Parameters affecting estimation of unsaturated permeability of soils

Abstract

Flow through the unsaturated zone of soil contributes to a variety of geotechnical and geo-environmental problems. Unsaturated permeability is the most important hydraulic property governing the flow process. Therefore, knowledge of unsaturated permeability is crucial in the analysis of the flow process in the unsaturated zone. Unsaturated permeability of soil can be directly measured in soil laboratories; however, this requires not only expensive equipment but also a well-trained technician. It is also a time consuming and tedious task. To overcome the high cost and challenges associated with direct measurement, permeability can be estimated by theoretical models from soil-water characteristic curve (SWCC). To date, there are numerous estimation models that can be used to obtain the permeability of unsaturated soil from SWCC. However, each model results in a different estimation curve. The reason for this difference is not well understood. The objective of this study is to identify the controlling parameters that affect the estimation of unsaturated permeability of soils. The SWCC equation, relative permeability equation and measured SWCC data range were identified as possible controlling parameters. A matrix of unsaturated permeability estimation models was proposed by combining four SWCC and three relative permeability (kr) equations to evaluate the effect of controlling parameters on estimation of unsaturated permeability of soils. The matrix of unsaturated permeability estimation models was evaluated using published data containing experimentally measured SWCC and permeability data. It was found that the range of SWCC measurements has the greatest effect on the estimated permeability functions, followed by the SWCC equation. The relative permeability equation has the least effect on variation in the estimation. A triaxial permeameter system was modified to directly measure the unsaturated permeability of six different kaolin-sand soil mixtures in this study. Soil-water characteristic curve tests were conducted over a wide range of suctions for these soil mixtures. The measured SWCC data were used to estimate the unsaturated permeability of the soils. The estimated values of unsaturated permeability of the soils were validated using the directly measured unsaturated permeability data. It was found from the results that the unsaturated permeability estimation models could reasonably estimate directly measured unsaturated permeability data when the full measurement of SWCC data covering a large range of suction were used. Even though, the full measurement of SWCC data needs less expensive equipment as compared to that of direct measurement of unsaturated permeability, the full measurement of SWCC data still requires a substantial testing time. Therefore, a procedure is proposed to estimate SWCC values at high suction using the grain size distribution and measured SWCC data at 100 kPa. The proposed procedure could reasonably estimate directly measured unsaturated permeability data. This procedure requires significantly shorter testing time and inexpensive equipment which are available in most of laboratories.