 3–18.

McLENNAN, J.M., RASIDI, J.S., HOLMES, R.L., ANDSMITH, G.C., 1990, The geology and petroleum potentialof the Western Arafura Sea. The APEA Journal, 30(1),91–127.

MASSON, D.G., PARSON, L.M., MILSOM, J., NICHOLS,G., SIKUMBANG, N., DWIYANTO, B. ANDKALLAGHER, H., 1990, Subduction of seamounts atthe Java Trench: a view with long-range sidescan sonar.Tectonophysics, 185, 51–65.

METCALFE, I. , 1998, Palaeozoic and Mesozoicgeological evolution of the SE Asian region:multidisciplinary constraints and implications forbiogeography. In: Hall, R. and Holloway, J.D. (Eds),Biogeography and Geological Evolution of SE Asia,25–41.

MILSOM, J., 1991, Gravity measurements and terranetectonics in the New Guinea region. Journal of SoutheastAsian Earth Sciences, 6, 319–328.

MILSOM, J., 2000, Stratigraphic constraints on suturemodels for eastern Indonesia. Journal of Asian EarthSciences, 18, 761–779.

MILSOM, J., KAYE, S. AND SARDJONO, 1996, Extension,collision and curvature in the eastern Banda arc. In: Hall,R. and Blundell, D. (Eds), Tectonic Evolution ofSoutheast Asia. Geological Society Special Publication,106, 85–94.

NAGURA, H., SUZUKI, I., TERAMOTO, T., HAYASHI, Y.,YOSHIDA, T., BANDJARNAHOR, H., KIHARA, K.,SWIECICKI, T. AND BIRD, R., 2003, The Abadi GasField. Proceedings of the 29th Annual IndonesianPetroleum Association Convention, Jakarta, October14–16, 2003, 451–466.

NORVICK, M.S., 1979, The tectonic history of the BandaArcs, eastern Indonesia: a review. Geological SocietyJournal, 136, 519–527.

PAIRAULT, A., 2002, Structural styles and tectonicevolution of the Seram Trough: Indonesia. RoyalHolloway University of London, M.Sc. report,unpublished.

PAIRAULT, A.A., HALL, R. AND ELDERS, C.J., 2003,Tectonic Evolution of the Seram Trough. Proceedings ofthe 29th Annual Indonesian Petroleum AssociationConvention, Jakarta, October 14–16 2003, 355–370.

PIGRAM, C.J. AND DAVIES, H.L., 1987, Terranes and theaccretion history of the New Guinea orogen. BMR Journalof Australian Geology & Geophysics, 10, 193–221.

PIGRAM C.J. AND PANGGABEAN H., 1981, Pre-Tertiarygeology of western Irian Jaya and Misool island:implications for the tectonic development of eastern

550 Geological development of eastern Indonesia and northern Australia collision zone

Indonesia: Proceedings of the Indonesian PetroleumAssociation, 10th Annual Convention, 385–399.

PIGRAM, C.J. AND PANGGABEAN, H., 1983, Age of theBanda Sea, eastern Indonesia. Nature, 301, 231–234.

PIGRAM, C.J. AND PANGGABEAN, H., 1984, Rifting ofthe northern margin of the Australian continent and theorigin of some microcontinents in eastern Indonesia.Tectonophysics, 107, 331–353.

PRICE, N.J. AND AUDLEY-CHARLES, M.D., 1987, Tectoniccollision processes after plate rupture. Tectonophysics,140, 121–129.

RICHARDSON, A.N. AND BLUNDELL, D.J., 1996,Continental collision in the Banda arc. In: Hall, R. andBlundell, D. (Eds), Tectonic Evolution of Southeast Asia.Geological Society Special Publication, 106, 47–60.

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 inthe Miocene. Journal of Asian Earth Sciences, 19, 453–479.

SNYDER, D.B., MILSOM, J. AND PRASETYO, H., 1996,Geophysical evidence for local indentor tectonics in theBanda Arc east of Timor. In: Hall, R. and Blundell, D.(Eds), Tectonic Evolution of Southeast Asia: GeologicalSociety Special Publication, 106, 61–74.

STRUCKMEYER, H.I.M., YEUNG, M. AND PIGRAM, C.J.,1993, Mesozoic to Cainozoic plate tectonic andpalaeogeographic evolution of the New Guinea region. In:Carman, G.J. and Carman Z. (Eds), Petroleum Exploration inPapua New Guinea: Proceedings of the Second PNGPetroleum Convention, Port Morseby, 261–290.

VEEVERS, J.J. (Ed.), 2000, Billion-year Earth History ofAustralia and Neighbours in Gondwanaland. GemocPress, Sydney.

Tom Fraser was born in Scotlandand educated at Queen MaryCollege, London University underthe eagle eye of Prof. JakeKirkaldy. Tom’s baptism into theoil and gas industry was bymudlogging in the North Sea. Heprogressively worked his way intointerpretation and New Venturestasks. He has worked in

17 different countries, including South America and WestAfrica. During the last 15 years he has worked principally inSouth East Asia, in both Mesozoic and Tertiary basins. Hisprincipal interests are in integration of seismic and geologicalinformation, and the quantification and display of PetroleumSystem attributes. He enjoys a significant after-hoursrelationship with his Fender bass.

Keith Myers graduated ingeology at Curtin University,Western Australia andundertook post-graduatestudies in petrology andmineralogy. After a period ingold exploration, he moved topetroleum consulting firm Dolan& Associates in 1992, workingon projects in various locations

in Australia, South East Asia and Africa. In 1999, he joinedWestern Geophysical as Manager of Exploration &Reservoir Services, with continued involvement in Asia,especially China and Indonesia. After the formation ofthe WesternGeco joint venture, he moved to Kuala Lumpuras Business Development Manager for the Asia Pacificregion. Keith is a member of PESA, SEAPEX and theGeological Society of Malaysia.

Robert Hall is Professor of Geologyand Director of the SE Asia ResearchGroup at Royal Holloway Universityof London. He completed his PhD in1974 on ophiolites and sutures ineastern Turkey and later worked inthe Eastern Mediterranean andMiddle East for several years. Since1984, he has worked in South EastAsia, mainly in Indonesia, carrying

out field-based research supported by industrial consortia, withpost-doctoral researchers and postgraduate research students.These studies have been the basis for computer animations ofthe plate tectonic reconstructions of SE Asia and the SWPacific for the Cenozoic. Research interests: Regional geologyof SE Asia and the western Pacific. Island arc origin andevolution. Plate tectonic reconstructions of SE Asia and thewest Pacific. Seismic tomography, mantle processes andtectonics of the region. Tropical sedimentation: links toprovenance, climate and tectonics. Implications of platetectonics for biogeography of SE Asia.

Peter Baillie is a graduate of theUniversity of Tasmania (geology)and Macquarie University inSydney (sedimentology, basinanalysis). He was employed at theTasmanian Department of Minesfrom 1970 until 1993 and heldvarious positions in regionalgeological mapping and petroleumexploration administration. In 1993,

he moved to the Department of Minerals and Energy in WesternAustralia as Manager of the Exploration and Production Branchin the Petroleum Division. He joined TGS-NOPEC GeophysicalCompany in 1997 and is Chief Geologist Asia Pacific. Formerlythe Managing Editor of the PESA Journal, Peter was DeputyChairman of WABS II (1998) and Chairman of WABS III (2002).