Earthquake supercycles on the Mentawai segment of the Sunda megathrust in the seventeenth century and earlier

Abstract

Over at least the past millennium, the Mentawai segment of the Sunda megathrust has failed in sequences of closely timed events rather than in single end-to-end ruptures—each the culmination of an earthquake “supercycle.” Here we synthesize the sixteenth- and seventeenth-century coral microatoll records into a chronology of interseismic and coseismic vertical deformation. We identify at least five discrete uplift events in about 1597, 1613, 1631, 1658, and 1703 that likely correspond to large megathrust ruptures. This sequence contrasts with the following supercycle culmination, which involved only two large ruptures in 1797 and 1833. Fault slip modeling suggests that together the five cascading ruptures involved failure of the entire Mentawai segment. Interseismic deformation rates also changed after the onset of the rupture sequence, as they did after the 1797 earthquake. We model this change as an altered distribution of fault coupling, presumably triggered by the ~1597 rupture. We also analyze the far less continuous microatoll record between A.D. 1 and 1500. While we cannot confidently delineate the extent of any megathrust rupture during that period, all evidence suggests that individual major ruptures involve only part of the Mentawai segment, often overlap below the central Mentawai Islands, often trigger coupling changes, and occur in clusters that cumulatively cover the entire Mentawai segment at the culmination of each supercycle. It is clear that each Mentawai rupture sequence evolves uniquely in terms of the order and grouping of asperities that rupture, suggesting heterogeneities in fault frictional properties at the ~100 km scale.