Evaluating the linear dielectric constant-electrical conductivity model using time-domain reflectometry

Abstract

Recently, Hilhorst (2000) presented a theoretical model describing a linear relationship between bulk electrical conductivity, sigma sub(a), and dielectric constant, K sub(a), in moist soil. Hilhorst (2000) found that using this linear relationship, measurements of the pore water electrical conductivity, sigma sub(p), can be made in a wide range of soil types without soil-specific calibrations. If accurate measurements were to be made, only one parameter had to be optimized. In the present study, the linear model was evaluated using detailed time-domain reflectometry (TDR) measurements in coarse sand, medium sand and loamy sand. It was shown that the measured slope of the linear model was lower than expected. A reason for this might be that the K sub(a) of the pore water is smaller than that of free water due to the presence of bound water. The results showed that the linear model was as good as other commonly used models for sigma sub(p) predictions.Original Abstract: Recemment, Hilhorst (2000) a presente un modele theorique decrivant la relation entre la conductivite electrique volumique sigma sub(a) et la constante dielectrique K sub(a) pour un sol humide sous la forme d'une relation lineaire. Hilhorst (2000) a mis en evidence que, grace a cette relation lineaire, les mesures de la conductivite electrique de l'eau des pores sigma sub(p) peuvent etre realisees sur une large gamme de sols sans calibration particuliere pour chaque sol. Si les mesures devaient etre precises, seul un parametre devait etre optimise. Dans cette etude, le modele lineaire a ete evalue grace a des mesures par reflectometrie en domaine temporel (TDR) dans des sables grossier, moyen et riche en matiere organique. La pente du modele lineaire obtenue est plus faible que prevu. Une explication possible peut etre que la valeur de K sub(a) est plus faible pour l'eau des pores que pour l'eau libre, a cause de la presence d'eau liee. Les resultats montrent que le modele lineaire est aussi bon que d'autres modeles couramment utilises pour l'estimation de sigma sub(p).