Comparison of Soil Nonlinearity (In SituStress–Strain Relation and G/Gmax Reduction) Observed in Strong‐Motion Databases and Modeled in Ground‐Motion Prediction Equations

Abstract

Earthquake ground motions are strongly affected by the upper tens of meters of the Earth's crust and consequently local site effects need to be included in any ground-motion prediction. It is increasingly common in ground motion prediction equations (GMPEs) to account for possible non-linear behavior of near-surface materials (soil). These non-linear site terms adjust observations made on soft soil sites to the ground motion expected on bedrock and hence allow these abundant soil records to be used within the regression analysis for the derivation of empirical GMPEs. These nonlinear site terms also allow rapid predictions of the expected ground motions on soil rather than requiring a site response analysis to be conducted. In this study we compare the signature on observed peak ground acceleration as a function of a strain proxy of non-linear soil behavior within four large strong-motion databases to the predicted signature from four recent GMPEs, three of which explicitly include nonlinear site terms. We find that observed non-linearity in the databases, interpreted in terms of strainstress relationships and reduction of shear modulus, is limited but that even this limited effect is underestimated by the non-linear site terms of the considered GMPEs, which suggests that predictions from these GMPEs could be biased for soft soil sites but also on bedrock. Some of this mismatch could be explained by the use of the average shear-wave velocity in the top 30m (Vs30) to characterize sites as well as errors in these values.

Long‐Time Relaxation Induced by Dynamic Forcing in Geomaterials

“…Alternative phenomenological models include “stick‐slip” models (e.g., Dieterich, ) and “soft‐ratchet” models (Vakhnenko et al, ). Another class of frictional models was considered for nonlinearly interacting seismic waves (e.g., Bonilla et al, , ; Guéguen et al, ; Wu, Peng, & Assimaki, ; Wu, Peng, & Ben‐Zion, ; Wu & Peng, ). A theoretical framework has also been proposed that uses images of crack networks in granular materials to construct a micropotential model for a medium containing elementary cracks with known properties (Aleshin & van den Abeele, ), as well as a hybrid P‐M space model (Gusev & Tournat, ) and a P‐M space thermal model (Scalerandi et al, ).…”

Testing Nonlinear Amplification Factors of Ground-Motion Models

“…We mainly follow the same data selection criteria as used for K18 (see Dataset and data selection criteria section), but we adopt an additional criterion to omit records that might have triggered non-linear soil response at certain sites. This means that we omit all records with PGA rock > 0.05 g from stations with V S30 < 760 m/s (Guéguen et al, 2019;Régnier et al, 2013), and use the records that contain only linear soil response to derive the GMM. This is to avoid biasing the GMM median predictions and the estimates of the δS2S s with non-linear soil response, making it a linear GMM.…”

“…However, for large ground motions and mostly for sites with soft soils, 1-D numerical simulations and several observations suggest that the site-specific amplification should decrease with increasing intensity of predicted ground motions for rock conditions (non-reference site method, Bonilla et al, 2005;Stewart et al, 2003;Field et al, 1997). Non-linear site effects have been shown to produce a shift of shear-wave energy towards frequencies lower than the fundamental resonance frequency of the soil column, accompanied by a relative decrease in amplification at high frequencies (Bonilla et al, 2005;Régnier et al, 2013;Guéguen et al, 2019).…”

A reconnaissance study in Izmir (Bornova Plain) affected by October 30, 2020 Samos earthquake

“…According to the recent studies in literature [39,[50][51][52][53] evolution of stress-strain relation on site can be traced by a stress-strain proxy represented by PGA (stress proxy) -PGV/V s (strain proxy). In some of the studies strain proxy is represented by PGV/V S,30 .…”