Society for Earthquake and Civil Engineering Dynamics
We summarize the results of an ongoing study within the Southern California Earthquake Center (SCEC) Technical Activity Group (TAG) focused on Ground Motion Simulation Validation (GMSV). The aim here is to address, on a statistical basis, whether simulated ground motions for four historical earthquakes are biased in terms of median spectral-shape- and duration-related intensity measures (IMs) in comparison with real records. In particular, the study considers the Graves and Pitarka’s (2010) hybrid broadband ground motion simulation methodology for the following historical events: 1979 M 6.5 Imperial Valley, 1989 M 6.8 Loma Prieta, 1992 M 7.2 Landers, and 1994 M 6.7 Northridge. We also compare the intra-event dispersion of recorded and simulated IMs. Hypothesis tests are carried out to assess the statistical significance of the results found for each IM and each event. The results from this study show that the simulation methodology matches well the recorded IMs and the observed differences are generally not statistically significant, providing confidence in using the simulation methodology for engineering application.