EXPERIMENTAL TESTING OF TRANSIENT UNSATURATED FLOW THEORY AT LOW WATER-CONTENT IN A CENTRIFUGAL FIELD

Experimental measurements and theoretical predictions of transient moisture conditions have been compared for a sandy soil approaching hydrostatic equilibrium in a centrifugal field. Starting near saturation, samples were centrifuged at constant speed with a constant suction at the outflow boundary. Water flowed freely out of the sample through a porous plate. Step increases in centrifuge speed produced transient moisture conditions suitable for comparison between experiment and theory. Measurements of electrical conductivity by a direct contact four‐electrode technique indicated the water content according to a calibration based on known moisture conditions at various equilibrium states. A specially modified centrifuge permitted electrical measurements during centrifugation. For comparison, the transient water contents were computed by a finite‐difference solution of Richards' equation (modified by replacing gravitational with centrifugal potential), using soil characteristics measured previously by steady state techniques. The time dependence of water content changes, used as the basis for comparison between experiment and theory, shows agreement which is reasonable given the degree of uncertainty of the measurements. The experiment confirms, within a factor of 4, the validity of Richards' equation for moisture conditions as dry as 25% of saturation, over a hydraulic conductivity range of 5 × 10−11 to 1 × 10−8 m/s, and in a centrifugal field up to about 200 g. This paper is not subject to U.S. copyright. Published in 1990 by the American Geophysical Union.