← Back to Sessions

Are Empirical Models Adequate for Risk Estimation?

Description: 

Seismic hazard poses a significant risk in California, particularly in regions with complex fault systems and dense populations. Physics-based simulations developed by the Statewide California Earthquake Center (SCEC), such as the CyberShake project, enhance our understanding of ground motion impact by capturing critical phenomena like rupture directivity, waveguide effects, and 3D basin amplification—factors usually not included in empirical models. In addition, data at distances of less than 10-20 km from the faults, as well as for large magnitude events, are sparse, severely limiting the accuracy of Ground Motion Models (GMMs), usually used to obtain ground motion estimates for risk estimation. Preliminary results from CyberShake Study 22.12 in Southern California demonstrate the unique insights provided by physics-based models, highlighting key differences in ground motions compared to empirical models. For example, comparisons for a M6.85 Puente Hills rupture scenario suggest that leading NGA-W2 GMMs on average may overpredict the spectral accelerations at a period of 0.3 s and peak ground accelerations by up to 30% and thus lead to overestimates of building damage within a distance of 20 km from the fault. In addition, the physics-based simulations have the potential to reduce the uncertainty associated with the model space, in particular at longer periods considering source- and site-specific effects, as compared to ergodic GMMs at critical near-fault sites, allowing for more targeted policy making. Our analysis will be extended to additional scenarios and rupture variations for earthquake sources affecting downtown Los Angeles.

Session: Accuracy and Variability of Physics-based Ground Motion Modeling - I

Type: Oral

Date: 4/15/2025

Presentation Time: 03:00 PM (local time)

Presenting Author: Brierton

Student Presenter: Yes

Invited Presentation: 

Poster Number:

Authors