Suitability of a hydraulic-conductivity model for predicting salt effects on swelling soils.

Abstract

Many empirical approaches have been developed to analyze changes in hydraulic conductivity due to concentration and composition of equilibrium solution. However, in swelling soils these approaches fail to perform satisfactorily, mainly due to the complex nature of clay minerals and soil–water interactions. The present study describes the changes in hydraulic conductivity of clay (Typic Haplustert) and clay-loam (Vertic Haplustept) soils with change in electrolyte concen- tration (TEC) and sodium-adsorption ratio (SAR) of equilibrium solution and assesses the suit- ability of a model developed by Russo and Bresler (1977) to describe the effects of mixed Na-Ca-Mg solutions on hydraulic conductivity. Four solutions encompassing two TEC levels viz., 5 and 50 mmol c L –1 and two SAR levels viz., 2.5 and 30 mmol 1/2 L –1/2 were synthesized to equili- brate the soil samples using pure chloride salts of Ca, Mg, and Na at Ca : Mg = 2:1. Diluting 50 mmol c L –1 solution to 5 mmol c L –1 reduced saturated hydraulic conductivity of both soils by 66%, and increasing SAR from 2.5 to 30 mmol 1/2 L –1/2 decreased saturated hydraulic conductiv- ity by 82% and 79% in clay and clay-loam soils, respectively. Near saturation, the magnitude of the change in unsaturated hydraulic conductivity due to the change in TEC and SAR was of 10 3 - and 10 2 -fold, and at volumetric water content of 0.20 cm 3 cm –3 , it was of 10 14 - and 10 6 -fold in clay and clay-loam soils, respectively. Differences between experimental and predicted values of saturated hydraulic conductivity ranged between 0.6% and 11% in clay and between 0.06% and 2.1% in clay-loam soils. Difference between experimental and predicted values of unsatu- rated hydraulic conductivity widened with drying in both soils. Predicted values were in good agreement with the experimental values of hydraulic conductivity in clay and clay-loam soils with R 2 values of 0.98 and 0.94, respectively. The model can be satisfactorily used to describe salt effects on hydraulic conductivity of swelling soils in arid and semiarid areas, where groundwater quality is poor.