The Influence of Impedance-Ratio Distributions on 1D Linear Site Response Proxies
Description:
Ground-motion site response is inherently site specific since it depends on in situ dynamic parameters, particularly shear-wave velocity, Vs. Thus, identifying the best proxies to predict site-response remains a challenge. Using Vs profiles developed at 193 central and eastern U.S. sites with underlying lithologies ranging from bedrock to deep (>100 m) soils, we investigated the influence of impedance ratio, IR, distributions on 1D linear site responses at the first peak frequencies, f0. We categorized each site as “gradient”, “single-layer”, or “multilayered”, depending on the number of Vs-profile interfaces with IR≥3.0, where gradient sites lack any such interface. We then evaluated correlations between various Vs-profile attributes with f0 and corresponding amplifications, A0, both calculated from 1D linear site-response analyses. Depth-based attributes, Z1.0, Z2.5, and ZIRmax, the depths to velocities of 1.0 km/s, 2.5 km/s, and the maximum IR, IRmax, respectively, correlate well with f0 at single- and multilayered sites, whereas Vs30, the average Vs in the upper 30 m, is the only strong predictor of f0 at gradient sites. Vs30 also correlates with f0 at other site types but is unreliable at low frequencies (≤~3 Hz) and slow Vs (≤~300 m/s). IRmax correlates with A0 for all site categories with the strongest correlation at single-layer sites and the weakest at gradient sites. Average Vs to Z1.0 correlates marginally (r<0.6) with A0 at single-layer sites as does Vs30 at gradient sites. These results suggest: 1) the most appropriate site response proxies depend on the distribution of strong IRs beneath a site, 2) no single proxy has skill to predict both f0 and A0, 3) IRmax and its depth are the most versatile and useful attribute pair, and 4) Vs-profile characteristics may contribute to variability in ground-motion model residuals.
Session: Understanding and Modeling the Uncertainties in Earthquake Ground Motions
Type: Oral
Date: 4/20/2023
Presentation Time: 10:15 AM (local time)
Presenting Author: Seth Carpenter
Student Presenter: No
Invited Presentation:
Authors
Seth Carpenter Presenting Author Corresponding Author seth.carpenter@uky.edu University of Kentucky |
Zhenming Wang zhenming.wang@uky.edu University of Kentucky |
|
|
|
|
|
|
|
The Influence of Impedance-Ratio Distributions on 1D Linear Site Response Proxies
Category
Understanding and Modeling the Uncertainties in Earthquake Ground Motions