Uncertainty in soil-water-characteristic curve and impacts on unsaturated shear strength predictions

Development in and on unsaturated soils is of particular importance in and semi-arid regions in the southwestern United States. Although a complete theory for the analysis of geotechnical problems involving unsaturated soils has been developed in the last three decades, it has not been widely implemented by practicing engineers. This fact might be attributed to several difficulties including (i) unsaturated soil properties are generally nonlinear functions of the pore water pressure; (ii) highly negative pore-water pressures are difficult to measure; (iii) laboratory and field-testing for unsaturated soils has proven to be costly, time consuming, and difficult to conduct; and (iv) the uncertainty associated with direct measurements and/or prediction of the unsaturated soil properties has not been addressed in detail by researchers.; The most important and accepted constitutive relations for unsaturated soils are intimately related to the soil matric: suction and hence with the soil-water characteristic curve (SWCC). For this reason, it is imperative to understand the impact of the uncertainty in the soil-water characteristic curve on predictions of the unsaturated soil properties. The uncertainty in the SWCC, and the impact of the corresponding uncertainty on the shear strength of unsaturated soils was emphasized in this research.; The uncertainty in the SWCC associated with direct suction measurements as well as the uncertainty associated with the prediction of the SWCC based on grain-size distribution (GSD) was investigated for three different soils that covered a typical range of soils frequently encountered within the field of geotechnical engineering. The uncertainty associated with direct suction measurements was found to be greatest around the air-entry value for the sand and the silt and for water contents at and lower than the residual condition for the clay. The range of variability was found to be related to the specific equation used to fit the suction data. The variability associated with the different methods available to measure suction was also investigated. Operator ability and experience, and the range of suction covered in the measurements were found to significantly impact the uncertainty in the SWCC.; The variability associated with the various predictive models of the SWCC based on GSD followed the same general trends observed for the experimental data. Surprisingly, the variability in the SWCC as predicted by GSD-based algorithms was found to be smaller than that associated with experimental measurements of suction by different operators.; The uncertainty in the experimentally measured SWCC was found to greatly impact the prediction of the unsaturated shear strength as the soil gets drier. The error in the prediction due to the uncertainty in the SWCC, was found as high as 10% at 500 kPa of suction for the sand, and as high as 40% at 1,000 kPa of suction for the silt and the clay.