Evaluation of the site amplification factors estimated by equivalent linear site response analysis using time series and random vibration theory based approaches

Abstract

The objective of this study is to estimate and compare the site amplification factors (AFs) using two different one dimensional (1-D) equivalent-linear (EQL) site response analysis approaches: the time series (TS) approach and the random vibration theory (RVT) based method. For this purpose, random soil profiles combined with different soil types, EQL soil properties, and unit weights are tested at several input ground motion levels. Analysis results showed that the AFs estimated by the TS-approach are systematically higher than the AFs estimated by the RVT-based method in the short period range (T < 0.5 s), especially when the bedrock peak ground acceleration is higher than 0.2 g. The relative difference between the AFs estimated in both methods is most prominent in granular soils: differences reach up to 35–40% compared to 10–15% for clays. On the other hand, the AFs calculated in this study are in good agreement with the empirical AF models utilized in recent ground motion models, indicating that the RVT-based AF models may be preferred in the future to cover a larger range of scenarios than the empirical datasets.