A shear-strain anomaly following the Loma Prieta earthquake

Abstract

BOREHOLE tensor strain instruments deployed along the San Andreas fault for the past ten years have provided sufficient resolution and stability to sample regional tectonic processes, potentially enhancing earthquake prediction capability. Data obtained from the instrument at San Juan Bautista, in the near-field region of the 17 October 1989 Loma Prieta earthquake (M_s = 7.1) provide the first opportunity to observe shear strain processes associated with a large earthquake. We previously reported1 a prominent shear-strain anomaly in those data for more than a year before the earthquake. This anomaly ceased immediately after the earthquake, but, as we report here, a new and higher rate of fault-parallel shear accumulation (2 microstrain per year) was established about four months later and has continued to the present. Associated changes in creep rate are apparent at a number of sites on the surface trace of the fault within 30 km of the strain meter. We propose that the observed strain accumulation results from increased slip around a nearby locked section of the fault, this slip arising from loading by the failed Loma Prieta source region to the north. This model is consistent with suggestions of an increased probability of a moderate earthquake near San Juan Bautista^9–10, and with evidence¹² that interactions between fault regions are important in earthquake processes.