Analysis of Oblique Plate Convergence along the Manila Trench and the Philippine Trench

The Manila Trench, a >1,200 km long, seismically active N-S trending trench located along the western margin of the Philippine archipelago, acts as the main convergence zone between the Philippine Mobile Belt (PMB) and the Sunda Block (a mobile fragment of the Eurasian Plate). We compare the ongoing subduction along the Manila Trench with that occurring along the opposing Philippine Trench/East Luzon Trough in the east, the boundary which separates the Philippine Mobile Belt from the Philippine Sea Plate. We use joint inversions of published geodetic velocity fields and focal mechanism data to obtain best-fit kinematic block models. From seismicity data, known geometries of faults and subduction dip angles, we construct models of tectonic blocks and their boundaries. We model the Manila and Philippine Trenches as opposite-dipping planes that confine rotating elastic blocks composing the Philippine Mobile Belt. We find that the convergence rate along the Manila Trench decreases progressively southwards, from >70 mm/yr near 19° N, to less than 20 mm/yr at its southern termination at Mindoro Island (~13° N). The systematic slowing reflects the ongoing collision process between Mindoro and the Palawan block; this region acts as a fulcrum, resulting in as much as 7°/My counterclockwise rotation of blocks of the Luzon arc with respect to the Sunda block. The near-orthogonal convergence along the Manila Trench contrasts with the ~40 mm/yr oblique convergence at the Philippine Trench, where the convergence angle changes from nearly westward in the southern PMB to NW near its northern termination in SE Luzon. We find that the Manila Trench is relatively poorly coupled compared to the moderately coupled Philippine Trench. Based on the rates and directions of convergence, we find that the northern Philippine Fault system accommodates the shear component of convergence along the two margins of the PMB, resulting in strain partitioning. The lower inferred seismic coupling along the Manila Trench as compared to the Philippine Trench may explain its relative low rate of historical seismic moment release in earthquakes. However, additional GPS sites close to the Manila and Philippine trenches are needed to reliably resolve coupling rates along these subduction boundaries.