Evolution of the boundary between the Philippine Sea Plate and Australia: palaeomagnetic evidence from eastern Indonesia

Abstract

The boundary between the Philippine Sea and Australian plates is the left-lateral Sorong Fault system of eastern Indonesia. Until recently, modelling this boundary for the period before about 5 Ma was difficult; the Tertiary motion of the Philippine Sea Plate was uncertain and palaeomagnetic data from areas adjacent to the fault were lacking. Recent geological and palaeomagnetic studies of the area north of the Sorong Fault have elucidated the Tertiary motion history of the Philippine Sea Plate, providing a reference for examining movements within the fault system. We report new palaeomagnetic data from within the Sorong Fault Zone, from the islands of Taliabu and Obi. Taliabu is part of the Sula Platform and is considered to be derived from Australia. Pelagic limestones from the Upper Cretaceous Tanamu Formation of Taliabu yielded a direction of D = 329.1°, I = −34.9° implying counter-clockwise rotation and a formation latitude of 19 ± 5°S. Sula and Misool are postulated to be part of a single microcontinent which had a different Late Cretaceous-mid-Tertiary movement history from Australia. The Sula Platform was transported to its present position by movement along the Sorong Fault system in the Late Miocene. Obi includes rocks of Philippine Sea and Australian origin; all the new sites are in rocks of Philippine Sea Plate origin. Since the Early Neogene the Philippine Sea Plate, which includes all islands north of the Sorong Fault, has rotated 40° clockwise and moved 10–15° northwards. Philippine Sea Plate rocks within the Sorong Fault Zone record similar latitude shifts, but different rotations. In north Obi, the Upper Oligocene Anggai River Formation and the Middle Miocene Woi Formation record ∼ 60° and ∼ 30° counter-clockwise rotations, respectively. The sense of rotation is consistent with motion within a left-lateral fault system, with the Philippine Sea and Australian plates providing the shear couple. In contrast, the Woi Formation in southeast Obi records 15–20° clockwise rotation; this area is separated from the zone of counter-clockwise movement in north Obi by a strand of the Sorong Fault. Arc volcaniclastic rocks from the Upper Cretaceous Leleobasso Formation of northwest Obi have a primary magnetisation with a mean direction of D = 357.1°, I = −21.9°. These rocks formed at ∼ 11°N or ∼ 11°S, depending on the interpreted rotation history, and indicate a Pacific rather than Indian Ocean origin. A volcanic arc at the southern edge of the Philippine Sea Plate collided with eastern New Guinea at ∼ 25 Ma. The Philippine Sea-Australia plate boundary then changed from subduction to strike-slip, as the Philippine Sea Plate began its Neogene rotation, initiating the Sorong Fault system. We suggest that many of the arc fragments in the New Guinea orogenic belt originated in the southern Philippine Sea Plate arc which has subsequently been dismembered by strike-slip faulting.