Relationship between seismicity and geologic structure in the Southern California region

Abstract

Data from 10,126 earthquakes that occurred in the southern California region between 1934 and 1963 have been synthesized in the attempt to understand better their relationship to regional geologic structure, which is here dominated by a system of faults related mainly to the San Andreas system. Most of these faults have been considered “active” from physiographic evidence, but both geologic and short-term seismic criteria for “active” versus “inactive” faults are generally inadequate. Of the large historic earthquakes that have been associated with surficial fault displacements, most and perhaps all were on major throughgoing faults having a previous history of extensive Quaternary displacements. The same relationship holds for most earthquakes down to magnitude 6.0, but smaller shocks are much more randomly spread throughout the region, and most are not clearly associated with any mappable surficial faults. Virtually all areas of high seismicity in this region fall within areas having numerous Quaternary fault scarps, but not all intensely faulted areas have been active during this particular 29-year period. Strain-release maps show high activity in the Salton trough, the Agua Blanca-San Miguel fault region of Baja California, most of the Transverse Ranges, the central Mojave Desert, and the Owens Valley-southern Sierra Nevada region. Areas of low activity include the San Diego region, the western and easternmost Mojave Desert, and the southern San Joaquin Valley. Because these areas also generally lack Quaternary faults, they probably represent truly stable blocks. In contrast, regions of low seismicity during this period that show widespread Quaternary faulting include the San Andreas fault within and north of the Transverse Ranges, the Garlock fault, and several quiescent zones along major faults within otherwise very active regions. We suspect that seismic quiescence in large areas may be temporary and that they represent likely candidates for future large earthquakes. Without more adequate geodetic control, however, it is not known that strain is necessarily accumulating in all of these areas. Even in areas of demonstrated regional shearing, the relative importance of elastic strain accumulation versus fault slippage is unknown, although slippage is clearly not taking place everywhere along major “active” faults of the region. Recurrence curves of earthquake magnitude versus frequency are presented for six tectonically distinct 8500-km^2 areas within the region. They suggest either that an area of this small size or that a sample period of only 29 years is insufficient for establishing valid recurrence expectancies; on this basis the San Andreas fault would be the least hazardous zone of the region, because only a few small earthquakes have occurred here during this particular period. Although recurrence expectancies apparently break down for these smaller areas, historic records suggest that the calculated recurrence rate of 52 years for M = 8.0 earthquakes for the entire region may well be valid. Neither a fault map nor the 29-year seismic record provides sufficient information for detailed seismic zoning maps; not only are many other geologic factors important in determining seismic risk, but the strain-release or epicenter map by itself may give a partially reversed picture of future seismic expectance. Seismic and structural relationships suggest that the fault theory still provides the most satisfactory explanation of earthquakes in this region.