Foundation performance is controlled significantly by the stress-strain behavior of the underlying soils. For geomaterials, the small-strain shear modulus Gmax is a fundamental stiffness applicable to both monotonic static and dynamic loading conditions, as well to both drained and undrained loading. Yet, Gmax is too stiff for direct use in computing foundation displacements. The main objectives of this research are to: (1) explore the scaled parallelism between the stress-strain-strength behavior of the single soil element response and the load-displacement-capacity of a shallow foundation system supported on soil; (2) develop a methodology for evaluating the performance of vertically-loaded footings using a rational framework based on the small-strain modulus Gmax, large-strain strength ( and #964;max or su) and strain at failure ( and #947;f); and (3) calibrate the proposed method using a foundation database of full-scale load tests under both undrained and drained conditions. In geotechnical practice, foundation bearing capacity is handled as a limit plasticity calculation, while footing displacements are evaluated separately via elastic continuum solutions. Herein, a hybrid approach is derived that combines these two facets into a closed-form analytical solution for vertical load-deflection-capacity based on numerical studies. Here, a non-linear elastic-plastic soil model was developed to simulate the stress-strain-strength curves for simple shearing mode (LOGNEP) for each soil element. The model was encoded into a subroutine within the finite difference program FLAC. A large mesh was used to generate load-displacement curves under circular and strip footings for undrained and drained loading conditions. With proper normalization, parametric foundation response curves were generated for a variety of initial stiffnesses, shear strengths, and degrees of non-linearity in the soil stress-strain-strength response. Soil stress-strain non-linearity is described by a logarithmic function (Puzrin and Burland, 1996, 1998) that utilizes a normalized strain xL that relates strain at failure and #947;f, shear strength ( and #964;max or su), and small-strain stiffness Gmax, all having physical meaning. A closed-form algorithm is proposed for generating non-linear load-displacement curves for footings and mats within an equivalent elastic framework. The proposed method was calibrated using a database of well-documented footing load tests where soil input parameters were available from laboratory and/or in-situ field test results.

This study is focused on developing a method to predict the dynamic behavior of mine tailings dams under earthquake loading. Tailings dams are a by-product of coal mining and processing activities. Mine tailings impoundments are prone to instability and failure under seismic loading as a result of the mechanical behavior of the tailings. Due to the existence of potential seismic sources in close proximity to the coal mining regions in the United States, it is necessary to assess the post-earthquake stability of these tailings dams. To develop the aforementioned methodology, 34 cyclic triaxial tests along with vane shear tests were performed on undisturbed mine tailings specimens from two impoundments in Kentucky. Therefore, the liquefaction resistance and the residual shear strength of the specimens were measured. The laboratory cyclic strength curves for the coal mine specimens were produced, and the relationship between plasticity, density, cyclic stress ratio, and number of cycles to liquefaction were identified. The samples from the Big Branch impoundment were generally loose samples, while the Abner Fork specimens were dense samples, older and slightly cemented. The data suggest that the number of loading cycles required to initiate liquefaction in mine tailings, NL, decreases with increasing CSR and with decreasing density. This trend is similar to what is typically observed in soil. For a number of selected specimens, using the results of a series of small-strain cyclic triaxial tests, the shear modulus reduction curves and damping ratio plots were created. The data obtained from laboratory experiments were correlated to the previously recorded geotechnical field data from the two impoundments. The field parameters including the SPT blow counts (N1)60, corrected CPT cone tip resistance (qt), and shear wave velocity (vs), were correlated to the laboratory measured cyclic resistance ratio (CRR). The results indicate that in general, the higher the (N1)60 and the tip resistance (qt), the higher the CSR was. Ultimately, practitioners will be able to use these correlations along with common state-of-practice geotechnical field methods to predict cyclic resistance in fine tailings to assess the liquefaction potential and post-earthquake stability of the impoundment structures.