One of Silver's principal research interests was seismic anisotropy and its implications for the tectonic evolution of the Earth. He organized and conducted seismic field experiments in northern Canada, southern Africa, Chile and Bolivia, China, and Tibet, as well in California and elsewhere in western North America. Silver and colleagues were the first to conduct, in 1989, a modern portable broadband seismic experiment. This experiment was designed to explore the deep structure of the North American continent, but also formed the starting point for the development of novel methods of seismological investigation: with Winston Chan, Martha Savage, and other colleagues, Silver elaborated on earlier work and deduced from the measurements the splitting of shear waves, a type of seismic anisotropy, for areas of the size of tectonic plates in order to determine the patterns of convection in the upper mantle and the deformation history of the continental and subcontinental lithosphere that record how the continent grew and evolved. This approach has been developed ever since and is now in widespread use to study the patterns of convective flow in the Earth’s interior and the processes by which the continents were assembled. His shear-wave splitting studies with Mark Behn and Clint Conrad showed that the pattern of seismic anisotropy under oceanic lithosphere can be explained as being caused by mantle flow driven by plate motions and mantle density heterogeneity. With Behn, he also made the controversial proposition that plate tectonics on Earth is intermittent and may have been temporarily interrupted in the past when subduction largely ceased after the closure of a large ocean basin. An important observation was made possible by his serendipitous observation of the 1994 Bolivia earthquake during a field campaign in the region: the data recorded by his broadband seismograph array showed that the source region of this event, which is the largest deep quake on record, is in conflict with the generally accepted view that such quakes are caused by phase transformations of mantle minerals. In a long-term study of small earthquakes triggered by a large event, the 1992 Landers, California earthquake, he and his colleagues discovered an annual cycle: fall had the greatest number of earthquakes, spring the least. The team found that this pattern could be related to barometric pressure changes: less pressure meant reduced stress on the faults, which permitted them to move more frequently. More recent work by him and his collaborators suggests that changes in the state of stress of the lithosphere induced by a large earthquake can alter the strength of faults and the seismic activity in an earthquake-prone area. In 2008, Silver was co-author of a paper showing there were subtle changes in the speed of seismic waves that preceded two small earthquakes, encouraging results for the field of earthquake forecasting In pursuit of his overarching goal of monitoring the deformation of the lithosphere on a continental scale, Silver played a key role in establishing the Plate Boundary Observatory, a part of the large EarthScope research program, which observes the tectonic activity throughout the western US and Alaska.
