This research is to complete analysis on coral data gathered earlier between the western coast of Sumatra and the trench of the Sumatran subduction zone, 240 km offshore. The samples consist of slabs of head corals that record sea-level changes caused by tectonic deformation related to earthquake occurrence. The nature of earthquake recurrence is poorly understood, in large part because physical models are constrained inadequately by sparse paleoseismic and paleogeodetic records and by instrumental data. This work has shown that coral "microatolls" offer an unrivalled opportunity to test models of earthquake recurrence, because they record vertical deformation through several large-earthquake cycles quite precisely in their morphology and annual banding. On the islands of the outer-arc Sumatran ridge, 120 to 130 km from the trench and about 25 km above the subduction interface, this project has shown sudden emergence of about 2 meters during the giant earthquake of 1833 and during two earlier events about AD 1608 and 1380. This implies 10 to 13 meters of slip on the subduction interface during each of these events. There is also evidence of episodic, interseismic submergence of the islands at rates that increase trenchward from about 4 to about 10 mm/yr, averaged over the past several decades and in the decades prior to each large earthquake. This grant will support the development of plausible physical models to explain the interseismic and coseismic deformation. This research is a component of the National Earthquake Hazard Reduction Program.
