Prediction of the relative soil hydraulic conductivity with fractal models

Abstract

The predictive capacity of four particular models of the general fractal model for hydraulic conductivity proposed by Fuentes et al. (2001) is evaluated: a) "small pore" model, b) "geometric mean pore" model, c) "neutralpore" model, and d) "large pore" model. The water retention curve is the basic information required by these models.

The power function reported by Braddock et al. (2001) with three form parameters is used to represent the water retention curve. This function contains the Brooks and Corey (1964) equation, with one form parametel; and the van Genuchten (1 980) equation, with two form parameters. In order to obtain closed-form equations for hydraulic conductivity, the form parameters are reduced to two; the corresponding water retention curve is implicit in the water content, and using one form parameter we obtain the van Genuchten equation. The prediction of the conductivity is realized with the closed-form corresponding to van Genuchten equation with one form parameter. We used fifty soils from the GRIZZLY soil database presented by Haverkamp et al. (1997). The comparison between experimental and predicted relative hydraulic conductivity permits to conclude that the four models can be used in order to estimate the relative hydraulic conductivity.