Evaluation of Six NEHRP B/C Crustal Amplification Models Proposed for Use in Western North America

Abstract

We evaluate six crustal amplification models based on National Earthquake Hazards Reduction Program (NEHRP) B/C crustal profiles proposed for use in western North America (WNA) and often used in other active crustal regions where crustal properties are unknown. One of the models is based on an interpolation of generic rock velocity profiles previously proposed for WNA and central and eastern North America (CENA), in conjunction with material densities based on an updated velocity-density relationship. A second model is based on the velocity profile used to develop amplification factors for the Next Generation Attenuation (NGA)-West2 project. A third model is based on a near-surface velocity profile developed from the NGA-West2 site database. A fourth model is based on velocity and density profiles originally proposed for use in CENA but recently used to represent crustal properties in California. We propose two alternatives to this latter model that more closely represent WNA crustal properties. We adopt a value of site attenuation (κ 0) for each model that is either recommended by the author of the model or proposed by us. Stochastic simulation is used to evaluate the Fourier amplification factors and their impact on response spectra associated with each model. Based on this evaluation, we conclude that among the available models evaluated in this study the NEHRP B/C amplification model of Boore (2016) best represents median crustal amplification in WNA, although the amplification models based on the crustal profiles of Kamai et al. (2013, 2016, unpublished manuscript, see Data and Resources) and Yenier and Atkinson (2015), the latter adjusted to WNA crustal properties, can be used to represent epistemic uncertainty.

Estimation ofκ0Implied by the High‐Frequency Shape of the NGA‐West2 Ground‐Motion Prediction Equations

“…We used the IRVT approach proposed by Al to calculate the FAS and values of the site attenuation spectral-decay parameter κ 0 (Anderson and Hough, 1984) that are implied by the high-frequency shapes of the GMPEs, when evaluated for an NEHRP B/C site condition (V S30 760 m=s), vertical strike-slip faulting, and default values of sediment depth, hypocentral depth, and depth to the top of the fault rupture plane. We used the IRVT methodology implemented in the computer program STRATA (Kottke and Rathje, 2008), together with the WNA NEHRP B/C crustal profile of Boore (2016), typical values of source velocity (β 3:5 km=s) and source density (ρ 2:72 g=cm 3 ) for WNA (Boore, 2016;Campbell and Boore, 2016), and the ground-motion duration model of Thompson (2014, 2015) to calculate the FAS of the GMPEs. The estimated κ 0 values were found to be a function of magnitude that led us to propose four tripartite piecewise linear models with the functional forms given by equation 2to model this trend.…”

“…For this purpose, we used the crustal amplification model developed by Boore (2016) for a typical NEHRP B/C crustal profile in WNA and other ACRs. Campbell and Boore (2016) determined this to be the most appropriate of five proposed WNA NEHRP B/C crustal profiles used by various investigators to represent the site profiles inherent in the GMPEs. Site amplifications given in Boore (2016) were calculated using the square-root impedance method (Boore, 2013) and do not include the attenuation effects parameterized by κ 0 .…”

Local and Moment Magnitude Analysis in the Ridgecrest Region, California: Impact on Interevent Ground-Motion Variability

“…In addition, we require that the reference stations are installed at sites with measured vs30 (average shear wave velocity of the uppermost 30m) above 700 m/s, where the vs30 are extracted from (Rekoske et al, 2020). Regarding the reference amplification function Γ(f ), we considered the crustal amplification model for the National Earthquake Hazards Program (NEHRP) B/C boundary multiplied by an exponential term with k 0 = 0.034 s (Campbell and Boore, 2016). In the following, in solving equation (2), the spectral attenuation is smoothed by requiring the second derivative with distance to be small (Castro et al, 1990).…”

New Challenges for Seismic Risk Mitigation in Urban Areas

“…5, they are composed of the product of the wave amplification term A(f) and the diminution term D(f), this latter accounting for high-frequency attenuation. Many studies in the literature have provided generic amplification curves for the term A(f), in the framework of the National Earthquake Hazard Reduction Program, NEHRP which is the seismic code adopted in the United States (e.g., Boore and Joyner, 1997;Boore, 2003;Boore 2016;Campbell and Boore, 2016). However, considering that the class thresholds for Italian and United States seismic design codes are different (see Table 4 in Pischiutta et al, 2021), in this study we choose to exploit such curves only for NTC-18 classes -B and -C (corresponding to NEHRP -C and -D, respectively).…”