Geological Survey of Japan, Japan

Articulating a complete and comprehensive plate tectonic theory requires understanding how new subduction zones form (subduction initiation, SI). Because SI is a tectonomagmatic singularity with no active examples, reconstructing SI is challenging. There is debate about whether SI requires compression as many geodynamicists suggest or can happen spontaneously, due to the excess density of old oceanic lithosphere alone. Studies of intra-oceanic forearcs provide clear and simple constraints for understanding SI, because forearc crust forms during SI and preserves a high-fidelity magmatic and stratigraphic record. We have heretofore been remarkably ignorant of this record, because the "naked forearcs" which expose SI crustal sections are distant from continents and lie in the deep trenches, making study and sampling expensive and difficult. Geophysical reflection and refraction studies cannot yet provide the level of detail needed, so samples of SI crust must be collected via "3D" dredging, diving, and drilling. Key advances in understanding forearc crustal structure were made as a result of DSDP Leg 60 and ODP Legs 125 and 126 expeditions to the Izu-Bonin-Mariana (IBM) forearc in the 1970's and 1980's; more forearc drilling is needed. Our studies of the IBM forearc (using Shinkai 6500) indicate that SI is accompanied by seafloor spreading in what ultimately became the forearc of the new convergent margin. IBM SI encompassed approximately 7 m.y. for the complete transition from initial seafloor spreading and eruption of voluminous MORB-like (fore-arc basalts, FAB) to normal arc volcanism, perhaps consistent with how long it might take for slowly subsiding lithosphere to sink approximately 100 km deep and for mantle motions to evolve from upwelling beneath the infant arc to downwelling beneath the magmatic front. Our best estimates are that FAB erupted at about 51-52 Ma, boninites began erupting around 48 Ma, and normal arc lavas began erupting around 44-45Ma, heralding establishment of a mature magmatic arc composed of discrete volcanic centers. The SI magmatic stratigraphy of the IBM forearc is seen in many SSZ ophiolites, consistent with the observation that most ophiolites are obducted forearc fragments. This understanding opens the door for on-land ophiolitologists to contribute fundamentally to understanding SI.