Potentially destructive earthquakes are inevitable in the Los Angeles region of California, but hazards prediction can provide a basis for reducing damage and loss. This volume identifies the principal geologically controlled earthquake hazards of the region (surface faulting, strong shaking, ground failure, and tsunamis), summarizes methods for characterizing their extent and severity, and suggests opportunities for their reduction. Two systems of active faults generate earthquakes in the Los Angeles region: northwest-trending, chiefly horizontal-slip faults, such as the San Andreas, and west-trending, chiefly vertical-slip faults, such as those of the Transverse Ranges. Faults in these two systems have produced more than 40 damaging earthquakes since 1800. Ninety-five faults have slipped in late Quaternary time (approximately the past 750,000 yr) and are judged capable of generating future moderate to large earthquakes and displacing the ground surface. Average rates of late Quaternary slip or separation along these faults provide an index of their relative activity. The San Andreas and San Jacinto faults have slip rates measured in tens of millimeters per year, but most other faults have rates of about 1 mm/yr or less. Intermediate rates of as much as 6 mm/yr characterize a belt of Transverse Ranges faults that extends from near Santa Barbara to near San Bernardino. The dimensions of late Quaternary faults provide a basis for estimating the maximum sizes of likely future earthquakes in the Los Angeles region: moment magnitude .(M) 8 for the San Andreas, M 7 for the other northwest-trending elements of that fault system, and M 7.5 for the Transverse Ranges faults. Geologic and seismologic evidence along these faults, however, suggests that, for planning and designing noncritical facilities, appropriate sizes would be M 8 for the San Andreas, M 7 for the San Jacinto, M 6.5 for other northwest-trending faults, and M 6.5 to 7 for the Transverse Ranges faults. The geologic and seismologic record indicates that parts of the San Andreas and San Jacinto faults have generated major earthquakes having recurrence intervals of several tens to a few hundred years. In contrast, the geologic evidence at points along other active faults suggests recurrence intervals measured in many hundreds to several thousands of years. The distribution and character of late Quaternary surface faulting permit estimation of the likely location, style, and amount of future surface displacements. An extensive body of geologic and geotechnical information is used to evaluate areal differences in future levels of shaking. Bedrock and alluvial deposits are differentiated according to the physical properties that control shaking response; maps of these properties are prepared by analyzing existing geologic and soils maps, the geomorphology of surficial units, and. geotechnical data obtained from boreholes. The shear-wave velocities of near-surface geologic units must be estimated for some methods of evaluating shaking potential. Regional-scale maps of highly generalized shearwave velocity groups, based on the age and texture of exposed geologic units and on a simple two-dimensional model of Quaternary sediment distribution, provide a first approximation of the areal variability in shaking response. More accurate depictions of near-surface shear-wave velocity useful for predicting ground-motion parameters take into account the thickness of the Quaternary deposits, vertical variations in sediment .type, and the correlation of shear-wave velocity with standard penetration resistance of different sediments. A map of the upper Santa Ana River basin showing shear-wave velocities to depths equal to one-quarter wavelength of a 1-s shear wave demonstrates the three-dimensional mapping procedure. Four methods for predicting the distribution and strength of shaking from future earthquakes are presented. These techniques use different measures of strong-motion