Probabilistic characterization of shear strain-dependency of shear modulus and damping ratio for italian clays

The quantitative modeling of the shear strain-dependency of shear modulus and damping ratio is a fundamental step in the characterization of the dynamic cyclic behavior of soils. Such dependency can be investigated directly through specialized cyclic dynamic laboratory testing on undisturbed soil samples and through the subsequent fitting of analytical models. This approach is difficult to implement in routine practice, and it is convenient to refer to empirical models relating dynamic soil properties to other geotechnical properties such as plasticity, which are known to influence dynamic behavior, and which are easily and routinely measured. Like many technical and engineering disciplines, geotechnical engineering is experiencing a momentous shift in paradigm from the traditional deterministic approach to an uncertainty-based approach in which the inherent variability in datasets, uncertainties in measurements and engineering models used to transform in-situ and laboratory data to design parameters are modelled, processed, and reported explicitly, for instance through statistical and probabilistic techniques. An ever-increasing corpus of evidence from research and practice supports the argument that the uncertainty-based approach facilitates the improved understanding and engineering modelling of the unavoidable complexity in geotechnical systems, as well as cost-performance optimization in design and construction. This paper presents a novel method for the probabilistic characterization of shear strain-dependency of shear modulus and damping ratio for clays of differing levels of plasticity. Simple analytical relationships, which only require the availability or estimate of the plasticity index and which are applicable for a wide range of probabilities of exceedance, are derived from a high-quality database of resonant column and cyclic torsional shear test results on 170 undisturbed clay specimens from Italian sites. The paper concludes by demonstrating how to apply the results practically in a simplified one-dimensional ground response analysis.