This paper deals with the applicability and validation of three predic
tive models for the estimation of soil hydraulic parameters (soil wate
r characteristic and unsaturated hydraulic conductivity) by using rout
inely available soil data. Comparisons were made of the models of Camp
bell [Campbell, G.S., 1974. A simple model for determining unsaturated
conductivity from moisture retention data. Soil Sci. 117, 311-314; Ca
mpbell, G.S., 1985. Soil Physics with BASIC. Elsevier, New York]; regr
ession equations of Vereecken et al. [Vereecken, H., Feyen, J., Maes,
J., Darius, P., 1989. Estimating the soil moisture retention character
istic from texture, bulk density, and carbon content. Soil Sci. 148, 3
89-403; Vereecken, H., Maes, J., Feyen, J., 1990. Estimating unsaturat
ed hydraulic conductivity from easily measured soil properties. Soil S
ci. 149, 1-12] and the model of Gregson, Hector, McGowan (GHM model) [
Gregson, K., Hector, D.J., McGowan, M., 1987. A one-parameter model fo
r the soil water characteristic. J. Soil Sci. 38, 483-486.]. The predi
ctions of the above models were evaluated based on measured and calcul
ated data from six German soils. Regression analysis showed that the G
HM model, based on a single paired measurement of water content vs. ma
tric head, produced noticeably better results than the other two model
s, which require input of textural properties of the soils. It is reco
mmended that hydraulic parameters be estimated from paired measurement
s of water content vs, matric head rather than from textural propertie
s of soils. (C) 1998 Elsevier Science B.V. All rights reserved.