As part of dynamic stability evaluations of earth embankments founded on laminated sand and clay deposits, the need to characterize their cyclic resistance became critical for the assessment of the embankment behavior and subsequent decisions on liquefaction mitigation measures. Due to lack of experimental and case history data on the effective stress behavior of such deposits, which are typically encountered in tidal and alluvial depositional environments, advanced laboratory tests on high quality undisturbed samples and numerical simulations using advanced constitutive models were performed to gain insight on liquefaction triggering and post-liquefaction accumulation of deformations under level and sloping ground conditions of such formations. Results indicated that the presence of clay laminations within sand deposits tends to increase the liquefaction triggering resistance. The increase in liquefaction resistance becomes more pronounced as the percentage of clay laminations increases. Numerical analyses results also indicated that void redistribution effects, often related to strain localization effects, tend to reduce as the thickness of sand laminations decreases, or as the clay lamination percentage increases.

Introduction

Performance-based concepts are increasingly used in earthquake engineering design practice. Nonlinear deformation analyses, involving dynamic finite element or finite difference methods, are frequently used for evaluating the effects of liquefaction on embankment dams and other major soil-structure systems during earthquakes. In engineering practice, the response of a geotechnical structure to strong ground motion is typically evaluated by means of empirical equations developed using either simplified system models or available observations from well-documented case histories. Despite their ease of use, empirical models may be overly simplistic in characterizing the response of complex systems and may not capture important phenomena associated with earthquake problems. In the context of performance-based design, numerical analyses combined with advanced cyclic testing for the calibration of constitutive models can offer an alternative, refined response model compared to simplified algebraic equations. Existing simplified procedures [1–۴] used for liquefaction assessment mainly focus on evaluating liquefaction triggering and post-liquefaction residual strength of sands based on in situ tests (i.e. CPT tip resistance or SPT blowcounts). In intertidal or alluvial environments, however, coarse-grained materials are frequently encountered within thinly layered deposits comprising alternating thin laminations of sands and clays. (Fig. 1). In such deposits, liquefaction assessment based on empirical correlations with CPT tip resistance may not be applicable due to the effect of the clay laminations on the CPT tip resistance measured within the thin “sandwiched” sand layers. An example CPT log in laminated sand and clay deposits including measured tip resistance, qc, friction ratio, Rf, pore water pressure response, u2 and soil behavior interpretations based on Robertson [5] soil classification is demonstrated in Fig. 2.