SIMULATING THE MOVEMENT OF A REACTIVE SOLUTE THROUGH A SOIL LYSIMETER COLUMN USING A FUNCTIONAL TRANSPORT MODEL

Trace Element Transport (TETrans), a one-dimensional, functional model of solute transport introduced by Corwin and Waggoner [1, 2], is tested for its ability to simulate the movement of boron through soil lysimeter columns under varying irrigation management strategies, irrigation water qualities and cropping regimes. The test period covers three years (1100 days) and represents one of the most extensive controlled tests both in duration and variation in boundary conditions for the transport and simulation of boron through an unsaturated soil system. Simulations for four weighing soil lysimeters are presented. Each lysimeter differs in irrigation management strategy, quality of the irrigation water applied and cropping strategy. The TE Trans model has the capability through a single parameter to account for deviations of water flow from strict piston displacement [13]. The parameter, termed the mobility coefficient, accounts for dispersion, diffusion and bypass, and is defined as the deviation of measured soil solution chloride concentrations from predicted chloride concentrations assuming piston-type displacement. By utilizing a temporally and spatially varying mobility coefficient to improve TETrans' ability to simulate the movement of chloride, and thereby water flow, it is postulated that the simulation of an accompanying reactive solute (i.e., boron) will also be improved. Comparisons of predicted concentrations to measured soil solution concentrations of boron show that the simulations using a temporally and spatially varying mobility coefficient based on chloride data were significantly improved over assumptions of strict piston-type displacement. However, the ability to simulate the transport of boron was still not completely successful when the conventional Langmuir adsorption isotherm model was used. Even the use of several alternative equilibrium models of adsorption-desorption was not completely successful. The best fit to the data was with a modified Langmuir adsorption model which was "kinetically irreversible." The unquestionable confidence in the simulation of the water flow and plant water uptake aspects of the transport phenomenon, as evidenced by excellent chloride simulations, led to the conclusion that the knowledge of the chemistry of boron in soils from a modeling standpoint is not well-known, particularly if approached from a simplified, practical standpoint which tries to keep measured physicochemical-transport parameters to a minimum. The use of nonhysteretic, equilibrium models of adsorption-desorption are insufficient to describe the chemical behavior of boron in a transient-state soil system where dynamic changes in ion composition, ionic strength and pH are occurring; and where complexities of adsorption-desorption hysteresis are occurring. A comparison of predicted soil solution boron concentrations using both TETrans and the Hanks' et al. [4] solute transport model to measured concentrations shows an improved ability of TETrans to simulate the movement of boron, particularly near the soil surface where preferential flow through cracks created from wetting and drying is a factor. The simulated results for both models were similar below 0.30 m.