Abstract In the Wilson cycle, there is a change from an opening to a closing ocean when subduction begins. Subduction initiation is commonly identified as a major problem in plate tectonics and is said to be nowhere observable, yet there are many young subduction zones at the west Pacific margins and in eastern Indonesia. Few studies have considered these examples. Banda subduction developed by the eastwards propagation of the Java trench into an oceanic embayment by tearing along a former ocean–continent boundary. The earlier subducted slab provided the driving force to drag down unsubducted oceanic lithosphere. Although this process may be common, it does not account for young subduction zones near Sulawesi at different stages of development. Subduction began there at the edges of ocean basins, not at former spreading centres or transforms. It initiated at a point where there were major differences in elevation between the ocean floor and the adjacent hot, weak and thickened arc/continental crust. The age of the ocean crust appears to be unimportant. A close relationship with extension is marked by the dramatic elevation of land, the exhumation of deep crust and the spectacular subsidence of basins, raising questions about the time required to move from no subduction to active subduction, and how initiation can be identified in the geological record.

Abstract Eastern Indonesia is the site of intense deformation related to convergence between Australia, Eurasia, the Pacific and the Philippine Sea Plate. Our analysis of the tectonic geomorphology, drainage patterns, exhumed faults and historical seismicity in this region has highlighted faults that have been active during the Quaternary (Pleistocene to present day), even if instrumental records suggest that some are presently inactive. Of the 27 largely onshore fault systems studied, 11 showed evidence of a maximal tectonic rate and a further five showed evidence of rapid tectonic activity. Three faults indicating a slow to minimal tectonic rate nonetheless showed indications of Quaternary activity and may simply have long interseismic periods. Although most studied fault systems are highly segmented, many are linked by narrow (<3 km) step-overs to form one or more long, quasi-continuous segment capable of producing M > 7.5 earthquakes. Sinistral shear across the soft-linked Yapen and Tarera–Aiduna faults and their continuation into the transpressive Seram fold–thrust belt represents perhaps the most active belt of deformation and hence the greatest seismic hazard in the region. However, the Palu–Koro Fault, which is long, straight and capable of generating super-shear ruptures, is considered to represent the greatest seismic risk of all the faults evaluated in this region in view of important strike-slip strands that appear to traverse the thick Quaternary basin-fill below Palu city.

Abstract Detrital garnet compositions can be used to help determine the provenance of sedimentary rocks. In this study a garnet compositional database consisting of more than 2500 wet chemical and electron microprobe analyses was compiled from the literature. For the garnets in the database the six principal garnet end-member compositions (pyrope, almandine, spessartine, uvarovite, grossular and andradite) were calculated. A multi-stage methodology was devised to match garnet compositions to source rocks, and a series of garnet provenance fields on ternary plots were identified. The method was tested using compositional data from detrital garnet studies in several areas where provenance has already been identified, with good results. The methodology was then used to assess the provenance of detrital garnets from Neogene sandstones of northern Sabah, Borneo for which provenance is unknown and in a region where there are few garnet analyses for comparison. Comparison of garnet compositions from possible sources on Palawan, Philippines and in Borneo combined with the ternary plots excludes some known garnet-bearing rocks as potential sources and suggests derivation of metamorphic and igneous garnets from Palawan during the Early Miocene. Supplementary material: The compositional database and garnet data plotting spreadsheets are available at www.geolsoc.org.uk/SUP18649 .

Detrital zircon U-Pb geochronology can make an extremely valuable contribution to provenance studies and paleogeographic reconstructions, but the technique cannot distinguish grains with similar ages derived from different sources. Hafnium isotope analysis of zircon crystals combined with U-Pb dating can help make such distinctions. Five Paleogene formations in West Java have U-Pb age populations of 80–50 Ma (Late Cretaceous–Paleogene), 145–74 Ma (Cretaceous), 298–202 Ma (Permian–Triassic), 653–480 Ma (mid-Neoproterozoic–latest Cambrian), and 1290–723 Ma (late Mesoproterozoic–early Neoproterozoic). Hf-isotopes have been analyzed for 311 zircons from these formations. Differences in zircon U-Pb age and Hf-isotope populations reflect changing sources with time. Late Cretaceous and Paleogene zircons are interpreted as having been derived from two temporally discrete volcanic arcs in Java and West Sulawesi, respectively. The Java arc was active before micro-continent collision, and the W Sulawesi arc developed later, on newly accreted crust at the SE Sundaland margin. The collision age is estimated to be ca. 80 Ma. U-Pb age and 176 Hf/ 177 Hf i characteristics allow a distinction to be made between Cretaceous granitic and volcanic arc sources. Zircons that are older than ca. 80 Ma have a continental Sundaland provenance. Mid-Cretaceous zircons in all upper Eocene and lower Oligocene formations were derived from granites of the Schwaner Mountains of SW Borneo. Permian–Triassic zircons were derived predominantly from granites in the SE Asian Tin Belt. 176 Hf/ 177 Hf i ratios permit distinction between Tin Belt granites in the Main Range and Eastern Provinces, and indicate that only the lower Oligocene Cijengkol Formation contains significant input from the Main Range Province, suggesting a partial change in drainage pattern. Older zircon ages are more difficult to interpret but probably record contributions from allochthonous basement and sedimentary rocks that were deposited prior to rifting of continental blocks from Gondwana in the early Mesozoic.

