Soil–Foundation–Structure Interaction of Inelastic Structural Systems on Unsaturated Soil Layers

Abstract

Recently, progress has been made toward understanding the seismic response of structures placed on unsaturated soil layers. A missing link, however, involves the influence and assessment of the underlying soil saturation conditions on the expected superstructure seismic demands. Simplified soil–structure interaction procedures that can be used to predict superstructure seismic demands have not been explicitly extended to incorporate the influence of unsaturated soil on the system response. In this paper, results from a series of six centrifuge tests are compared. In each test, an inelastic single-degree-of-freedom physical model was shallowly embedded in a sandy silt with a distinct water table elevation or a completely dry soil condition. The soil-structure system was subjected to a series of earthquake motions. The response of the system was evaluated to assess the influence of the soil saturation condition on the seismic response. Specifically, a conventional analytical procedure for predicting the influence of inertial interaction on the seismic response of the structure was extended to consider the water table elevation and underlying soil saturation condition and evaluated for its reliability. Analytical flexible-base modal parameters were compared with those determined from experimental results to judge the potential of the analytical procedure to be used in practice. Experimental results suggest that as the water table elevation was lowered from the fully saturated condition, both the flexible-base system period and damping ratio reduced. Therefore, the system behaved stiffer in the unsaturated soil compared with the dry and fully saturated conditions. The stiffer response reduced the seismically induced foundation settlements and rotations but amplified superstructure seismic demands in the form of accelerations, flexural drifts, and bending strains.