Cenozoic geological and plate tectonic evolution of SE Asia and the SW Pacific: computer-based reconstructions, model and animations

Abstract A plate tectonic model for the Cenozoic development of the region of SE Asia and the SW Pacific is presented and its implications are discussed. The model is accompanied by computer animations in a variety of formats, which can be viewed on most desktop computers. GPS measurements and present seismicity illustrate the high rates of motions and tectonic complexity of the region, but provide little help in long-term reconstruction. Plate boundaries shifted rapidly in the Cenozoic. During convergence of the major plates, there were numerous important episodes of extension, forming ocean basins and causing subsidence within continental regions, probably driven by subduction. Within eastern Indonesia, New Guinea and the Melanesian arcs, there are multiple Cenozoic sutures, with very short histories compared to most well-known older orogenic belts. They preserve a record of major changes in tectonics, including subduction polarity reversals, elimination of volcanic arcs, changing plate boundaries and extension within an overall contractional setting. Rapid tectonic changes have occurred within periods of less than 5 Ma. Many events would be overlooked or ignored in older orogenic belts, even when evidence is preserved, because high resolution dating is required to identify them, and the inference of almost simultaneous contraction and extension seems contradictory. There were three important periods in regional development: at about 45, 25 and 5 Ma. At these times, plate boundaries and motions changed, probably because of major collision events. The 45 Ma plate reorganisation may be related to India–Asia collision, although some important intra-Pacific events, such as voluminous Eocene boninite magmatism, seem to be older and require other causes. Indentation of Asia by India modified the Asian continent, but there is little indication that India has been the driving force of tectonics in most of SE Asia. The most important Cenozoic plate boundary reorganisation was at about 25 Ma. The New Guinea passive margin collided with the East Philippines–Halmahera–South Caroline Arc system. The Australian margin, in the Bird's Head region, also began to collide with the SE Asian margin in Sulawesi. The Ontong Java Plateau collided with the Melanesian Arc. These collisions caused a major change in the character of plate boundaries between about 25 and 20 Ma. Since 25 Ma, tectonic events east of Eurasia were driven by motion of the Pacific Plate. Further, west, the movement of Australia northwards caused rotations of blocks and accretion of microcontinental fragments to SE Asia. Plate motions and boundaries changed again at about 5 Ma, for uncertain reasons, possibly as a consequence of Pacific Plate motion changes, arc–continent collision in Taiwan, or other boundary changes at the Pacific margin, for example in the Philippines. Areas to the west and east of New Guinea, the Banda Sea and Woodlark Basin, illustrate the speed of change, the unexpected interplay of convergence and extension, and the importance of subduction as the engine of change. Subduction has been the principal driving mechanism for tectonic change, although its manifestations are varied. They include collision-related phenomena, partitioning of oblique convergence, and effects of hinge roll-back and pull forces of subducting slabs. Magmatism is not always associated with subduction, depending on the movement of subduction hinge, and there may be important extension of the upper plate both perpendicular and parallel to the length of subduction zones. Strike-slip faulting is observably very important within the Pacific–Australia–Eurasia convergent setting, yet appears in few tectonic models. Long-term strike-slip deformation can explain some of the complexities of areas such as New Guinea, including magmatism and its absence, and thermo-chronological data showing very young and rapid cooling of the mobile belt and fold belt. The inadequacies of the tectonic model reflect in part the difficulties of applying rigid plate tectonics, when there is clear evidence of changing shapes of fragments. Geological knowledge of the region is still inadequate and significant improvements to regional data sets, such as palaeomagnetic data and isotopic ages, are required. New tomographic techniques offer an important means of testing this and other reconstructions. However, valuable insights could also be obtained from simple data sets, such as sediment volumes, if more information that is complete were available in the public domain. Two-dimensional plate tectonic cartoons of small areas are no longer adequate descriptions or tools for understanding. It is essential to test plate tectonic models by using animation techniques with reconstructions drawn at short time intervals, which expose flaws in models, show major gaps in knowledge and help identify truly regional events. Observations of present-day tectonics, and all geological evidence, indicate that the model presented here is over-simplified. Improvements in this, or new models, will inevitably be more complex than the reconstructions described here.