Influence of Bidirectional Horizontal Shaking on Seismic Response of Structure on Liquefiable Soils

In practice, a unidirectional horizontal ground motion is often selected as input during the design of geotechnical systems. This recording is obtained either as the component with the greatest intensity [e.g., in terms of a primary intensity measure, IM, like peak ground acceleration (PGA)] or as the maximum rotated component (RotD100, again in terms of an IM like PGA). Although seismic ground deformations are generally assumed to be governed by the direction of maximum shaking intensity under unidirectional shaking (1D), prior experimental research has shown that bidirectional horizontal shaking (BD) can increase seismic settlements in dry sand or the buildup of excess pore water pressures in saturated granular soils. In this paper, we explore the influence and relative importance of bidirectional horizontal shaking numerically (as compared to maximum rotated 1D shaking) on the seismic response of shallow-founded structures on liquefiable soils. A suite of 3D, fully coupled, nonlinear finite-element analyses, previously validated with centrifuge recordings, are employed to evaluate soil-structure interaction (SSI) and its performance under bidirectional seismic loading. In general, bidirectional shaking is shown to amplify the foundation's permanent settlement and permanent rotation by up to 10%-50% and 5.2-6.8 times greater than that under unidirectional shaking, respectively, due to its greater net seismic energy content compared to the models subject to unidirectional shaking. The limited numerical sensitivity study showed that evolutionary IMs correlated better with permanent foundation settlements, while the peak transient IMs could provide a better correlation with residual foundation rotations. Although much experimental and numerical research is needed in this area, the limited results presented here point to the importance of considering bidirectional shaking (as opposed to unidirectional maximum rotated component) when evaluating the key engineering demand parameters that quantify building performance on liquefiable soils.