The San Miguel strand of the San Miguel-Vallecitos fault zone displaces an alluvial ridge 23+5/−10 m in the vicinity of Dolores Mountain. The same section of fault produced surface rupture during the M6.8 earthquake of 1956. We estimate a minimum age of the ridge at between 110+150/−60 ka by comparing soil development on the offset ridge to two soil chronosequences in southern California. Dividing the measured offset by the age of the ridge places a maximum range on the fault slip rate of 0.2+0.35/−0.15 mm/yr. Excavations along and across the fault at the same site revealed at least 80 cm and probably 115 cm of right-lateral offset of a now-buried stream channel deposit. We attribute the 115-cm offset to the 1956 M6.8 earthquake, a displacement that is slightly larger than previously reported for this earthquake. Dividing the 80 to 115 cm of displacement by the calculated fault slip rate yields an estimated return time for similar-sized events of about 5.8+17/−4.3 ka. Similarly, dividing the slip expected for rupture of the entire 160-km length of the San Miguel-Vallecitos fault zone by the fault slip rate yields an estimate of return time of Mw7.8 earthquakes of about 80 ka. The slip rate determined from this study is at least an order of magnitude less than that contributed by the Agua Blanca fault and indicates that the San Miguel fault zone transfers less than about 1% of the plate motions.
The San Miguel fault shows a complex fault trace and registers a small value of cumulative geologic offset (maximum of 0.6 km). When combining the geological estimates of magnitude and return time of the largest earthquakes with magnitude-frequency data recorded along the San Miguel fault by the RESNOR seismic network during the period 1976 to 1991, we observe that the shape of the magnitude-frequency distribution along the fault may be described by the Gutenberg-Richter relationship Log n = a − bM. In contrast, along-strike-slip faults in southern California that are characterized by orders of magnitude more cumulative offset and less complex fault traces, we observe that extrapolation of earthquake frequency statistics from the instrumental record of seismicity severely underestimates the rate of occurrence of maximum expected events. We speculate that the relatively high productivity of small earthquakes along the San Miguel fault reflects a more heterogeneous stress field associated with the incipient and complex nature of the fault trace.