SCEC Award Number 20138 View PDF
Proposal Category Individual Proposal (Integration and Theory)
Proposal Title Multi-cycle dynamics of the Big Bend of the San Andreas fault
Investigator(s)
Name Organization
Benchun Duan Texas A&M University
Other Participants Dunyu Liu (Postdoc)
SCEC Priorities 1d, 2e, 5a SCEC Groups FARM, SAFS, Geology
Report Due Date 03/15/2021 Date Report Submitted 03/10/2021
Project Abstract
 This project has two objectives. First, we explore mechanical conditions that produce similar levels of slip rates on the two sides of the Big Bend along the San Andreas Fault (SAF), which is a feature in geological observations that seems inconsistent with the restraining Big Bend. Second, we explore rupture patterns along the fault system comprising the southern SAF and the San Jacino Fault (SJF) that are consistent with paleoseismological observations in recurrent intervals and their COVs. We use a 2D multicycle dynamic modeling method to achieve these objectives. The method consists of a finite element model for coseismic spontaneous ruptures and an analytical Maxwell viscoelastic model for the interseismic fault stress loading and relaxation. In this method, the fault shear and normal stresses evolve spontaneously over multiple earthquake cycles, providing an initial stress condition for dynamic rupture models that is consistent with complex fault geometry and fault rupture history. We find that heterogeneous loading along the fault strike, supported by geodetic observations, can produce the feature in slip rate distribution. By fitting with paleoseismological observations, we constrain other model parameters to explore rupture patterns on the fault system. We observe the 1857, 1812, 1918 types of ruptures and also ruptures confined to the north of the Big Bend in the models. These rupture patterns are favorably comparable to rupture scenarios reported in paleosesimological studies. Observations-constrained multicycle dynamic models provide physical insights into rupture behaviors of the fault system, contributing to seismic hazard assessment in Southern California.
Intellectual Merit The research integrates numerical modeling with geological and paleoseismological observations 1) to improve physical understanding of observed features along the San Andreas fault system, and 2) to provide physics-based guidance for seismic hazard analyses in Southern California. Both align well with SCEC research objectives. Integration nature of the research also contributes to an effort of interdisciplinary research and collaboration SCEC has been fostering. To better integrate numerical models with geological, paleoseismological, and seismological observations, the models need to handle complexities in geologic structure, including complex fault geometry. Our multicycle dynamic models target for geometrically complex faults, which are important in controlling rupture extent and path (thus earthquake sizes and scenarios). Integrating several complex and challenging aspects of earthquake studies in one project (such as this one), including matching geological and palesoseismological observations, handling complex fault geometry, and simulating spontaneously dynamic ruptures within the context of earthquake cycles, is creative and should be a future direction in earthquake science research.
Broader Impacts The project trains a postdoctoral researcher in fundamental research in earthquake science and its application in seismic hazard analysis, including integration of physics-based models with geological and paleosemological observations. The research results have been used in undergraduate geophysics course teaching by the PI, contributing to the teaching and learning of undergraduate students at Texas A&M University. The project also facilitates collaborations between modelers and geologists, contributing to interdisciplinary team formation. The research contributes to seismic hazard assessment and thus hazard mitigation in Southern California.
Exemplary Figure Figure 3. Five rupture patterns from Model A in this study. In each sub-figure (a-f), slip distribution (top panel) and rupture time (bottom or middle panel) are given. Rupture time curves seem noisy, but reflect dynamic triggering of some fault portions before the main rupture reaches them: relatively continuous portion of curves clearly shows rupture propagation, with zero rupture time location corresponding to nucleation of the main rupture. Bottom panel in (d-f) gives the fault geometry for reference. (Liu et al., 2021, in preparation).