SCEC Award Number 18168 View PDF
Proposal Category Individual Proposal (Integration and Theory)
Proposal Title Implementation and Validation of Multi-surface Plasticity in a Discontinuous Mesh Finite Difference Code
Investigator(s)
Name Organization
Daniel Roten San Diego State University Kim Olsen San Diego State University Yifeng Cui University of California, San Diego
Other Participants
SCEC Priorities 4a, 4b, 4c SCEC Groups GM, CS, Seismology
Report Due Date 03/15/2019 Date Report Submitted 06/06/2019
Project Abstract
 SCEC project 18168 aims to implement an Iwan-type plasticity model in the discontinuous mesh version of the AWP-GPU finite difference (FD) code. A further scope of the project is to investigate the impact of Iwan-type plasticity on realistic earthquake ground motion scenarios, in particular a M7.8 (ShakeOut) earthquake scenario on the southern San Andreas fault. As of this writing, the implementation of multi-surface plasticity in AWP-GPU has not yet been fully completed, and verification of the method against the CPU code of AWP and other 1D or 2D wave propagation codes is still pending. This delay was mainly caused by migration from the Titan supercomputer to the newer system Summit (both at OLCF) as well as the decommission of Blue Waters (NCSA). Porting of the code to the newer Volta GPUs revealed several preexisting bugs that needed to be addressed before development could continue. In this preliminary report, we summarize the current state of code porting and development. We als present new results of ShakeOut scenario simulations using the Iwan model performed with the CPU version of AWP, and the sensitivity of ground motions to the choice of rock strength parameters.
Intellectual Merit This project directly addresses the question of how strong ground motions depend on the complexities and nonlinearities of earthquake systems, one of the five basic questions outlined in
SCEC5. Implementation of realistic soil nonlinearity into the efficient, discontinuous mesh finite difference code provides the community with a powerful tool to model the dynamics of the source, wave propagation and site response in a single step. The performed research aligns with SCEC5's goal to depart from an isolated treatment of source, site and path effects, and to consider surface ground motions as the nonlinear response of soils and rocks from source to surface.
Broader Impacts The scenario of a large earthquake on the southern San Andreas fault dominates the hazard for high-rise buildings in the Los Angeles region. Our result show that more accurate predictions of long-period ground motions during such event can be simulated with advanced plasticity models. Ongoing research aims to include such nonlinear effects in the Cybershake simulations, which will result in more accurate physics-based seismic hazard maps.
Exemplary Figure Figure 5: 3s-SAs inside rectangular area include main waveguide amplification patch (Fig. 4) obtained in the linear case, the nonlinear case using a single von Mises yield surface, and a multi-surface Iwan model using different definitions of the reference strain.