Skip to content Skip to navigation

Selection of CyberShake Time Series for Engineering Building Code Analysis

Graduate Researcher(s): 
Ganyu Teng (
Faculty Advisor/PI: 
Jack Baker
Project Sponsor: 
Southern California Earthquake Center

This project searches the CyberShake ground motion simulation catalog in order to select suites of ground motions that satisfy building code requirements and could be used to evaluate a tall building design at several potential sites in Los Angeles. The search and selection of ground motions provides an understanding of the suitability of CyberShake ground motions for practical engineering analysis. Comparable suites of recorded ground motions are selected for the same sites, in order to illustrate the relative advantages and disadvantages of using the two data sources. 

Figure 1. Locations of two selected sites: Los Angeles downtown (LADT) and Pasadena (PAS)

Figure 2. Comparison of the target spectrum and selected response spectra from CyberShake and NGA-West2

A number of ground motion metrics, including directional polarization and shaking duration, are evaluated for the suites of simulations, in order to evaluate their reasonableness for representing future earthquake shaking. Major observations include 1) Selected ground motions from CyberShake and NGA-West2 share similar features, in terms of response spectra and polarization; 2) When selecting records from CyberShake, it is easy to specify magnitudes, distances and faults to match the deaggregation results; 3) CyberShake durations on soil are consistent with the empirical model proposed by Afshari and Stewart (2016), whereas durations on rock are shorter. This inconsistency may be due to limited recorded earthquakes. 4) Strong frequency-dependent polarization is observed for San Andreas Fault motions. Those records are usually excluded after the ground motion selection. In summary, the results provide further insights regarding the value of physics-based ground motion simulations for use in engineering analyses.