A Model for the Termination of the Ryukyu Subduction Zone against Taiwan A Triple Junction of Collision, Subduction/Separation and Subduction Boundaries

In the vicinity of Taiwan, the Philippine Sea plate moves in the direction of N50^{O}W to collide with the Eurasian continent and at the same time subducts toward the north along the Ryukyu subduction zone. The geometry of the junction between the Ryukyu subduction and Taiwan is obscured by intense deformation near the island. Based on tomographic velocity images for the subduction zone and relocated seismicity the junction is determined to be near the Longitudinal Valley at the latitude of about 23.7^{O}N, i.e., the northern part of the Coastal Range overlies the subduction zone. North of this latitude the subduction becomes deeper and only the portion of the Eurasian lithosphere that is in contact with the subduction zone is engaging in collision. Above the subduction zone the Eurasian plate undergoes stretching due to the shortening of the plate on the Taiwan side and the lack of shortening of the Eurasian plate above the subducting Philippine Sea plate. When the subducting Philippine Sea reaches the depths of the lower part of the lithosphere or asthenosphere then the Eurasian plate is no longer under compression and a bight in the surface structure is created. Adding to these actions is the opening of the Okinawa Trough and the resulting southward migration of the Ryukyu Trench. Because of the opening of Okinawa this junction is a dynamically changing triple junction that involves collision between about 23^{O}N and 23.7^{O}N, collision and separation, with normal and strike-slip motion along the boundary, to its north and subduction along the Ryukyu Trench. The triple junction migrates southward with time. The plate configuration in this junction resembles that in the eastern syntaxis of Himalaya where the Indian plate moves northwestward to collide along the Himalayan front and subducts under Northeastern India and Myanmar. While the Taiwan collision began about four million years the Indian collision has probably gone on for at least a few tens of million years. The Taiwan situation can provide a scenario of the earlier development of the Indian/Eurasian collision regime.