Optimization of Hydraulic Functions from Transient Outflow and Soil Water Pressure Data

Inverse solution techniques currently used for estimating unsaturated soil hydraulic functions from laboratory outflow experiments use cumulative outflow only in combination with initial and final soil water pressure head values. Additional soil water information is needed to improve the estimation procedure and to minimize uniqueness problems. It was the objective of this study to experimentally explore the feasibility of using both cumulative outflow and soil water pressure head data in the inverse solution for laboratory determination of soil hydraulic functions using both one-step and multistep outflow experiments. Soil water pressure head was measured with a microtensiometer and pressure transducer. Desorption experiments were performed under both pressure and suction for the following soils: Yolo silt loam (flne-silty, mixed, nonacid, thermic Typic Xerorthent), Panoche loam (fine-loamy, mixed [calcareous], thermic TypicTorriorthent), Hanford sandy loam (coarse-loamy, mixed, nonacid, thermic Typic Xerorthent), and Oso-Flaco fine sand (coarse-loamy, mixed Typic Cryorthod-fine-loamy, mixed, raesic Ustollic Haplargid). Water retention curves optimized from cumulative outflow alone were compared with those obtained from outflow and soil water pressure head measurements for both one-step and multistep outflow experiments. Computer optimization of the retention curve by the inverse solution technique using transient outflow experiments was greatly improved when cumulative transient outflow data were combined with simultaneously measured soil water pressure bead data. Pressure and suction experiments yielded equally good results for one-step and multistep desorption. Moreover, the addition of soil water pressure head data resulted in unique parameter values for the optimized soil hydraulic functions under our experimental conditions. P OF WATER MOVEMENT in soil as well as the fate of salts and contaminants under different combinations of soil, climatic, and management conditions is dependent on availability of representative Hydrologic Science, Dep. of Land, Air, and Water Resources, Veihmeyer Hall, Univ. of California, Davis, CA 95616. Received 4 June 1992. *Corresponding author. Published in Soil Sci. Soc. Am. J. 57:1167-1175 (1993). soil hydraulic properties. Both the hydraulic conductivity, K, as a function of water content, 0, or soil water pressure head, h, and the soil water retention function 6(h) influence the rate at which water and solute move in the vadose zone and thus affect travel time of contaminants toward groundwater. Direct methods for the determination of these highly nonlinear functions exist (Klute, 1986; Klute and Dirksen, 1986). These methods, however, require the experiments to reach several stages of steady-state or equilibrium conditions. Moreover, they require restrictive initial and boundary conditions, which make them time consuming and expensive (van Dam et al., 1990). The parameter estimation technique involves the indirect estimation of soil hydraulic functions by numerical solution of the equation governing the flow process, subjected to the imposed boundary conditions. First, the hydraulic properties are assumed to be described by an analytical model with unknown parameter values. An experiment is set up under controlled conditions with prescribed initial and boundary conditions. During the experiment, one or more flowcontrolled attributes (auxiliary variables) are measured. Subsequently, the flow equation is solved numerically using the parameterized hydraulic functions with initial estimates provided. The parameters of the hydraulic functions are optimized by minimization of the objective function containing the sums of squared deviations between observed and predicted auxiliary variables, using repeated numerical simulation of the flow process. This iterative inversion of the flow equation is in contrast to direct inversion techniques as used in analytical solutions. By far, laboratory outflow experiments have been the most attractive for estimation of soil hydraulic functions by computer optimization. The outflow method was introduced by Gardner (1956) for determining the unsaturated soil water diffusivity and unsaturated hydraulic conductivity by analytical techniques. This analytical technique has since been