7^th IASPEI / IAEE International Symposium · Effects of Surface Geology on Seismic Motion

Grenoble, France · October 19–21, 2026

Integration of Non-Ergodic Site Response into NEHRP Provisions & ASCE-7

Jonathan P. Stewart¹, David P. Teague², C.B. Crouse³, Gyimah Kasali⁴, Silvia Mazzoni⁵, Debra Murphy⁶, Shahram Pezeshk⁷, Melanie A. Walling⁸, Tessa Williams⁹, and Zia Zafir¹⁰

Author affiliations

¹Sabol-Scott Professor, UCLA Samueli Engineering, Los Angeles, CA
²Principal, Engeo, San Ramon, CA
³Principal Engineer, AECOM, Seattle, WA
⁴Principal Geotechnical Engineer, Rutherford + Chekene, San Francisco, CA
⁵Consultant, Santa Monica, CA
⁶Principal Engineer, Slate Geotechnical Consultants, Oakland, CA
⁷Professor and Chair, University of Memphis, Memphis, TN
⁸Principal, MAW Seismic, LLC, Seattle, WA
⁹Senior Engineer, Slate Geotechnical Consultants, Oakland, CA
¹⁰Principal Engineer, Kleinfelder, Rancho Cordova, CA

Section 1

NEHRP Provisions

Chapter 21: Site-Specific Ground Motion Procedures for Seismic Design

Chapter 21 serves as an alternative to default design motions (Chapter 11).
Section 21.1 pertains to site-specific site response, which can be applied in two contexts (Figure 1):
1. Option 1 (Figure 1a): Site-specific PSHA applied with ergodic site factors (in which case Section 21.1 is not used) or site-specific site response (Section 21.1).
2. Option 2 (Figure 1b): PSHA for reference site from NSHM (Petersen et al. 2024) for reference site with site-specific site response (Section 21.1).
Section 21.1 can be applied using Site-Specific Site Response or Hybrid Ground Motions approaches.

Section 2

Current Procedures (NEHRP 2020 and Before)

Requirements

Base motion
Site condition model
Ground response analyses (GRA)
Applied with hybrid approach

Figure 2. Schematic illustration of adjustments required to ensure consistency between halfspace velocity and reference site condition in calculation of surface intensity measure. The multiplication of base IMs by AF to estimate surface IMs represents the use of a hybrid approach. Source: Kramer & Stewart (2025)

Issues Motivating New Procedures

Period limitation of GRA
Hazard level not preserved
Lack of guidance on non-ergodic site response

Section 3

Site-Specific Site Response

Required model elements

Mean model, μ_lnAF
Aleatory variability
Epistemic uncertainty
f₂ represents the slope of the log amplification with the log of x_IMr
f₃ (in the same units as x_IMr) represents a transitional level of x_IMr between linear and strongly nonlinear response.

Can be derived from

Approach 1: Ground motion data
Approach 2: Ground response analyses
Combinations of 1 & 2

Equation mu lnAF = f1 + f2 ln((x_IMr + f3)/f3), with the nonlinear term F_nl plotted against x_IMr (PGA) from 0.01 to 1 g, decreasing from 0 toward -0.7; f3 marks the transition and f2 is the slope at large x_IMr. — Figure 3. Mean site response model and physical representation of coefficients f₂ and f₃. *Source: Kramer & Stewart (2025)*

Hazard curves of annual frequency of exceedance versus PSA(1 s) for a reference site with VS30 = 760 m/s. The single-station curve (blue) and reduced-sigma curves fall below the full aleatory curve at low exceedance frequencies. — Figure 4. Hazard curve for reference site (V_S30 = 760 m/s) with full aleatory variability (σ), aleatory variability reduced by 0.1 and 0.2, and single-station sigma (blue).

Allowable reductions

Point of comparison is MCE_R spectrum from USGS seismic design Geo-Database
NEHRP 2020 (ASCE-7/22) and before: maximum 20% reduction
NEHRP 2026, site-specific approach: maximum 30% reduction

Section 4

Hybrid Ground Motions

Hybrid Ground Motions Approach

Amplification computed using GRA with MCE_R and ⅔MCE_R
No amplification models (f₁, f₂, f₃) required
Hazard computed using full φ_S2S (IM^ref)
Enveloped hybrid approach:
${IM}^{surf} = max ({AF}^{{MCE}_{R}}, {AF}^{\frac{2}{3} {MCE}_{R}}) \times {IM}^{ref}$
Max 20% reduction

Ratio PSA/PSA_conv versus period from 0.01 to 10 s. Hybrid (MCE_R) in yellow drops as low as about 0.84 near 0.5 s; Hybrid (2/3 MCE_R) in red stays near 0.95 to 1.09; the dashed envelope follows the larger of the two. — Figure 5. Ratio of pseudo-spectral acceleration (PSA) from hybrid procedures to PSA from convolution. Results shown using hybrid with full input motion amplitude (labeled *MCE*_R) and 2/3 of full input, along with recommended envelope of the two results.

Section 5

Closure

Approvals

Approved for 2026 NEHRP Provisions and Commentary. Publication expected November 2026.
Final approval for ASCE 7/28 pending.

We advanced this proposal because

Non-ergodic site response is being used; needs to be reflected in NEHRP Provisions and Commentary for guidance / guardrails.
Current procedures in Sec. 21.1 are frequently unconservative.
Other improvements were needed.

References

Building Seismic Safety Council, BSSC (2020). NEHRP Recommended Seismic Provisions for New Buildings and Other Structures. Volume I: Part 1 Provisions, Part 2 Commentary. Report No. FEMA P-2082-1, Federal Emergency Management Agency, Washington DC.
Kramer SL and JP Stewart (2025). Geotechnical Earthquake Engineering, 2nd Edition. Taylor & Francis, New York, NY. 1060 pages.
Petersen MD, AM Shumway, PM Powers, EH Field, MP Moschetti, KS Jaiswal, et al. (2024). The 2023 US 50-State National Seismic Hazard Model: Overview and implications. Earthquake Spectra 40(1): 5–88.