I first worked in Alaska as a member of the geological team evaluating fault rupture hazard for the Trans-Alaska pipeline. Three decades later, the great Denali earthquake drew me back to study the 325-km long rupture that produced the earthquake. That rekindled my interest in the neotectonics of the Alaska region.

The obliquely convergent Sumatran plate boundary is an extraordinary natural laboratory for understanding the influence of fault geometry on rupture processes and the sequential rupture of active faults. We are working on both principal components of the system - the Sumatran fault and the Sumatran subduction zone.

The Sumatran subduction zone is an exceptional natural laboratory for investigating the behavior of a large fault through several earthquake cycles. This is because the corals fringing the islands of its outer-arc ridge provide an unusually complete and long record of strain accumulation and relief above a subduction interface. As part of the Sumatran Project of the Tectonics Observatory, we have installed a network of continuously recording GPS stations and seismic instruments there. Some of these instruments recorded coseismic and postseismic deformations of the giant 2004 and 2005 earthquakes.

Investigation of fault ruptures and other tectonic deformation immediately following earthquakes has taught us a great deal about the nature of earthquake sources. In particular, we have investigated the Landers earthquake (1992), the Imperial Valley earthquake (1979), the Chi-Chi, Taiwan, earthquake (1999), the Izmit, Turkey, earthquake (1999), the great Denali earthquake, Alaska (2002), and the giant Sumatran earthquakes (2004 and 2005).