Importance of Site-Specific Dynamic Soil Properties for Seismic Ground Response Studies: Ground Response Analysis

Introduction

In seismic Ground Response Analysis (GRA), the effect of local geology is significantly important since the response of any structure, subjected to the vertically propagating horizontal shear waves, depends on the regional seismicity, source mechanism, geology and local soil conditions (Boore 1972; Kramer 1996; Nath et al. 2013; Kumar et al., 2014a, 2014b). The bedrock motions get amplified or de-amplified within the soil stratum and, therefore, the estimation of dynamic soil properties of the subsurface of any site, prior to conducting the GRA, is extremely important. GRA is one of the important step to foresee the potential consequences of earthquake motion prior to the earthquake occurrence. Several GRA methodologies (such as linear, equivalent linear and non-linear) are available to evaluate different response parameters of the site (Kramer, 1996). The response spectrum at various layers of a stratified soil deposit, obtained as an outcome of GRA, are useful for earthquake based design of geotechnical structures. Almost all methodologies require strain dependent dynamic properties of soils, in terms of the shear modulus reduction and damping ratio curves, as the essential input parameters (Kumar et al., 2017a). Several researchers have performed GRA for Indian cities using DEEPSOIL (Hashash et al., 2016) and SHAKE2000 (Ordonez, 2000), as summarized Table 1. GRA for Guwahati city has been reported by Kumar and Krishna (2013) and Basu et al. (2017) using existing material model proposed by Seed and Idriss (1970) for sandy soil, Vucetic and Dobry (1991) for clay soil, or Ishibashi and Zhang (1993) for sandy and clay soils. Since these soil models have been established for the soils of different regions and of different compositions, the direct adoption of the same may not be suitable for the regional soils considered in the present study. Although the existing soil models were used by Kumar and Krishna (2013), Basu et al. (2017) and Kumar et al. (2017b), the use of dynamic soil properties of regional soils were not reported in the earlier literatures. Shukla and Choudhury (2012) have recommended the use of actual dynamic soil properties for more realistic outcomes from GRA. It can be stated that, in the absence of proper site-specific dynamic soil properties, use of existing soil models might lead to inaccurate estimation of the parameters involved in the earthquake resistant designs.