Improving the resilience of infrastructure systems is a significant challenge for engineers. When studying the performance of distributed infrastructure in earthquakes, spatial variations in strong ground motion have a significant impact. Currently, spatial ground motion variations in future earthquakes are predicted empirically, and calibrated using ground motion observations from densely recorded earthquakes. While useful, that calibration process requires strong assumptions about stationarity and isotropy of correlations. This project reports results from conducting analogous spatial variation estimation using physics-based simulations from the CyberShake platform. This platform contains simulated ground motions from tens of thousands of earthquake rupture scenarios, at locations throughout Southern California, providing a synthetic ground motion catalog that is much richer than we could ever hope to achieve from recordings. That richness allows significant relaxation of stationarity and isotropy assumptions, and provides new insights regarding the role of source and path heterogeneity on the spatial correlation of ground motion amplitudes.
We first verified CyberShake simulations by comparing their spatial correlations with an empirical model under the same assumptions. The results show that their correlation coefficients of residuals in spectral acceleration decrease with distance in roughly the same as the empirical model, but with a slightly higher overall correlation level. Additionally the results showed that CyberShake correlations are magnified in a sedimentary basin, probably due to the enhancement of common path effects. Further work is planned to examine the dependence of spatial correlations on period and rupture properties. The results indicate that advanced ground motion simulations holdpromise in allowing refinement of stationarity and isotropy assumptions in empirical ground motion correlation models.
Figure 1 Heat map of correlation coefficients with the location denoted with a triangle, for SA(3s), (left) Calculation from CyberShake simulations; (right) Empirical model