Textural averages of saturated soil hydraulic conductivity predicted from water retention data

Modeling of transport processes in soils relies on predicted values of saturated hydraulic conductivity, K,. The objective of this work was to derive the parameters of a K, power-law model from water retention (WR) properties defined at the inflection point of s-shaped WR functions. The proposed model uses water content, theta(inf), pressure potential, psi(inf), and the slope, S, at the inflection point to derive an effective porosity, phi(inf), (air filled porosity at theta(inf)) and a pore distribution index, lambda*, (estimated from WR data between pressure potential at air-entry and psi(inf) or as S/theta(inf)). The phi(inf) is raised to 3 -lambda* or 3 -S/theta(inf) and multiplied by C(inf) D(inf) where C(inf) is a constant and pore diameter is defined as D(inf)proportional to psi(-1)(inf). The model was developed from: A) a dataset containing 374 measurements of K, together with detailed WR and particle size distribution information, B) a subset of the HYPRES dataset containing 1827 values of K. and their corresponding sand, silt and clay percentages and parameter values from an s-shaped WR model; and tested for consistency with C) a published dataset containing textural averages of parameters of an s-shape WR model and K, values developed from 2134 and 1036 samples, respectively. Linear regression between 16 (A), 10 (B) and 12 (C) textural averages of logK(s) and their predicted values with C(inf)=20 resulted in R(2) values of 0.95, 0.82 and 0.82, respectively. The proposed model is physically sound and except for C(inf) all its parameters are derived from one point in the WR curve. (c) 2008 Elsevier B.V. All rights reserved.