On the Applicability of Shear Strain Index as a Proxy for Site Response Nonlinearity

Several recent studies have emphasized the advantages of conducting fully nonlinear site response analyses over the more conventional equivalent linear method, especially for strong ground motion and high frequency response predictions. Since the maximum soil shear strain (γ_(max)) reflects the degree of nonlinearity of the soil, recent studies have introduced the concept of a threshold γ_(max) to determine the conditions under which nonlinear analyses are necessary to provide credible site response predictions. However, as γ_(max) cannot be calculated a priori (i.e., before conducting any site response analysis), a proxy defined as the ratio of peak velocity of the input motion (PGV) to the time-averaged shear-wave velocity in the top 30 meters (V_(S30)), i.e., PGV/V_(S30), has been proposed instead. In this study, we quantify the appropriateness of PGV/V_(S30) (also referred to as shear strain index, Iγ) as a proxy for γ_(max), using a statistically significant number of ground motions and sites and using nonlinear site response simulations. We find that I_γ is a reliable proxy for γ_(max) only for sites with no sharp shear-wave velocity (V_S) impedance contrasts; otherwise it can underestimate γ_(max) by an order of magnitude or more. We lastly propose a correction factor called heterogeneity factor (HF), which is dependent on the velocity contrast of the V_S profile, and when applied upon I_γ, improves the mean and reduces the standard deviation of the I_γ – γ_(max) correlation