cal record includes a number of moderate to large earthquakes, however. The greatest seismic hazard in the southeastern United States is thought to be near Charleston, South Carolina. In 1886, an M 7.3 earthquake caused widespread damage and liquefaction in a broad area. Recent studies of paleoliquefaction features indicate that earthquakes the size of the 1886 event occur about every 600 years (52). The geological structure responsible for these events is still uncertain, although satellite images and river drainage deflections suggest a linear feature that could be the expression of a fault in the basement beneath the thick coastal plain sediments (53). Seismicity in eastern Tennessee encompasses the cities of Knoxville and Chattanooga (54) in a 200-kilometer-long, north-east-trending band. No historic earthquakes greater than body wave magnitude (mb) 4.5 have occurred within this band, but the high seismicity is consistent with the potential occurrence of larger events. Other notable earthquakes in the southeast include the 1897 Giles County earthquake (mb 5.5) in western Virginia and the 1916 Jefferson County (mb 5.1) event near Birmingham, Alabama.
The largest historic event in the northeastern United States was the Cape Ann earthquake (mb 6) off the coast of Massachusetts in 1755. A repeat of this event would have serious consequences for Boston, which contains numerous older, more vulnerable structures. Similar concerns have been raised for the New York metropolitan area, which experienced moderate (M 5) earthquakes in 1737 and 1884. Analysis of seismicity along the Boston-to-Washington corridor suggests that an mb 6 or greater event should occur about every 400 years (55). Such an earthquake would likely cause substantial damage.
Unlike plate boundary zones, where the source of the deformation is fairly clear, the seismicity of the central and eastern United States arises from deformations that are poorly described and not well understood. They are most commonly associated with relic geologic features on the southern periphery of the ancient Canadian craton, inherited from previous episodes of plate tectonics. These epicratonic features often comprise buried fault systems of formerly active plate boundaries. To understand why they have been reactivated as weak structures in present-day tectonics will require better data from regional seismic and geodetic networks, along with more extensive paleoseismic mapping at the surface and subsurface imaging by active geophysical techniques.
The hazard map of Alaska is dominated by the Alaska-Aleutian megathrust, the longest fault zone with the highest rate of seismicity in the United States (Figure 3.14). It stretches 3600 kilometers from Kam-