SCEC Award Number 13083 View PDF
Proposal Category Travel Only Proposal (SCEC Annual Meeting)
Proposal Title Developing a statistical framework for earthquake rupture process for physics-based ground motion simulation
Investigator(s)
Name Organization
Seok Goo Song Eidgenössische Technische Hochschule Zürich (Switzerland) Luis Angel Dalguer Eidgenössische Technische Hochschule Zürich (Switzerland)
Other Participants Youbing Zhang (PhD student) and 1 master student
SCEC Priorities 6b, 6e, 4d SCEC Groups GMP, CME, SIV
Report Due Date 03/15/2014 Date Report Submitted N/A
Project Abstract
This project was funded in 2009, 2010, 2012, and 2013 by SCEC. The main research objective is to develop a statistical framework for quantifying the variability (aleatory uncertainty) of finite source process and to understand its relationship to near-source ground motion characteristics. We developed a stochastic model that governs the finite source process with 1-point and 2-point statistics of kinematic source parameters and developed a pseudo-dynamic rupture model generator (SongRMG, Ver 1.0). Especially in the project year, we investigated the effect of 1-point and 2-point statistics of kinematic source parameters on near-source ground motions by systematically perturbing 1-point and 2-point statistics of input source parameters (Song et al. 2014)
Intellectual Merit Rupture dynamics enables us to understand earthquake rupture process in a physics-based way. We can study many interesting and complicated features of earthquake rupture by dynamic modeling. On the other hand, earthquake statistics enables us to quantify the variability of earthquake rupture for future events. We aim to develop a stochastic model for finite source process with simple correlation structures. This is an exciting research work because we can simulate finite source models by stochastic modeling.
Broader Impacts Seismologists and earthquake engineers can use stochastic finite source modeling tools developed in this project and generate a number of rupture scenarios for simulating ground motions. They can also study the effect of finite source process on near-source ground motion characteristics in a systematic sense.
Exemplary Figure Figure 4. Ground motion residuals from pseudo-dynamic models (compared to dynamic models) for different levels of source correlations. Simulations show that source correlations produce stronger ground motions, and appropriate correlation in source modeling leads to less randomness in ground motion predictions (Song et al. 2014, GJI)