Using bimodal lognormal functions to describe soil hydraulic properties.

Accurate parameterization of the soil hydraulic properties represents a key issue for the modeling of soil water transport processes. Th e more complex the soil structure, the more crucial this requirement becomes. In dealing with this problem for structured and well-aggregated soils, we have pursued the general objective of developing hydraulic relationships whose parameters characterize the soil’s pore size distributions, thereby providing a physically based framework for the hydraulic relationships of bimodal soils. In our work, we assumed that the soil water retention function is determined by linear superposition of two distinct pore domains, which can be associated with textural and structural retention behaviors, respectively. Th e composite soil water retention function was described by Kosugi’s lognormal function, with parameters being directly associated with the mean and variance of the soil pore size distribution for each pore domain. Th e two components of soil water retention were linked by a weighting factor to which a physical meaning can also be given. An important and practical advantage of the proposed bimodal water retention function is that a closed-form analytical expression is obtained for the bimodal hydraulic conductivity function using pore size distribution parameters. Th is is relevant because we suggest that soil hydraulic properties can be characterized by the soil particle size distribution. Sensitivity analysis and comparisons with experimental data were used to evaluate the proposed bimodal lognormal hydraulic functions and to demonstrate their increased eff ectiveness in predicting the hydraulic conductivity characteristic of soils.