Evaluating the efficacy of fractal models for soil water retention curve estimation from particle‐size distribution for Hanford sediments

Abstract Soil water retention curve (SWRC) is a key characteristic widely used to study single‐ and multi‐phase flow and transport in porous media. Since its direct measurement is time‐ and labor‐consuming, numerous empirical and theoretical approaches were developed to estimate it from particle‐size distribution (PSD). In this study, we apply concepts of fractal theory and more specifically the Bird et al. and Hunt and Gee models to estimate the fractal dimension ( D ) and SWRC from the PSD for 138 samples from the Hanford site located in southcentral Washington State. To estimate D , we use the portion of the PSD showing fractal behavior, that is, the truncated power‐law trend between the particle mass and its size. We also predict the SWRC using the estimated fractal dimension as well as an optimized air‐entry value and soil porosity. Results show that the Bird et al. model estimates the SWRC more reasonably than the Hunt and Gee model, although both models underestimate D . For the Hanford samples used in this study, the pore‐solid fractal‐based Bird et al. model appears to be a better representation of the SWRC than the PSD‐based Hunt and Gee model. This suggests that compared to the Hunt and Gee model, the SWRC estimation by the Bird et al. model and its possible use in flow and transport modeling can produce more accurate results.