Paleogeographic reconstruction and origin of the Philippine Sea

Our reconstruction of the Philippine Sea suggests that it formed by two distinct episodes of back-arc spreading, each of which resulted from seaward retreat of the trench. In the first episode, the protoIzu-Bonin Trench retreated northward and the West Philippine Basin formed behind the northern half of the Palau-Kyushu Ridge. In the second episode, the Izu-Mariana Trench retreated eastward and the Shikoku and Parece Vela Basins formed behind it. During the last 17 Ma, the Philippine Sea basin has been moving northwestward with respect to Eurasia shifting the TTT triple junction off central Japan westward by about 50 km. The motion of the Philippine Sea with respect to Eurasia at the triple junction changed from north-northwestward to west-northwestward 10-5 Ma ago. For the period before 17 Ma ago, we construct two models, retreating trench model and anchored slab model. The Izu-Bonin Trench migrated from south to northeast rotating in a clock-wise sense since 48 Ma ago in the retreating trench model. In the anchored slab model, the trench has been fixed with respect to Eurasian margin since 43 Ma ago. We prefer the retreating trench model because the deformation of the plate boundary along the eastern margin of Eurasia during 30-17 Ma ago is much simpler for this model than for the anchored slab model. Furthermore rotations of the Bonin-Mariana islands are consistent with those predicted from the retreating trench model. The 48 Ma ages of the northern part of the Palau-Kyushu Ridge and of Chichi-Jima of the Bonin Islands indicate that there was subduction beneath the northern half of the ridge beginning at least 48 Ma ago. From this and the subparallelism in trend between the northern part of the Palau-Kyushu Ridge and the Central Basin Ridge, we propose that the major part of the West Philippine Basin formed by back-arc spreading in a N-S direction behind the northern part of the Palau-Kyushu Ridge. The Pacific plate was moving northward with respect to hot-spots from 48 to 43 Ma ago, which implies that the Pacific plate is not likely to have been subducting beneath the West Philippine Basin during this time. We speculate that another plate existed south of the Pacific plate and thai it was subducting beneath both tile West Philippine Basin and Australia. The annihilation of this plate might he a cause for the sudden change of the Pacific plate motion at 43 Ma ago.