Estimating unique soil hydraulic parameters for sandy media from multi-step outflow experiments

Estimating unique soil hydraulic parameters is required to provide input for numerical models simulating transient water flow in the vadose zone. In this paper, we analyze the capability of six soil hydraulic functions to provide unique parameter sets for sandy soils from multi-step outflow data. Initial parameter estimates and experimental boundary conditions were explored to determine their affect on the uniqueness of soil hydraulic functions. Of the hydraulic functions tested, the lognormal distribution–Mualem (LDM) function provided the best performance and a unique solution for error-free numerically generated multi-step outflow data. For experimental multi-step outflow data with inherent measurement errors, the LDM function again showed better performance and uniqueness than the van Genuchten–Mualem and Gardner–Mualem functions. In experiments with different boundary conditions, the LDM function provided the best fitting ability, resulting in unique parameter sets when the intrinsic permeability (k) was fixed at its measured value. The experiment that had a greater number of pneumatic pressure steps, thereby causing a lower flow rate, provided better fitting ability and more unique solutions than faster experiments.