Factors affecting site-specific response analysis

Abstract

The engineering purpose of a site-specific response analysis is to estimate the uniform hazard acceleration spectrum on the ground surface for a selected hazard level. One of the mandatory components for site response analyses is one or more representative acceleration time histories that need to be scaled with respect to the calculated seismic hazard level for the selected site. The selection and scaling procedures of earthquake acceleration records play an important role in this approach. The effects and differences in using two different scaling approaches are studied: scaling with respect to ground motion parameters and response spectrum scaling. A set of homogeneous ground motion prediction relationships are developed for peak ground acceleration, peak ground velocity, root-mean-square acceleration, Arias intensity, cumulative absolute velocity, maximum spectral acceleration, response spectrum intensity, and acceleration spectrum intensity based on a uniform set of acceleration records for ground motion parameter scaling.

The uncertainties associated with site response analysis are considered as epistemic and aleatory uncertainties in source characteristics, soil profile, and soil properties. Aleatory variability is due to the intrinsic randomness of natural systems; it cannot be reduced with additional data (Passeri et al. 2020), however; its variability may be modeled by probability distribution functions. Thus, one possibility is to determine the probability distribution of the acceleration spectrum calculated on the ground surface for all possible input acceleration records, site profiles, and dynamic soil properties. The variability in the earthquake source and path effects are considered using a large number of acceleration records compatible with the site-dependent earthquake hazard in terms of fault mechanism, magnitude, and distance range recorded on stiff site conditions. Likewise, a large number of soil profiles may be considered to account for the site condition variability. The uncertainties related to dynamic soil properties may be considered as possible variability of maximum dynamic shear modulus in site response analyses. A methodology is proposed to estimate a uniform hazard acceleration spectrum on the ground surface based on the probabilistic assessment of the factors involved in site response analysis. The uniform hazard acceleration spectra obtained from a case study are compared with the spectra calculated by probabilistic models proposed in the literature.