Indirect Methods

This chapter distinguishes between semiempirical and empirical methods to predict the soil water retention and hydraulic conductivity. Two scenarios are prevalent for semiempirical predictions. First, particle-size distribution (PSD) models have been widely used to predict the hydraulic conductivity curve, whose measurement is more cumbersome than the retention curve, from experimental results. Secondly, PSD models can be employed to predict the retention curve from the particle-size distribution, bulk and particle densities, and other soil properties. Although there are many different functions that can be used to describe and predict unsaturated hydraulic properties, the models by Mualem and van Genuchten have been particularly popular and appear to provide an adequate description of the hydraulic properties of most soils. Empirical pedotransfer functions are used to predict soil hydraulic properties from soil texture and other readily available soil information without relying on a particular conceptual model. Empirical methods include regression analysis and neural network analysis.