Abstract The SE Asian gateway is the connection from the Pacific to the Indian Ocean and it has diminished from a wide ocean to a complex narrow passage with deep barriers ( Gordon et al. 2003 ) as plate movements caused Australia to collide with SE Asia. It is one of several major ocean passages that existed during the Cenozoic but has received much less attention than others that opened, such as the Drake Passage, Tasman Gateway, Arctic Gateway or Bering Straits, or that closed, such as the Panama Gateway or Tethyan Gateway (e.g. von der Heydt & Dijkstra 2006 ; Lyle et al. 2007 , 2008 ). It is not entirely clear why there has been this comparative neglect, but it may reflect the relative limited knowledge of the large and remote areas of Indonesia and the western Pacific, in particular their geological history, and the relatively small number of active researchers in this large region. Unlike the Panama Gateway and Tethyan Gateway the SE Asian gateway is still partly open and the ocean currents that flow between the Pacific and Indian Oceans have been the subject of much recent work by oceanographers (e.g. Gordon 2005 ). We now know that the Indonesian Throughflow, the name given to the waters that pass through the only remaining low latitude oceanic passage on the Earth, plays an important role in Indo-Pacific and global thermohaline flow ( Gordon 1986 ; Godfrey 1996 ), and it is therefore probable

Abstract Continental SE Asia is the site of an extensive Cretaceous–Paleocene regional unconformity that extends from Indochina to Java, covering an area of c . 5 600 000 km 2 . The unconformity has previously been related to microcontinental collision at the Java margin that halted subduction of Tethyan oceanic lithosphere in the Late Cretaceous. However, given the disparity in size between the accreted continental fragments and area of the unconformity, together with lack of evidence for requisite crustal shortening and thickening, the unconformity is unlikely to have resulted from collisional tectonics alone. Instead, mapping of the spatial extent of the mid–Late Cretaceous subduction zone and the Cretaceous–Paleocene unconformity suggests that the unconformity could be a consequence of subduction-driven mantle processes. Cessation of subduction, descent of a northward dipping slab into the mantle, and consequent uplift and denudation of a sediment-filled Late Jurassic and Early Cretaceous dynamic topographic low help explain the extent and timing of the unconformity. Sediments started to accumulate above the unconformity from the Middle Eocene when subduction recommenced beneath Sundaland.

Abstract The Sundaland core of SE Asia is a heterogeneous assemblage of Tethyan sutures and Gondwana fragments. Its complex basement structure was one major influence on Cenozoic tectonics; the rifting history of the north Australian margin was another. Fragments that rifted from Australia in the Jurassic collided with Sundaland in the Cretaceous and terminated subduction. From 90 to 45 Ma Sundaland was largely surrounded by inactive margins with localized strike-slip deformation, extension and subduction. At 45 Ma Australia began to move north, and subduction resumed beneath Sundaland. At 23 Ma the Sula Spur promontory collided with the Sundaland margin. From 15 Ma there was subduction hinge rollback into the Banda oceanic embayment, major extension, and later collision of the Banda volcanic arc with the southern margin of the embayment. However, this plate tectonic framework cannot be reduced to a microplate scale to explain Cenozoic deformation. Sundaland has a weak thin lithosphere, highly responsive to plate boundary forces and a hot weak deep crust has flowed in response to tectonic and topographic forces, and sedimentary loading. Gravity-driven movements of the upper crust, unusually rapid vertical motions, exceptionally high rates of erosion, and massive movements of sediment have characterized this region.

Abstract High resolution multibeam bathymetric and seismic data from the area north of the Banggai-Sula Islands, Indonesia, provide a new insight into the geological history of the boundary between the East Sulawesi ophiolite, the Banggai-Sula microcontinent and the Molucca Sea collision zone. Major continuous faults such as the Sula Thrust and the North Sula–Sorong Fault, previously interpreted to bound and pass through the area are not seen. The south-verging Batui Thrust previously interpreted offshore to the east of Poh Head cannot be identified. In the areas where the thrust was interpreted there is a north-vergent thrust and fold zone overlain by almost undeformed sediments. Gently dipping strata of the Banggai-Sula microcontinent margin can be traced northwards beneath younger rocks. In the east, rocks of the Molucca Sea collision complex are deformed by multigenerational folds, thrusts and strike-slip faults. There is a series of small thrusts between the leading edge of the collision complex and the foot of the slope. In the west a zone of transpression close to the East Arm of Sulawesi is the termination of the dextral strike-slip Balantak Fault extending east from Poh Head.

