3-DIMENSIONAL FINITE-DIFFERENCE MODELING OF THE SAN-ANDREAS FAULT - SOURCE PARAMETERIZATION AND GROUND-MOTION LEVELS

Olsen et al. (1995) recently simulated an M-w 7.75 earthquake on the S an Andreas fault, predicting long-period (T > 2.5 sec) ground velociti es of 140 cm/sec in the Los Angeles basin, about 60 km from the fault. These motions are much larger than estimates derived from empirical r elations or other numerical simulations. Standard area-magnitude relat ions predict that the 170 x 16 km fault used in the simulations would produce an M-w 7.5 earthquake, giving a moment of 2.0 x 10(27) dyne-cm , which is 2.4 times smaller than the moment used by Olsen et al. (199 5). Further, self-similar scaling predicts a rise time of 3 sec for an M-w 7.75 event and 2.2 sec for an M-w 7.5 event. The filtered impulse slip function used by Olsen et al. (1995) has an effective rise time of 1.6 sec, yielding a response that is about 2 times larger than expe cted for periods less than 5 sec. This combination of high seismic mom ent and short rise time, along with the use of a uniform slip distribu tion, leads to the extreme ground-motion levels predicted by Olsen et al. (1995). To quantify the sensitivity of the long-period ground-moti on response to source parameterization, we have performed 3D finite-di fference (FD) simulations using various combinations of seismic moment , source rise time, and slip heterogeneity. These calculations incorpo rate the same grid dimensions, fault size, and bandwidth employed by O lsen et al. (1995). With a moment of 2.0 x 10(27) dyne-cm, a rise time of 2 sec, and a smoothly heterogeneous slip distribution, we simulate peak long-period ground velocities of 155 cm/sec in the near-fault re gion and 40 cm/sec in the Los Angeles basin. These values are much clo ser to (although still higher than) empirical predictions. A uniform s lip distribution produces the largest peak motions, both in the near-f ault region and in the Los Angeles basin, whereas a rough asperity sli p distribution noticeably reduces the maximum near-fault ground veloci ties. Our results indicate that the accurate simulation of long-period ground motions requires a realistic source parameterization, includin g appropriate choices of seismic moment and rise time, as well as the use of spatial and temporal variations in slip distribution.