For dynamic excitation, it is convenient to idealize homogeneous soil under a base mat by a semi‐infinite truncated cone. It is easy to analyze the cone model for vertical and horizontal translation, as well as for rocking and torsional rotation. The accuracy by comparison to rigorous half‐space solutions is quite adequate for practical applications. Time‐domain computational methods for translational and rotational motions are described in both the stiffness and flexibility formulations and elucidated by examples. The infinite cone is dynamically equivalent to a discrete element representation of the soil, consisting of an interconnection of a small number of masses, springs, and dashpots. As an alternative to the physical‐component model, the response may be determined directly by simple recursive numerical procedures. The recursive methods are exact and particularly well suited for hand calculations of short‐duration excitations.

It is quite difficult to compute rigorously the dynamic response of a footing or base mat on a soil layer resting on rigid rock. As an alternative to the rigorous approach, the semi‐finite truncated cone models appropriate for footings on unlayered soil may be employed in modified form. The physically motivated concept of echoes converts the response of the footing on unlayered soil to that of the footing on the layer. The calculations, performed exclusively in the familiar time domain, are in remarkable agreement with rigorous solutions, both for static and dynamic excitation. The necessary echo constants are evaluated for translational and rotational motions in either the flexibility or stiffness formulation of the response. The concept of echoes captures the important effect of the cutoff frequency. The computational procedures, which may be implemented by hand for excitations of short duration, are illustrated by a practical example.

The development of shear strain in a soft, marine clay undergoing monotonic shear in torsional and simple shear equipment has been studied. The ratio between the irrecoverable, or plastic, shear strain to the overall shear strain that develops on a particular plane was found to be related to the overall shear strain developed on that plane by a continuous, increasing function. The function was found to be linear on a semilogarithmic scale up to a shear strain of the order of 25%, and it appears to be relatively unaffected by the previous development of shear strains on other planes. A similar functional relationship was found to be applicable, also, for shear strain increments. The Masing model was found to reasonably predict the shape of the unloading branch of the shear stress‐strain curve under various boundary conditions, making it possible to estimate plastic strains from monotonic loading tests. On the basis of observations made in the study, it appears that kinematic rather than isotropic work‐hardening models are best suited to describe the stress‐strain behavior of soft clay.

To study the statistics of the settlement of pile foundations, the uncertainty in the Young's modulus of the soil must first be estimated. This paper considers the practical aspects of estimating the uncertainty in the model for quantifying the Young”s modulus of the soil as well as the uncertainty due to the inherent variability of the basic soil property, namely, the shear strength. The uncertainty in the model is evaluated by treating the correlation parameter as a random variable; the soil is modeled as a spatially random medium. Through a numerical example it is shown how these two uncertainties can be simultaneously considered in evaluating the reliability of the pile foundation against an allowable settlement limit using design charts. The design charts were formulated based on first‐order computations and validated in this paper by comparing with second‐order results over the range of values of practical interest.

The response of the soil plug within driven, open‐ended pipe piles is very different tinder the dynamic conditions of installation and the static conditions of loading during service. This paper addresses the latter aspect and describes a combined experimental and numerical study of the response of soil plugs in open‐ended pipe piles. The work focuses on the partially drained (static) loading relevant to offshore applications, and the experimental work is conducted using calcareous sand from Bass Strait, Australia. Model tests are conducted in pipe piles of 25‐mm and 100‐mm internal diameters, with loading rates as great as 6 MPa/s, using a downward hydraulic gradient to achieve appropriate effective stress profiles in the soil plug. The experimental results are assessed within the framework of analytical solutions of the drained and undrained response of the soil plug and numerical studies of the partially drained problem that allow the results to be extrapolated to prototype conditions. Example applications are given for pile geometries and loading rates typical for Bass Strait.

Shaking‐table tests to measure the dynamic interface shear strength properties between geotextiles and geomembranes were performed and the results are reported. It was observed that there is a limited shear stress, hence a limited acceleration, that can be transmitted from one geosynthetic to another. Beyond this acceleration, relative displacement along the geosynthetic interface is initiated. Hence, the primary concern about the dynamic response of a geotechnical facility that incorporates geosynthetics should be the permanent relative displacement that may accumulate along the geosynthetics interfaces. The dynamic interface friction angles measured at the onset of relative displacement between the geosynthetics are not appreciably different from those obtained from static tests. It was also observed that the dynamic interface friction angle increased slightly as the table acceleration increased beyond the level causing initial sliding. This increase could not be attributed to the potential effect of the frequency of motion of the shaking table and normal stress. Finally, an innovative use of geosynthetics as base isolation for earthquake hazard mitigation is proposed and its applicability is demonstrated. Further exploration of this application of geosynthetics may lead to products that can provide varying degrees of base isolation, especially in geotechnical earthquake engineering.

The possibility of using experimentally determined foundation impedance functions (dynamic stiffness) to infer the dynamic properties of the underlying soil deposit is explored. The foundation is modeled as a rigid block resting on a layered viscoelastic half‐space consisting of several parallel layers overlying a half‐space. The shear‐wave velocities and material damping ratios in the layers are determined by minimizing the difference between the experimental impedance functions obtained from forced vibration tests of the foundation and theoretical impedance funtions obtained by an integral equation technqiue. The effects that the number and range of the selected frequencies used in the analysis have on the accuracy of the identification process are studied in detail. The effects of using an incomplete set of impedance functions as well as those resulting from random errors in the data are also investigated. Finally, the use of a variability penalty to minimize the variation of inferred properties from layer to layer is considered.

This paper presents a detailed investigation of the influence of key mechanical and geometrical parameters on the rocking impedance of rigid strip foundations embedded in layered soil. Geometrical parameters include the depth of embedment, the thickness of the top soil layer, and the degree of contact of foundation sidewalls with surrounding backfill soil. The influence of the relative shear wave velocities of the layers for a two‐layered soil profile is also investigated. The results are presented in the form of simple and versatile dimensionless graphs. This investigation is conducted using an advanced algorithm of the boundary element method incorporating isoparametric boundary elements. Higher order quadratic elements are used since they can model accurately the wavy nature of the dynamic problem.