Laboratory study of evaporative fluxes in homogeneous and layered soils
Bruch, Philip G.
Many problems faced by geotechnical engineers require the prediction of the water flux boundary condition at the soil surface. Soil covers for the decommissioning of landfills and tailings piles are one of the primary engineering applications in this category. The purpose of a cover may be to minimize infiltration, reduce radon fluxes, or prevent the generation of acidic leachates. Two cover concepts which are currently being evaluated for various applications are capillary barriers and moisture retaining covers. Both of these concepts involve multi-layered soil covers in which the soils and the layering combinations are chosen to achieve the design requirements. The processes of evaporation and moisture redistribution in homogeneous and layered soils were studied through the use of column evaporation tests. Six soil columns were tested: three homogeneous profiles, and three layered configurations. The soils used for the column evaporation tests were an aeolian sand (Beaver Creek sand), a natural silt, and a processed silt. The column evaporation tests were conducted with a constant head boundary condition (representing a shallow water table) at the base of the column for 31 days, after which the lower boundary condition was changed to a zero flux condition. The column evaporation tests were continued for approximately 30 days with the zero flux boundary condition at the base of the column. Measurements taken during the column evaporation tests included gravimetric water contents, temperatures, suctions (using tensiometers) and evaporation rates. A sensitivity analysis was conducted to determine the influence of soil hydraulic properties (saturated hydraulic conductivity, air entryvalue, andpore size distribution index) on the maximum evaporation rate sustainable from soil profiles corresponding to those used in the column evaporation tests. The sensitivity analyses indicated that the saturated hydraulic conductivity and the air entry value have a greater effect on the maximum evaporation rate than the pore size distribution index for the soils and layering combinations analyzed. The column evaporation tests were modelled using the computer program SWIM (Soil Water Infiltration and Movement). SWIM is based on the Richards equation for flow in unsaturated porous media and accounts only for liquid phase flow. The SWIM model was limited by the curve fitting functions used to determine moisture retention curves and hydraulic conductivity functions. The computed moisture contents from SWIM showed reasonable agreement for the natural silt and processed silt. The results for the Beaver Creek sand showed poor agreement due to the steep moisture retention curve for this material. This thesis indicates that selective layering of multi-layered soil covers has potential in designing capillary barriers or moisture retaining covers where the covers will be subjected to predominantly evaporative conditions.