S060-0016
Construction of a Database of Validated Simulated Ground Motions for the Evaluation of the Seismic Risk Posed to Distributed Infrastructure Systems: The San Francisco Bay Area Case Study

Wednesday, 16 December 2020
Poster
Floriana Petrone1, Norman Abrahamson2, David McCallen1, Arben Pitarka3 and Arthur J Rodgers4, (1)University of Nevada, Reno, Civil and Environmental Engineering, Reno, NV, United States, (2)University of California Berkeley, Civil and Environmental Engineering, Berkeley, CA, United States, (3)Lawrence Livermore National Laboratory, Livermore, CA, United States, (4)Lawrence Livermore National Laboratory, Atmospheric, Earth and Energy Division, Livermore, CA, United States
Abstract:
Current approaches to earthquake structural risk rely on the use of Uniform Hazard Spectra (UHS) for the selection and scaling of a suite of real records to perform nonlinear dynamic analyses (NLDA) of structural systems. However, for several seismically active regions where the hazard-controlled earthquake scenarios are typically large magnitude events (Mw>6.5) with short source-to-site distances (<15km), the database of real records is severely limited. Furthermore, having spatially sparse records does not allow to fully account for the potentially large site-specific variability of the demand, the interaction between complex incident seismic waves and the sensitivity of the structural response to specific ground-motion characteristics that are not explicitly represented by UHS (e.g., frequency content, duration, multi-directionality, and pulse-like component). Ground motions simulated with high-resolution 3D Earth models supplement the database of real records and enable site-specific evaluations of the earthquake risk to distributed infrastructure. This study was conducted within a project supported by the Pacific Earthquake Engineering Research (PEER) Center to construct a database of validated simulated ground motions, computed for eight Mw7 Hayward Fault rupture earthquake scenarios in the San Francisco Bay Area. The simulations with 5Hz resolution are performed with the SW4 computer program, and the kinematic rupture scenarios were generated with the Graves and Pitarka hybrid rupture modeling method. Results represent a set of records with aleatory variability derived from the characteristics of the fault rupture only, with the objective of evaluating the variability of the demand posed to the infrastructure residing in the region. The simulations are validated through comparison against Ground Motion Models (GMMs) and real records, with a specific focus on near-fault areas and sites along the West San Francisco-Oakland Bay Bridge.