Contribution of capillary pressure to effective stress for unsaturated soils: Role of wet area fraction and water retention curve

Estimating the contribution of capillary pressure to the effective stress is critical for studying the mechanical and hydraulic behavior of unsaturated soils. Based on experimental evidence, we clarify that the degree of saturation (Sr) rather than matric suction (s) is more suitable as a variable for modeling Bishop's effective stress parameter (chi). The wet area fraction, defined as the ratio of the wet external area of soil particles to the total surface area, is then introduced as an alternative to the parameter chi. Because the contact area between soil particles and pore water is affected by both hydraulic (matric suction) and mechanical (volumetric deformation) states, the soil water retention curve (SWRC) is embedded in the proposed model to simulate the influences of pore size distribution and hydraulic hysteresis on shear strength. The proposed model requires only the material parameters of soils in the reference state and then can be used to predict the strength characteristics in other mechanical or hydraulic states. The shear strength data under different test conditions and the corresponding SWRC data are adopted to validate the new model. The results reveal that the wet area fraction could well predict the contribution of capillary pressure to the shear strength of unsaturated soils.

Automated summary

Estimating capillary pressure's contribution to effective stress is essential for studying mechanical and hydraulic behavior of unsaturated soils. Based on experiments, we find that degree of saturation (Sr) is a better variable than matric suction (s) for modeling Bishop's effective stress parameter (chi). We introduce wet area fraction as an alternative to chi, considering the influence of hydraulic hysteresis and pore size distribution on shear strength through the soil water retention curve (SWRC). This model requires soil material parameters in the reference state and can predict strength characteristics in different mechanical or hydraulic states.