Estimating Microscale Spatial Distribution of Conductivity and Pore Continuity Using Computed Tomography

Despite the importance of information on the spatial distribution of unsaturated hydraulic conductivity (K-us) at microscale in soil, experimental determination of this property is difficult. The objectives of this study were to: (i) seek a simple, but reliable, procedure for the estimation of K-us at microscale and (ii) determine the sensitivity of the estimates to wetting-induced changes in selected structural parameters (porosity, epsilon, surface fractal dimension, D, and pore continuity, PC). Using computer assisted tomography (CAT), spatial distributions of soil water content (theta) and changes in epsilon that occurred during wetting were monitored at 2 by 2 mm resolution at 1-cm depth increments in water-stable (WSA) and unstable aggregate (USA) columns. The Fuentes theoretical equation, which requires data on saturated hydraulic conductivity (K-epsilon), theta, epsilon, and D, was used for the estimation of the spatial distribution of K-us. The spatial distribution of theta in WSA columns ranged from 0.113 to 0.317 cm(3) cm(-3) and from 0.175 to 0.567 cm(3) cm(-3) in the USA columns. The spatial distribution of epsilon(us) ranged from 0.46 to 0.74 and was used in the computation of D and PC. Values of K-epsilon ranged from 0.006 to 0.745 cm h(-1). The spatial distribution of K-us, ranged from 6.87 x 10(-4) to 1.49 x 10(-2) cm h(-1) in WSA compared with 7.3 x 10(-4) to 4.11 x 10(-2) cm h(-1) in USA. Pore continuity, theta, D, and initial aggregate diameter (x) accounted for 94 to 95% of the variability in K-us distributions. The results indicate that reliable estimates of K-us distributions at microscale ran be computed from single-source CAT derived data on theta and epsilon.