An evaluation of linear site-response parameters in the central and eastern United States and the importance of empirical site-response estimations

N. Seth Carpenter; Zhenming Wang; Edward W. Woolery

doi:10.1785/0120190217

An evaluation of linear site-response parameters in the central and eastern United States and the importance of empirical site-response estimations

N. Seth Carpenter, Zhenming Wang, Edward W. Woolery

Research output: Contribution to journal › Article › peer-review

7 Scopus citations

Abstract

Site response at many locations in the central and eastern United States is a resonance-based phenomenon, which site factors based on the V_S30 proxy may not reliably capture. Alternative factors, fundamental frequency, f₀, and fundamental-mode amplification, A₀, calculated from simplified expressions were evaluated against 1D linear, viscoelastic SH-wave full-resonance (FR) site responses. Tests were conducted using S-wave velocity profiles to bedrock at 11 seismic stations. The results showed that simplified expressions approximate FR f₀ and A₀ at most stations. However, at two sites with intermediate-depth strong impedance contrasts, the simplified approach underestimates A₀ by ∼ 40%. In addi-tion, FR f₀ and A₀ were compared with weak-motion earthquake S-wave horizontal-to-vertical (H/V) spectral ratios. Measuring f₀ and A₀ from S-wave H/V for our comparisons required considering multiple observations at the regional scale. The first S-wave H/V peaks occurred between ∼ 1 and ∼ 3 Hz at all seven stations in the Illinois basin (IB), which is a much lower f₀ than expected for five of these sites. Thus, we used the first S-wave H/V peaks at f > 3 Hz at these five sites in our comparisons. The S-wave H/V peaks we evaluated indicate that S-wave H/V can approximate FR f₀ . However, although the A₀ measured from S-wave H/V is positively correlated with theoretical A₀, it overestimated FR A₀ by more than 40% at all but three sites, indicating that additional study is needed to determine the appropriate use of A₀ from S-wave H/V. The observed, unmodeled amplifications between ∼ 1 and ∼ 3 Hz are of similar magnitudes as those from the shallower sediment layers at most stations, which highlights the importance of collecting empirical site-response estimations. Furthermore, the amplifications indicate the need to investigate deeper velocity structures in the IB to account for site responses within the frequency band of engineering interest.

Original language	English
Pages (from-to)	489-507
Number of pages	19
Journal	Bulletin of the Seismological Society of America
Volume	110
Issue number	2
DOIs	https://doi.org/10.1785/0120190217
State	Published - Apr 1 2020

Bibliographical note

Publisher Copyright:
© Seismological Society of America.

Funding

The authors are grateful for continued support from the Kentucky Geological Survey (KGS) and the Department of Earth and Environmental Sciences, both at the University of Kentucky, for the first author’s dissertation research support. The authors appreciate the editorial assistance of Meg Smath at KGS. The authors also thank John Hickman and Steve Greb at KGS for discussions related to Illinois basin geology. The authors are grateful for the helpful and constructive reviews they received from Bill Stephenson and an anonymous reviewer, which greatly improved this article. The authors are grateful for continued support from the Kentucky Geological Survey (KGS) and the Department of Earth and Environmental Sciences, both at the University of Kentucky, for the first author?s dissertation research support. The authors appreciate the editorial assistance of Meg Smath at KGS. The authors also thank John Hickman and Steve Greb at KGS for discussions related to Illinois basin geology. The authors are grateful for the helpful and constructive reviews they received from Bill Stephenson and an anonymous reviewer, which greatly improved this article.

Funders	Funder number
University of Kentucky Department of Earth and Environmental Sciences
Kentucky Geological Survey
U.S. Geological Survey
University of Kentucky

ASJC Scopus subject areas

Geophysics
Geochemistry and Petrology

Access to Document

10.1785/0120190217

Cite this

@article{e47ba1b0f58f4f6ebf903df48b75be54,

title = "An evaluation of linear site-response parameters in the central and eastern United States and the importance of empirical site-response estimations",

abstract = "Site response at many locations in the central and eastern United States is a resonance-based phenomenon, which site factors based on the VS30 proxy may not reliably capture. Alternative factors, fundamental frequency, f0, and fundamental-mode amplification, A0, calculated from simplified expressions were evaluated against 1D linear, viscoelastic SH-wave full-resonance (FR) site responses. Tests were conducted using S-wave velocity profiles to bedrock at 11 seismic stations. The results showed that simplified expressions approximate FR f0 and A0 at most stations. However, at two sites with intermediate-depth strong impedance contrasts, the simplified approach underestimates A0 by ∼ 40%. In addi-tion, FR f0 and A0 were compared with weak-motion earthquake S-wave horizontal-to-vertical (H/V) spectral ratios. Measuring f0 and A0 from S-wave H/V for our comparisons required considering multiple observations at the regional scale. The first S-wave H/V peaks occurred between ∼ 1 and ∼ 3 Hz at all seven stations in the Illinois basin (IB), which is a much lower f0 than expected for five of these sites. Thus, we used the first S-wave H/V peaks at f > 3 Hz at these five sites in our comparisons. The S-wave H/V peaks we evaluated indicate that S-wave H/V can approximate FR f0 . However, although the A0 measured from S-wave H/V is positively correlated with theoretical A0, it overestimated FR A0 by more than 40% at all but three sites, indicating that additional study is needed to determine the appropriate use of A0 from S-wave H/V. The observed, unmodeled amplifications between ∼ 1 and ∼ 3 Hz are of similar magnitudes as those from the shallower sediment layers at most stations, which highlights the importance of collecting empirical site-response estimations. Furthermore, the amplifications indicate the need to investigate deeper velocity structures in the IB to account for site responses within the frequency band of engineering interest.",

