Characterization of spatial correlations in ground motions—insights from physics-based simulations

Jack W. Baker, & Yilin Chen

Submitted August 7, 2017, SCEC Contribution #7425, 2017 SCEC Annual Meeting Talk on Mon 11:30

This talk presents statistical methods to quantify spatial correlations in the intensity of ground shaking across a region in a given earthquake, using observed and simulated ground motions. Risk assessment of spatially distributed building portfolios or infrastructure systems requires such quantification in order to predict the spatial extent of damage and impacts. This talk will first review techniques to perform such analysis using observed ground motions from past earthquakes, and describe how these results can be used in probabilistic seismic risk assessment of spatially-distributed lifelines. Quantification using past earthquakes is challenging due to limited observational data, and requires assumptions regarding stationarity of correlations across space and from earthquake to earthquake. As an alternative, physics-based simulations can be used to estimate correlations, and can overcome the constraints from limited data faced when performing empirical estimation. Results from CyberShake simulations are presented, and used to illustrate apparent nonstationarity present in the simulated ground motions from location to location and rupture to rupture. While measurement of these correlations is not necessary in some sense, because they are already implicitly present in the simulations, the results do serve two useful purposes. First, the results serve as a validation metric that can be used to compare simulations to observed ground motions. Second, the results provide insights into the role of source and path properties on resulting spatial correlations, and these insights are useful in regions of the world where comparable simulated ground motions are not available.

Key Words
seismic hazard, ground motion simulations, CyberShake

Baker, J. W., & Chen, Y. (2017, 08). Characterization of spatial correlations in ground motions—insights from physics-based simulations. Oral Presentation at 2017 SCEC Annual Meeting.

Related Projects & Working Groups
Earthquake Engineering Implementation Interface (EEII)