Estimating variations in locking depth for the Mojave segment of the San Andreas fault over the past 1500 years from paleoseismic stress drop
Teira Solis, & Bridget R. Smith-KonterIn Preparation 2015, SCEC Contribution #6116
The recurrence and slip of earthquakes along active fault systems such as the San Andreas Fault System (SAFS), to first order, should depend largely on the stress that accumulates between slip events and the release mechanism, or stress drop, of a fault segment. This model, however, often fails to explain the large differences in earthquake recurrence intervals and slip for published paleoseismic datasets. To better understand how stress accumulation can vary over multiple earthquake cycles, we investigate the role of paleoseismic fault locking depth of the Mojave segment of the SAFS. We use published date and slip estimates spanning the last 1500 years [Weldon et al., 2004] and a 4-D earthquake stress model [Smith-Konter and Sandwell, 2009] to tune model-derived stress drops for each earthquake cycle to fit stress drop estimates derived from paleoseismic data. For each earthquake cycle, we assume a constant slip rate and systematically adjust the locking depth of the Mojave segment to best match the paleoseismic data. For 9 of the 14 events studied, we find that model-derived stress drops can accurately predict paleoseismic stress drop, given the range of tested fault depths (5-25 km), when slip uncertainties are accounted for. We also identify a positive correlation between event locking depth and stress drop, where low stress drop events are best modeled with a larger fault depth and higher stress drop events are best modeled with a moderate fault depth. In terms of the paleoseismic data, this relationship suggests that when stress is stored by fault that is more deeply locked, stress will accumulate at slower rate throughout an earthquake cycle, producing an overall lower stress drop than would be expected for a shallow fault. From these results, we propose that realistic variations in locking depth spanning 9 of the 14 earthquake cycles of the Mojave segment could have generated enough variation in earthquake cycle stress rate to support non-characteristic stress drop behavior.
Citation
Solis, T., & Smith-Konter, B. R. (2015). Estimating variations in locking depth for the Mojave segment of the San Andreas fault over the past 1500 years from paleoseismic stress drop. Journal of Geophysical Research, (in preparation).