Project Title/ID Number Physically Based Source Input for Strong Ground Motion Simulation—Lifelines 1C08
Start/End Dates 10/1/02—9/30/03
Project Leader Greg Beroza (Stanford/Faculty)
Team Members Mariagiovanna Guatteri (Stanford/Post Doc), Seok Goo Song (Stanford/Grad Student)
Project goals and objectives

The goal of this project is to develop a library of rupture models for scenario earthquakes that are consistent with both the scaling and spatial variability of slip determined for past earthquakes and with simple notions of earthquake faulting physics.

Role of this project in supporting PEER’s vision
This project is designed to play an important role in validation exercises for the PEER/Lifelines Group. In particular, the use of a common source model, with consistently defined and realistic source properties, is designed to allow the different ground motion modeling groups in the PEER/Lifelines program, to compare results obtained by their different methods.
Methodology employed

We have used a spatial random field model developed by Mai and Beroza [2002] to create realizations of scenario earthquakes. From these we create dynamic models that are consistent with them using the method of Guatteri et al. [2003] based on a modified boundary integral formulation. These inputs are used as a basis for predicting an approximate, physically consistent source model (viz., the pseudo-dynamic approximation).

Brief description of past year’s accomplishments and more detail on expected Year 6 accomplishments

We have carried out dynamic rupture modeling using the BIM algorithm developed by Quin and Das [1989]. The method is strictly valid only for an elastic whole-space. That is, it does not account for the fact that the Earth has a free surface, nor does it account for the fact that the Earth’s material properties change with depth, which is usually represented with horizontal layering. The pseudo-dynamic source model we are developing under PEER sponsorship is based on the output of this dynamic rupture-modeling algorithm. Because we expect the free surface and horizontal layering to exert an important influence on the dynamics of earthquake rupture propagation, it is important to take them into account. We have demonstrated that we can accurately reproduce the effects of the free surface and horizontal layering in dynamic rupture modeling within the framework of the boundary integral method (BIM) using two approximations that are valid for vertical faults. This was presented at the December quarterly PEER/Lifelines workshop. We have also developed a library of source models for scenario strike-slip earthquakes. This was presented at the March quarterly PEER/Lifelines Workshop.


Figure 1. Example representing the pseudo-dynamic procedure.

The starting point is a slip realization generated as a spatial random field [Mai and Beroza, 2002], which yields a corresponding stress drop distribution computed using the method of Andrews [1981] shown in the upper left panel. Upper right panel shows a fracture energy distribution determined in such a way the the average rupture velocity is subshear. The crack resistance (middle left) determines a direct mapping to an approximate local rupture velocity (middle right). Rupture time (lower left) is determined from this and rise time is estimated based on the overall dimensions of the slipped region.
Larger View

Other similar work being conducted within and outside PEER and how this project differs

We had funding from SCEC to work on the initial development of this model. We also have funding from NSF-Engineering (PI: Cornell) to study the effects of severe ground motions in the extreme near field of earthquakes. This work differs in that we have developed a library of improved pseudo-dynamic rupture models for use by the PEER lifelines programs, we are in the process of transforming this into formats that will be readily useable by the different modeling groups, and we will be distributing the algorithm we have used to generate the pseudo-dynamic models.

Plans for Year 7 if this project is expected to be continued
We plan to extend the model to include the effects of the free surface for non-vertical faults. We plan to develop a parallel library and modeling capability for dipping faults. We also plan to explore the effects interaction of the free surface with a dynamically propagating rupture on the generation of strong ground motion. We also hope to improve our approximation to more realistically predict the time dependence of slip after fault rupture.
Describe any instances where you are aware that your results have been used in industry


Expected milestones

Library of Source models delivered to modeling groups. Program and documentation for pseudo-dynamic modeling made available on-line.


Source model library, pseudo-dynamic modeling program.