Abstract The Savu Basin is located in the Sunda–Banda fore-arc at the position of change from oceanic subduction to continent–arc collision. It narrows eastward and is bounded to the west by the island of Sumba that obliquely crosses the fore-arc. New seismic data and published geological observations are used to interpret Australia–Sundaland convergence history. We suggest the basin is underlain by continental crust and was close to sea level in the Early Miocene. Normal faulting in the Middle Miocene and rapid subsidence to several kilometres was driven by subduction rollback. Arc-derived volcaniclastic turbidites were transported ESE, parallel to the Sumba Ridge, and then NE. The ridge was elevated as the Australian continental margin arrived at the Banda Trench, causing debris flows and turbidites to flow northwards into the basin which is little deformed except for tilting and slumping. South of the ridge fore-arc sediments and Australian sedimentary cover were incorporated in a large accretionary complex formed as continental crust was thrust beneath the fore-arc. This is bounded to the north by the Savu and Roti Thrusts and to the south by a trough connecting the Java Trench and Timor Trough which formed by south-directed thrusting and loading.

Abstract SE Asian carbonate formations have been reviewed with the aim of understanding the influence of tectonics on their development and reservoir potential through the Cenozoic. Regional tectonics, via plate movements, extensional basin formation, and uplift, was the dominant control on the location of carbonate deposits. These processes controlled the movement of shallow marine areas into the tropics, together with their emergence and disappearance. Although ∼ 70% of the 250 shallow marine carbonate formations in SE Asia were initiated as attached features, 90% of economic hydrocarbon discoveries are in carbonate strata developed over antecedent topography, of which more than 75% were isolated platforms. Faulted highs influenced the siting of nearly two thirds of carbonates developed over antecedent topography. Around a third of carbonate units formed in intra- and interarc areas; however, economic reservoirs are mainly in backarc and rift-margin settings (∼ 40% each). Carbonate edifices show evidence of syntectonic sedimentation through: (1) fault-margin collapse and resedimentation, (2) fault segmentation of platforms, (3) tilted strata and differential generation of accommodation space, and (4) modification of internal sequence character and facies distribution. The demise of many platforms, particularly those forming economic reservoirs, was influenced by tectonic subsidence, often in combination with eustatic sea-level rise and environmental perturbations. Fractures, if open or widened by dissolution, enhance reservoir quality. However, fracturing may also result in compartmentalization of reservoirs through formation of fault gouge, or fault leakage via compromised seal integrity. This study will help in reservoir prediction in complex tectonic regions as the petroleum industry focuses on further exploration and development of economically important carbonate reservoirs.

Abstract The Eurasian margin in SE Asia is a geologically complex region situated at the edge of the Sundaland continent, and is mainly within Indonesia. The external margins of Sundaland are tectonically active zones characterized by intense seismicity and volcanic activity. The region is an obvious modern analogue for older orogens, with a continental core reassembled from blocks rifted from Gondwana, and surrounded by subduction zones for much of the Mesozoic and Cenozoic. It is a mountain belt in the process of formation, and contains many features typically associated with older Pacific margin orogens: there is active subduction, transfer of material at subduction and strike-slip boundaries, collision of oceanic plate buoyant features, arcs and continents, and abundant magmatism. The orogenic belt surrounds Sundaland and stretches from Sumatra into eastern Indonesia and the Philippines. The orogen changes character and width from west to east. Its development can be tectonically described only in terms of several small plates and it includes several suture zones. The western part of the orogenic belt, where the Indian plate is subducted beneath continental crust, is a relatively narrow single suture. Further east the orogenic belt includes multiple sutures and is up to 2000 km wide; there is less continental crust and more arc and ophiolitic crust, and there are several marginal oceanic basins. The orogen has grown to its present size during the Mesozoic and Cenozoic as a result of subduction. Continental growth has occurred in an episodic way, related primarily to arrival of continental fragments at subduction margins, after which subduction resumed in new locations. There have been subordinate contributions from ophiolite accretion, and arc magmatism. Relatively small amounts of material have been accreted during subduction from the downgoing plate. In eastern Indonesia the wide plate boundary zone includes continental fragments and several arcs, but the arcs are most vulnerable to destruction and disappearance. Rollback in the Banda region has produced major extension within the collision zone, but future contraction will eliminate most of the evidence for it, leaving a collage of continental fragments, similar to the older parts of Sundaland.