author = "{Seth Carpenter}, N. and Zhenming Wang and Woolery, {Edward W.}",

note = "Publisher Copyright: {\textcopyright} Seismological Society of America.",

year = "2020",

month = apr,

day = "1",

doi = "10.1785/0120190217",

language = "English",

volume = "110",

pages = "489--507",

number = "2",

}

TY - JOUR

T1 - An evaluation of linear site-response parameters in the central and eastern United States and the importance of empirical site-response estimations

AU - Seth Carpenter, N.

AU - Wang, Zhenming

AU - Woolery, Edward W.

N1 - Publisher Copyright: © Seismological Society of America.

PY - 2020/4/1

Y1 - 2020/4/1

N2 - Site response at many locations in the central and eastern United States is a resonance-based phenomenon, which site factors based on the VS30 proxy may not reliably capture. Alternative factors, fundamental frequency, f0, and fundamental-mode amplification, A0, calculated from simplified expressions were evaluated against 1D linear, viscoelastic SH-wave full-resonance (FR) site responses. Tests were conducted using S-wave velocity profiles to bedrock at 11 seismic stations. The results showed that simplified expressions approximate FR f0 and A0 at most stations. However, at two sites with intermediate-depth strong impedance contrasts, the simplified approach underestimates A0 by ∼ 40%. In addi-tion, FR f0 and A0 were compared with weak-motion earthquake S-wave horizontal-to-vertical (H/V) spectral ratios. Measuring f0 and A0 from S-wave H/V for our comparisons required considering multiple observations at the regional scale. The first S-wave H/V peaks occurred between ∼ 1 and ∼ 3 Hz at all seven stations in the Illinois basin (IB), which is a much lower f0 than expected for five of these sites. Thus, we used the first S-wave H/V peaks at f > 3 Hz at these five sites in our comparisons. The S-wave H/V peaks we evaluated indicate that S-wave H/V can approximate FR f0 . However, although the A0 measured from S-wave H/V is positively correlated with theoretical A0, it overestimated FR A0 by more than 40% at all but three sites, indicating that additional study is needed to determine the appropriate use of A0 from S-wave H/V. The observed, unmodeled amplifications between ∼ 1 and ∼ 3 Hz are of similar magnitudes as those from the shallower sediment layers at most stations, which highlights the importance of collecting empirical site-response estimations. Furthermore, the amplifications indicate the need to investigate deeper velocity structures in the IB to account for site responses within the frequency band of engineering interest.

AB - Site response at many locations in the central and eastern United States is a resonance-based phenomenon, which site factors based on the VS30 proxy may not reliably capture. Alternative factors, fundamental frequency, f0, and fundamental-mode amplification, A0, calculated from simplified expressions were evaluated against 1D linear, viscoelastic SH-wave full-resonance (FR) site responses. Tests were conducted using S-wave velocity profiles to bedrock at 11 seismic stations. The results showed that simplified expressions approximate FR f0 and A0 at most stations. However, at two sites with intermediate-depth strong impedance contrasts, the simplified approach underestimates A0 by ∼ 40%. In addi-tion, FR f0 and A0 were compared with weak-motion earthquake S-wave horizontal-to-vertical (H/V) spectral ratios. Measuring f0 and A0 from S-wave H/V for our comparisons required considering multiple observations at the regional scale. The first S-wave H/V peaks occurred between ∼ 1 and ∼ 3 Hz at all seven stations in the Illinois basin (IB), which is a much lower f0 than expected for five of these sites. Thus, we used the first S-wave H/V peaks at f > 3 Hz at these five sites in our comparisons. The S-wave H/V peaks we evaluated indicate that S-wave H/V can approximate FR f0 . However, although the A0 measured from S-wave H/V is positively correlated with theoretical A0, it overestimated FR A0 by more than 40% at all but three sites, indicating that additional study is needed to determine the appropriate use of A0 from S-wave H/V. The observed, unmodeled amplifications between ∼ 1 and ∼ 3 Hz are of similar magnitudes as those from the shallower sediment layers at most stations, which highlights the importance of collecting empirical site-response estimations. Furthermore, the amplifications indicate the need to investigate deeper velocity structures in the IB to account for site responses within the frequency band of engineering interest.

UR - http://www.scopus.com/inward/record.url?scp=85083081884&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85083081884&partnerID=8YFLogxK

U2 - 10.1785/0120190217

DO - 10.1785/0120190217

M3 - Article

AN - SCOPUS:85083081884

SN - 0037-1106

VL - 110

SP - 489

EP - 507

JO - Bulletin of the Seismological Society of America

JF - Bulletin of the Seismological Society of America

IS - 2

ER -

An evaluation of linear site-response parameters in the central and eastern United States and the importance of empirical site-response estimations

Abstract

Bibliographical note

Funding

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this