Decomposing Source and Path Terms in High-Frequency Ground Motion Data
Date: 4/24/2019
Time: 04:15 PM
Room: Pine
Earthquake source effects are notoriously difficult to separate from those of path and site attenuation, both for earthquake source seismologists and researchers in the ground-motion engineering community. Yet correctly modeling the source properties can aid in accurately understanding attenuation trends and patterns. The positive relationship between earthquake stress drop and ground-motion event terms (i.e., residuals between ground-motion prediction equations and high frequency data) is taken as a certainty, given that events with larger Brune stress drops generate more high-frequency energy and thus more ground motion. However, this connection is contingent on those ground-motion event terms truly representing the earthquake source, thus being unbiased with distance or attenuation. We consider several cases in which event and attenuation effects are difficult to resolve. Specifically, we study high-frequency ground-motion data, PGA and PGV which should have a clear relationship to stress drop to demonstrate that seismological understanding can aid our interpretation. Furthermore, assuming anelastic attenuation has a small effect at close distances, we only consider close-in records with Rrup<20km to avoid contamination from path attenuation; we then model the remaining path attenuation, i.e., Q, and site effect, such as kappa. For example, in the case of the 2014 M6.0 South Napa, California earthquake, the high-frequency ground-motion residuals show a strong dependence on distance, strongly biasing the event term and incorrectly mapping a specific regional attenuation effect into the source term. This is one situation when the event term from high-frequency ground-motion prediction equation residuals is clearly not equal to the earthquake stress drop and suggests how both event terms and regional attenuation should be estimated to be more indicative of physical behavior. Ground-motion models that correctly reflect the source and attenuation properties can thus be correctly extrapolated into different magnitude or distance ranges, or new geological regions.
Presenting Author: Annemarie S. Baltay
Authors
Annemarie S Baltay abaltay@usgs.gov U.S. Geological Survey, Menlo Park, California, United States Presenting Author
Corresponding Author
|
Thomas C Hanks thanks@usgs.gov U.S. Geological Survey, Earthquake Science Center, Menlo Park, California, United States |
Decomposing Source and Path Terms in High-Frequency Ground Motion Data
Category
Modeling and Understanding of High-frequency Ground Motion