dued geomorphic expression of deformation may be due to broad basinal or domal uplift. These lower tectonic rates are generally accompanied by less earthquake activity than for interplate regions. In general this has caused more focus and awareness of earthquake hazard in interplate regions and less concern in the intraplate regions. An excellent summary of intraplate seismicity (and other intraplate tectonism) is given by Sykes (1978), who noted that intraplate seismicity appears to be localized along pre-existing zones of crustal weakness. Seismicity in the eastern United States within the North American plate exemplifies this concept.
Seismic activity in the New Madrid area, where the large earthquakes of 1811 and 1812 occurred, appears to be localized along a pre-existing rift structure in the continent (Sykes, 1978). Although at the time the rift was formed the continent was under extension in that region, recent activity appears to be related to compression. This change in stress regime shows how a pre-existing structure (e.g., rift) can be reactivated in later, different strain episodes. The crustal weakness appears to be the locus of earthquake-related strain release. The 1886 Charleston earthquake was located in a Paleozoic orogenic belt, an inactive or relic interplate region on the eastern edge of the continent. Low seismicity and cover of young sediments has not allowed the tectonics and source zones of the Charleston earthquake to be understood as well as in the New Madrid area.
The Meers Fault of Oklahoma is another example of reactivation of an ancient geologic structure. The Meers Fault, a segment of the Frontal Fault zone, was formed about 500 m.y. ago as part of the southern Oklahoma rift. During a later orogeny it underwent extensive compressional (reverse) as well as left-lateral displacement. Studies of the Meers Fault show that the fault is currently active and has a dominant lateral component (Ramelli and Slemmons, 1985; Slemmons et al., 1985).
The potential size of intraplate earthquakes, although they are infrequent, is great. The New Madrid, Missouri, earthquakes of 1811 and 1812 are estimated to have MS magnitudes of over 8; the Charleston, South Carolina, earthquake of 1886 was about MS=7. The recently discovered Meers Fault has characteristic features that suggest MS=6.5 to 7.5 earthquakes in the late Quaternary (Slemmons et al., 1985).
Many active faults are poorly delineated on most pre-1950 geologic, structural, or tectonic maps. These commonly emphasize ancient and inactive tectonic features rather than neotectonic structures. Recognition and detailed mapping of historical and Quaternary faults in many zones of neotectonic activity, particularly at or near plate boundaries, have led to recent improvements in the delineation of active faults. In addition, the possible reactivation of intraplate faults such as the Meers Fault has emphasized the need to re-examine other faults and folds in central and eastern United States.
Remote-sensing methods can be effective in detecting, delineating, and describing the character of active faults and neotectonic features. The most effective methods accentuate fault scarps by employing imaging techniques using special illumination angles, wavelengths, or stereographic effects. Some of the methods for earthquake-hazard analysis are summarized by Glass and Slemmons (1978), but newer equipment and methods and rapid developments in analytical techniques require continued adaptation of many of these concepts. Examples of instrumental and structural analysis are in Williams (1983) in the section on geological applications, and for applications in nuclear power-plant site investigations in McEldowney and Pascucci (1979).
Low Sun angle and radar imagery methods are especially effective in detecting and delineating active faults and folds. Special low-Sun angle photography of faults can have the advantages of relatively low cost and appropriate scale and optimum shadowing or highlighting of scarps by selection of solar azimuth and altitude. Since this is the most effective single method of assessment of active faults, it is one of the most widely used methods for aerial photography and reconnaissance (Glass and Slemmons, 1978). Radar imagery of some areas is available at much higher cost and on smaller scales, may have the advantages of some ground penetration in arid regions, and can be taken at any azimuth, time of day or night, and in cloud or fog cover.
Recent studies using ground penetrating radar for fault trace identification have been very successful. Black et al. (1983) studied an area along the San Andreas Fault zone that has been extensively trenched and found good correlation of the ground penetrating radar records and the trench logs. Bilham and Seeber (1985) used subsurface radar profiling to detect colluvial wedges associated with former movements along the Lost River Fault and wide zones of faulting along the San Andreas Fault system. As this method is refined it will become an even more powerful method of fault detection and delineation.