The Quaternary Silver Creek Fault Beneath the Santa Clara Valley, California

Abstract

The northwest-trending Silver Creek Fault is a 40-km-long strike-slip fault in the eastern Santa Clara Valley, California, that has exhibited different behaviors within a changing San Andreas Fault system over the past 10-15 Ma. Quaternary alluvium several hundred meters thick that buries the northern half of the Silver Creek Fault, and that has been sampled by drilling and imaged in a detailed seismic reflection profile, provides a record of the Quaternary history of the fault. We assemble evidence from areal geology, stratigraphy, paleomagnetics, ground-water hydrology, potential-field geophysics, and reflection and earthquake seismology to determine the long history of the fault in order to evaluate its current behavior. The fault formed in the Miocene more than 100 km to the southeast, as the southwestern fault in a 5-km-wide right step to the Hayward Fault, within which the 40-km-long Evergreen pull-apart basin formed. Later, this basin was obliquely cut by the newly recognized Mt. Misery Fault to form a more direct connection to the Hayward Fault, although continued growth of the basin was sufficient to accommodate at least some late Pliocene alluvium. Large offset along the San Andreas-Calaveras-Mt Misery-Hayward Faults carried the basin northwestward almost to its present position when, about 2 Ma, the fault system was reorganized. This led to near abandonment of the faults bounding the pull-apart basin in favor of right slip extending the Calaveras Fault farther north before stepping west to the Hayward Fault, as it does today. Despite these changes, the Silver Creek Fault experienced a further 200 m of dip slip in the early Quaternary, from which we infer an associated 1.6 km or so of right slip, based on the ratio of the 40-km length of the strike-slip fault to a 5-km depth of the Evergreen Basin. This dip slip ends at a mid-Quaternary unconformity, above which the upper 300 m of alluvial cover exhibits a structural sag at the fault that we interpret as a negative flower structure. This structure implies some continuing strike slip on the Silver Creek Fault in the late Quaternary as well, with a transtensional component but no dip slip. Our only basis for estimating the rate of this later Quaternary strike slip on the Silver Creek Fault is to assume continuation of the inferred early Quaternary rate of less than 2 mm/yr. Faulting evident in a detailed seismic reflection profile across the Silver Creek Fault extends up to the limit of data at a depth of 50 m and age of about 140 ka, and the course of Coyote Creek suggests Holocene capture in a structural depression along the fault. No surface trace is evident on the alluvial plain, however, and convincing evidence of Holocene offset is lacking. Few instrumentally recorded earthquakes are located near the fault, and those that are near its southern end represent cross-fault shortening, not strike slip. The fault might have been responsible, however, for two poorly located moderate earthquakes that occurred in the area in 1903. Its southeastern end does mark an abrupt change in the pattern of abundant instrumentally recorded earthquakes along the Calaveras Fault-in both its strike and in the depth distribution of hypocenters-that could indicate continuing influence by the Silver Creek Fault. In the absence of convincing evidence to the contrary, and as a conservative estimate, we presume that the Silver Creek Fault has continued its strike-slip movement through the Holocene, but at a very slow rate. Such a slow rate would, at most, yield very infrequent damaging earthquakes. If the 1903 earthquakes did, in fact, occur on the Silver Creek Fault, they would have greatly reduced the short-term future potential for large earthquakes on the fault.