Estimating evolving shear tractions on the southern San Andreas and Wasatch fault systems
Grasshopper Anderson-Merritt
University of Massachusetts Amherst
- Date & Time
- Location
- Online-only seminar via Microsoft Teams
- Host
- Chad
- Summary
Estimates of evolving shear stress can inform the interpretation of earthquake recurrence and open intervals, inform the stress conditions that generate ground-rupturing earthquakes, and shed light on the relationship between accumulated stress, earthquake rupture length, and fault system geometry. Estimates of stress conditions on the eve of past earthquakes can also provide critical inputs for dynamic rupture models, which are sensitive to their initial traction conditions. In fault systems where rich paleoseismic records are available, numerical models can provide estimates of past fault stress accumulation and release over hundreds to thousands of years. I simulate evolving shear tractions in the southern San Andreas fault system since 1 ka and the Wasatch fault system since 5 ka, using Monte Carlo simulations to assess uncertainties in traction estimates due to uncertainties in earthquake timing, upper crustal rheology, and earthquake stress drop. In the southern San Andreas fault system, pre-earthquake traction estimates suggest that earthquakes with long rupture length require high shear tractions over a large area to occur, but high tractions may not be sufficient to allow ruptures to propagate over long distances through geometric complexities. This spatial pattern arises regardless of varying earthquake timing, upper crustal rheology, or the completeness of the stress drop. Uncertainty in upper crustal rheology has the greatest impact both on overall shear traction estimates and spatial variability in shear traction. In the Wasatch fault system, uncertainty in upper crustal rheology has the biggest impact on shear traction estimates, but near segment boundaries, the relative timing of earthquakes also makes a significant contribution to shear traction uncertainty. The segmented rupture model used for these simulations results in zero slip and high accumulated shear tractions near segment boundaries, suggesting that a rupture model that permits spillover or multisegment ruptures may better reflect the behavior of this fault system.