Ground-Motion Site Response and New Physics-Based Site Correction Factors for Design Response Spectrum

Session: Near-Surface Effects: Advances in Site Response Estimation and Its Applications
Type: Oral
Date: 4/28/2020
Time: 10:45 AM
Room: 110 + 140
Description: 

Site response can be physically quantified by two parameters: the fundamental (i.e., base mode) site period, T_f and its peak amplification, A₀. The fundamental site period and its peak amplification can be estimated by empirical and theoretical methods, including (1) the soil-to-rock spectral ratio, (2) the horizontal-to-vertical spectral ratios (HVSR) of earthquake S-waves, (3) 1-D site response analysis and (4) simplified one-layer sediment over bedrock. The fundamental site period and its peak amplification also depend on the level of input ground-motion due to the nonlinear behavior of soil. The effect of nonlinearity can also be estimated by empirical and theoretical methods.

Currently, the time-weighted average shear-wave velocity of the top 30 m of the soil/rock column, Vs30, is used to estimate site response in earthquake engineering. However, Vs30 may not be an appropriate parameter because it does not physically and uniquely correlate to site response, site resonance in particular. We propose directly utilizing the fundamental site period and its peak amplification to develop site correction factors for engineering applications. Specifically, we propose to substitute T_f for T_S and the square root of peak amplification (A₀^-2) for the site coefficient (F_a) for determining design spectral ground motion parameters, per the ASCE 7 methodology. The resulting design response spectrum is:

T_S=T_f; T₀ = 0.1T_s; S_T0 is the response acceleration on bedrock at T₀ or 0.2s if T₀≤0.2s; S_DS= A₀^-2S_T0; S_a = S_DS (0.4+0.6T/T₀) for T ≤ T₀; S_a = S_DS for T₀ < T ≤ T_S; S_a = S_DST_S/T for T_S ≤ T.

We applied this new physics-based method to develop design response spectra at six sites in the central United States. Our results demonstrate that the design response spectra developed using the new method can characterize site response better than the current Vs30-based approach.

Presenting Author: Zhenming Wang

---

Authors

Zhenming Wang Presenting Author Corresponding Author zmwang@uky.edu University of Kentucky, Lexington, Kentucky, United States Presenting Author Corresponding Author

Seth Carpenter seth.carpenter@uky.edu University of Kentucky, Lexington, Kentucky, United States

Edward W Woolery ewoolery@uky.edu University of Kentucky, Lexington, Kentucky, United States

Michael E Kalinski michael.kalinski@uky.edu University of Kentucky, Lexington, Kentucky, United States