An Integrated Analysis of Deformation Data at Parkfield, California: Detection of a Long-Term Strain Transient

More than a decade of high quality data from geodetic and borehole strain instrumentation in the Parkfield, California area (1985-1997) reveals a significant transient in apparent slip-rate along the San Andreas fault. It consists of two phases: decreased slip-rate (by 16 ± 5%) during 1991.0 - 1993.0 followed by increased rate (by 34 ± 6%) from 1993.0 for at least 3 years. This transient is observed on all the deformation data used in this study, including those from 2-color laser geodimeters, borehole strain dilatometers, and creepmeters. The second phase has been observed by tensor strain data (Gwyther et al, 1996). Compared to pre-1991 levels, seismicity is greater during this transient, especially for the period of increased slip-rate, when the four largest earthquakes in the observing period occurred. Since there was also a significant increase in precipitation during the period 1991-1996, compared to pre-1991, we tested for the relative contributions of tectonic and hydrologic influences to the observed signals using the geodimeter data, and found that a tectonic origin was more likely. Using the decade-long secular change in baseline length as a measure of tectonic influence, and the amplitude of the annual cycle as a measure of hydrologic influence, we found that the anomaly is significantly correlated with tectonic influence at the 95% confidence level, but not with hydrologic influence. The transient can be simply modeled as two slip events on the San Andreas Fault. The two color and creep data for phase two (1993-1996) are consistent with a model with 22 mm and 8 mm of slip north and south of Parkfield respectively. The borehole strain data (dilatometer and tensor) are marginally consistent with this model but additionally suggest the existence of a component of fault-normal collapse. The phase-1 event can be modeled by uniform slip of 3 mm along the fault. Although the transient is most probably tectonic, the post-1991 increase in precipitation suggests that hydrology may actually play a role in the tectonics. We speculate that during phase 1, precipitation acts as a surface load, which immediately increases the normal stress and thus frictional resistance across vertical faults. But as fluid diffuses to seismogenic depths it will increase fluid pore pressure in the fault zone and reduce frictional resistance. At some point, the pore-pressure effect will dominate, with a consequent increase in slip-rate. This would mark the end of phase 1 and the beginning of phase 2. The collapse component could represent the ultimate expulsion of some of this fluid from the fault zone