Recent decades have seen dramatic improvement in the ability of earth scientists to resolve the geometries of earthquake sources and to relate these geometries to local and global tectonic processes. Well-studied seismogenic faults consist of individual segments that are offset from each other along strike or that have different strikes. The boundaries between segments help control or limit rupture propagation during large earthquakes. On many segments of the plate-bounding San Andreas fault system, fault orientation and structure at depth are well represented by the surface fault trace. In the Cordillera away from the San Andreas system, faults that have produced large earthquakes seem commonly to be expressed at the surface by traces that are distorted representations of the faults at depth. In the United States east of the Cordillera, it has not been possible to associate earthquakes with geologically mapped faults as directly as in the western United States. The following observations, however, suggest that many eastern earthquakes occur on reactivated pre-Cenozoic faults: (1) epicenters in some sources coincide with mapped pre-Cenozoic faults; (2) in other sources, planar zones of hypocenters and instrumentally inferred fault planes parallel regional structural trends; and (3) some pre-Cenozoic faults show Cenozoic displacement. Global tectonic models have been extended to account for the distribution and focal depths of earthquakes that occur in California away from the principal plate boundary and to account for the orientations of regional stress tensors east of the plate boundary. Plate-tectonic models have also led to the hypothesis that a large thrust-interface earthquake might someday occur in the Pacific Northwest subduction zone, notwithstanding the absence of such earthquakes in the historic record. on July 26, 2015 memoirs.gsapubs.org Downloaded from Earthquakes, faults, and seismotectonic framework, U.S. 543 95 90 85 80 75 70 65the Geological Society of America, Decade of North American Geology (DNAG), seismicity map of North America (Engdahl and Rinehart, 1988), using the same conventions to assign magnitudes to plotted events. Where more than one earthquake has been assigned to the same epicenter (to within 4 km), only a single symbol for the largest earthquake is plotted. on July 26, 2015 memoirs.gsapubs.org Downloaded from on July 26, 2015 memoirs.gsapubs.org Downloaded fromEarthquakes, faults, and seismotectonic framework, U.S.
545North American plates in the Pacific Northwest is a subject of controversy; it, too, may be a temporary quiescence, or it may reflect aseismic subduction.The seismicity of the United States east of the Cordillera, though much lower on average than that of the Cordillera and Pacific Coast, is nonetheless important as an indicator of midplate tectonic processes and as a potential hazard to human beings. The largest historic earthquakes in the central and eastern United States had magnitudes greater than 6.5 and probably involved fault rupture through tens of kilomete...