Estimating the hydraulic functions of dual‐permeability models from bulk soil data

Abstract

Predictions of nonequilibrium preferential flow and transport are limited as long as the fracture and the matrix pore system cannot be represented by separate hydraulic functions. Previous approaches to describe hydraulic properties of soils with bimodal or multimodal pore size distributions are still restricted to soils with a single porous continuum. The objective of this study was to develop and test a fitting procedure for estimating retention and conductivity functions for dual‐permeability models using bulk soil data. The estimation is based on a set of van Genuchten‐Mualem functions for the two pore systems. A stepwise procedure is carried out that (1) assumes 3 out of 11 parameters to be known, (2) determines initial values for the remaining eight parameters to be fitted, and (3) fits the set of hydraulic functions simultaneously to θ(h) and K(h) data to obtain the parameter estimates. This procedure was evaluated using a synthetic data set, and it was applied to laboratory‐measured θ(h) and K(h) data from a loamy Dystric Gleysol soil. The results show that the effect of a fracture pore system on the unsaturated hydraulic conductivity function might be underestimated if only observations of water retention are considered. The fitted dual‐continuum hydraulic conductivity functions match the unsaturated soil's conductivity data better than a bimodal single‐domain approach that predicts hydraulic conductivity based on fitted water retention functions. Moreover, the dual‐continuum functions provide hydraulic parameters for dual‐permeability preferential flow models.