Reconstructing subducted oceanic lithosphere by “reverse-engineering” slab geometries: The northern Philippine Sea Plate

Abstract

Subducting slabs commonly acquire complex geometries from the migration of subduction trenches, slab-mantle interaction, slab tearing, and collision of slabs at depth. Although it is possible to construct three-dimensional models of subducted slabs using earthquake hypocenter locations and tomographic models, it is often not possible to rigorously test their accuracy. Here we present a methodology for performing such a test, by “reverse-engineering” the presubduction configuration of a slab of oceanic lithosphere from interpretations of its present-day morphology. We illustrate our approach for the Ryukyu and Shikoku slabs, northwest Philippine Sea, having simulated them as viscoelastic sheets that we unfolded and “floated” to the surface. The net strain distribution of the floated mesh indicated which parts of the original slab model were geometrically viable (minimal net strain) and which parts of the mesh required additional tears and/or zones of localized ductile extension to have enabled the slab to deform during subduction. In the instance of the Ryukyu and Shikoku slabs, the Palau-Kyushu and Gagua ridges are shown to have both acted as planes of weakness that broke into major vertical slab tears. These subducted ridges are connected by a trench-parallel tear that represented the former contact between the Huatung and West Philippine Basins. The fossil spreading center of the Shikoku Basin formed a separate slab window upon subduction along the Nankai Trough. The methodology presented herein is a powerful tool to evaluate complex slab morphologies, infer the locations of slab tears, and therefore reconstruct intricate configurations of subducted oceanic lithosphere.