SCEC Award Number 13143 View PDF
Proposal Category Individual Proposal (Integration and Theory)
Proposal Title GPS-Constrained Stressing Rates and Earthquake Cycle Models: Contributions to the SCEC Community Stress Model
Investigator(s)
Name Organization
Elizabeth Hearn Consulting Geophysicist
Other Participants
SCEC Priorities 2, 1, 1 SCEC Groups SDOT, Geodesy
Report Due Date 03/15/2014 Date Report Submitted N/A
Project Abstract
 I proposed two reasearch activities for 2013: incorporating stressing rates from UCERF3 deformation models into the SCEC Community Stress Model (CSM), and assessing the importance of earthquake cycle-related perturbations to the GPS surface velocity field in southern California. Stressing rates from the three UCERF3 models, were added to the CSM and presented and discussed at the May 2013 CSM workshop. In general, differential stresses and sH1max orientations agree reasonably well (especially along the SAF). The results differ in detail because of different modeling techniques (e.g. block- versus non-block models; discretization differences). The finite-element earthquake-cycle models suggest that viscoelastic perturbations to the GPS velocity field may matter solely for segments producing the largest (M 7.8) earthquakes. This means that models of stress transfer between faults may not have to represent earthquake cycles on all southern California faults, which means that developing such models may be less computationally expensive than previously thought.
Intellectual Merit My earthquake-cycle modeling (and recent studies by K. Johnson, T. Wright, B. Meade and others) suggests that interseismic viscoelastic perturbations to surface deformation may be significant solely for large earthquakes, and that deformation models of southern California need not include viscoelastic earthquake cycle effects for most of its active faults. That is, the computationally expensive task of “spinning up” the model over thousands of timesteps (for each active fault segment) may be unnecessary. This finding contributes to moving our field forward in that it opens the door for the development of detailed yet computationally tractable, three-dimensional deformation models of the southern California lithosphere, calibrated to both geodetic and geologic (slip rate) data. It also suggests that we should shift our focus to modeling upper crustal processes or features that may have more influence on stress transfer bewteen faults than viscoelastic relaxation of the lower crust and upper mantle (e.g. effects of fault geometry and connection, inelastic deformation, shear zone creep and poroelastic rebound).

The suite of finite-fault earthquake cycle models incorporating viscous shear zones addresses a longstanding conundrum: how can we explain rapidly decaying, large amplitude postseismic transients over a broad region with highly localized and almost stationary interseismic deformation seen around the world’s strike-slip faults? A low-viscosity shear zone embedded in a strong lithosphere, together with low-viscosity mantle asthenosphere, can explain the geodetic observations, as long as the shear zone’s effective viscosity increases with depth and/or time. These conditions are likely met in the Earth, given transient and power-law rheologies, the transition from quartz and feldspar to olivine rheologies the Moho, and the near-ubiquitous presence of localized viscous shear zones in association with major faults. This finding contributes to advancing knowledge and understanding in our general field, and it argues that viscous shear zones should be represented in any regional-scale stress transfer model of southern California (at least, below major faults).
Broader Impacts The UCERF3 model stressing rate contributions to the SCEC CSM add to a resource that is of obvious importance to understanding deformation dynamics and seismic hazard in southern California. Our ductile deformation workshop (and the CSM/CGM workshops) brought together diverse groups of colleagues working on understanding stresses in the southern CA lithosphere, and helped to distill a diffuse range of knowledge and opinions into a statement of what most of us agree to be true, and what the outstanding problems are. My earthquake cycle modeling work addresses how we estimate fault slip rates from GPS velocity fields, which feeds directly into earthquake probability calculations and the societally important task of assessing seismic hazard in southern California and other tectonically active regions.
Exemplary Figure Figure 1. Maximum principal stressing rate axis orientations (a-d) and differential stress per unit time (e-h) at a depth of 5 km from the three UCERF3 deformation models and (for reference) the block model of Loveless and Meade (2011). sH1max orientations range from -10° (N10W) to 35° (N35E), and differential stressing rates range from 0 to 50 kPa per year. a and e: Zeng buried dislocation model; b and f: average block model (K. Johnson); c and g: NeoKinema model (P. Bird); d and h: Loveless and Meade (2011) model. All of these results (as well as metadata and references) are available via the SCEC CSM website.

(note: I can provide an Illustrator file if need be.)