A SIMPLE METHOD FOR DETERMINING SOIL HYDRAULIC-PROPERTIES IN THE LABORATORY

Both soil water retention curve and unsaturated hydraulic conductivity data are often necessary for solving soil unsaturated flow problems. This study investigated two versions of a simple laboratory method for determining soil hydraulic properties of homogeneous and rigid soil samples. Three evaporation experiments (soil columns of 106-mm diam. and 60-mm height) were made with a silt loam soil packed at different bulk densities (0.96, 1.31, and 1.54 Mg m-3). Bulk density and water content profiles were determined by gamma attenuation (Cs-137 and Am-241), while pressure head profiles were recorded using microtensiometers, and average column water contents were computed using sample weight measurements. Evaporation experiments were also simulated by solving Richards' equation with a Galerkin finite-clement method. Three methods for the estimation of the hydraulic properties were used: the reference method, which requires the pressure head and water content profiles at several times, and the original and modified Wind methods, which require the average column water content at several times instead of the water content profile. Experimental results showed that the hydraulic properties obtained with the modified Wind method agreed well with those obtained with the reference method. Simulated data were used to compare the original and modified Wind methods, and to determine the effect of added noise on the determination of soil hydraulic properties. Using the numerical experiments, it was shown that the modified Wind method provided results slightly better than those of the original Wind method when the number of tensiometers was greater than three. When measurement errors were taken into account, estimation of the water retention curves using the modified Wind method was not very sensitive to experimental errors, but small uncertainties in tensiometric data influenced greatly the hydraulic conductivities determined in wet conditions.