Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Recorded Versus Synthetic Spectral-compatible Ground Motions: A Comparative Analysis of Structural Seismic Responses

Jungho Kim, Maijia Su, Ziqi Wang, Marco Broccardo

TL;DR

The paper assesses how well spectral-compatible synthetic ground motions reproduce recorded earthquakes in driving structural seismic responses. By calibrating both synthetic and recorded motions to a common unconditional target spectrum and evaluating five stochastic metrics on a 12-story and a 110-story structure, it leverages variance-based Sobol' analysis to separate aleatory from epistemic uncertainties. The results show synthetic motions closely match global response characteristics (mean, variance, distributions, correlations, and sensitivity patterns) within roughly 10–20%, supporting their use for routine design and parametric studies. However, extreme-event metrics, particularly for long-period structures, diverge substantially (often >50%), due to non-Gaussian features in recorded data that SGMMs underrepresent, highlighting limitations for rare-event risk assessment and tall-building dynamics. The findings advocate using synthetic motions for general seismic-response analyses while exercising caution for tail-dominated design scenarios, and point to future work in enhancing non-Gaussian aspects and exploring alternative spectrum-compatible generation approaches.

Abstract

This paper presents a comparative analysis of structural seismic responses under two types of ground motion inputs: (i) synthetic motions generated by stochastic spectral-compatible ground motion models and (ii) recorded motions from an earthquake database. Both ground motion datasets are calibrated to a shared target response spectrum to ensure consistent spectral median, variance, and correlation structure. Five key stochastic response metrics-probability distributions, statistical moments, correlations, tail indices, and variance-based global sensitivity indices-are systematically evaluated for two representative structures: a medium-period building and a limiting case of a long-period tower. The comparison accounts for uncertainties both from ground motion and structural parameters. The results reveal that synthetic motions closely replicate recorded motions in terms of global response behavior-including distributions, mean and variance, correlation structure, and dominant uncertainty sources-indicating their suitability for routine seismic design and parametric studies. However, substantial differences emerge in response extremes for long-period structures, particularly in metrics governed by rare events, such as higher-order moments and tail behavior. These differences, which often exceed 50%, can be attributed to the non-Gaussian features and complex characteristics inherent in recorded motions, which are less pronounced in synthetic datasets. The findings support the use of synthetic ground motions for evaluating global seismic response characteristics, while highlighting their limitations in capturing rare-event behavior and long-period structural dynamics.

Recorded Versus Synthetic Spectral-compatible Ground Motions: A Comparative Analysis of Structural Seismic Responses

TL;DR

The paper assesses how well spectral-compatible synthetic ground motions reproduce recorded earthquakes in driving structural seismic responses. By calibrating both synthetic and recorded motions to a common unconditional target spectrum and evaluating five stochastic metrics on a 12-story and a 110-story structure, it leverages variance-based Sobol' analysis to separate aleatory from epistemic uncertainties. The results show synthetic motions closely match global response characteristics (mean, variance, distributions, correlations, and sensitivity patterns) within roughly 10–20%, supporting their use for routine design and parametric studies. However, extreme-event metrics, particularly for long-period structures, diverge substantially (often >50%), due to non-Gaussian features in recorded data that SGMMs underrepresent, highlighting limitations for rare-event risk assessment and tall-building dynamics. The findings advocate using synthetic motions for general seismic-response analyses while exercising caution for tail-dominated design scenarios, and point to future work in enhancing non-Gaussian aspects and exploring alternative spectrum-compatible generation approaches.

Abstract

This paper presents a comparative analysis of structural seismic responses under two types of ground motion inputs: (i) synthetic motions generated by stochastic spectral-compatible ground motion models and (ii) recorded motions from an earthquake database. Both ground motion datasets are calibrated to a shared target response spectrum to ensure consistent spectral median, variance, and correlation structure. Five key stochastic response metrics-probability distributions, statistical moments, correlations, tail indices, and variance-based global sensitivity indices-are systematically evaluated for two representative structures: a medium-period building and a limiting case of a long-period tower. The comparison accounts for uncertainties both from ground motion and structural parameters. The results reveal that synthetic motions closely replicate recorded motions in terms of global response behavior-including distributions, mean and variance, correlation structure, and dominant uncertainty sources-indicating their suitability for routine seismic design and parametric studies. However, substantial differences emerge in response extremes for long-period structures, particularly in metrics governed by rare events, such as higher-order moments and tail behavior. These differences, which often exceed 50%, can be attributed to the non-Gaussian features and complex characteristics inherent in recorded motions, which are less pronounced in synthetic datasets. The findings support the use of synthetic ground motions for evaluating global seismic response characteristics, while highlighting their limitations in capturing rare-event behavior and long-period structural dynamics.

Paper Structure

This paper contains 16 sections, 2 equations, 16 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Background and comparative framework
  3. Input uncertainties in seismic analysis
  4. Metrics for stochastic seismic responses
  5. Synthetic and recorded ground motion datasets
  6. Structural models
  7. Comparative analysis of seismic responses: synthetic versus recorded motions
  8. Metric 1: Probability distributions
  9. Metric 2: Statistical moments
  10. Metric 3: Correlations between EDPs
  11. Metric 4: Tail indices
  12. Metric 5: Sensitivity indices to aleatory and epistemic uncertainties
  13. Synthesis of the comparative analysis
  14. Conclusions
  15. Algorithms for ground motion selection and generation
  16. ...and 1 more sections

Figures (16)

  • Figure 1: Response spectra of (a) recorded and (b) synthetic ground motions matching the target spectrum. Each plot shows 100 response spectra, with the median and 2.5%-97.5% quantiles of the target spectrum superimposed.
  • Figure 2: Comparison of the (a) spectral median and (b) variability between the target spectrum and the recorded/synthetic ground motions. Statistics are computed using 2,000 ground motions in each dataset.
  • Figure 3: Spectral correlations: (a) target correlation structure, (b) empirical correlations from recorded motions, and (c) empirical correlations from synthetic motions. Contours represent correlation coefficients between spectral accelerations at periods $T_1$ and $T_2$. Panels (b) and (c) are computed using 2,000 ground motions in each dataset.
  • Figure 4: Comparison of ground motion intensity measures between datasets: (a) PGA, (b) Arias intensity, and (c) significant duration. Histograms are based on 2,000 ground motions for each dataset.
  • Figure 5: Structural models: (a) medium-period building, (b) long-period tower, and (c) PDFs of uncertain structural parameters across four cases.
  • ...and 11 more figures