Fully Nonergodic Ground Motion Models in Central and Northern California Using NGA-West2 and SCEC Cybershake Datasets

Session: Numerical Modeling of Rupture Dynamics, Earthquake Ground Motion and Seismic Noise [Poster]
Type: Poster
Date: 4/29/2020
Time: 08:00 AM
Room: Ballroom
Description: 

A key input to probabilistic seismic hazard analyses (PSHA) is the total standard deviation of the misfit between ground motion observations and the median ground motion models (GMMs), commonly known as s_tot. The most promising way to reduce hazard is to reduce s_tot through the removal of the ergodic assumption. This allows for regional and site-specific effects to be accounted for by nonergodic medians instead of in an inflated standard deviation against and incorrect median. Although strong motion networks have been rapidly growing in recent decades, the empirical data are still too sparse to establish a fully nonergodic model. In comparison, numerical simulations can generate large ground motion datasets at any desired site, which are optimal to reduce s_tot by identifying and removing repeatable effects in a fully nonergodic model. However, before such PSHA products based on simulations are accepted for engineering applications, it is crucial that they are validated against empirical data and models.

In this study, we evaluate the ability of the physics-based CyberShake platform (CS) to capture the repeatable source, site and path effects from the empirical data, by applying the Mixed Effects Regression (MER). We use CS 17.3, which was generated 285 million seismograms, distributed at 438 sites in central California and CS 18.8, which was computed for northern California with 290 million seismograms at 869 sites. We first modify the functional form to address correlation leakage among source, site and path terms in the MER. Next, we apply the MER to the CS dataset at 2, 3, 5, 7.5 and 10s periods and compare its results from MER estimates based on subsets of NGA-West2 data in the same area. Preliminary results suggest that CS results are consistent with the observed median ground motions and their variability. CS can produces large ground motion realizations not currently captured by the empirical dataset. The identification of the conditions and causes of those extreme values is an important part of our work and we will present our results to date.

Presenting Author: Xiaofeng Meng

Authors