UNSATCHEM: Unsaturated Water and Solute Transport Model with Equilibrium and Kinetic Chemistry

Numerous models have been developed for predicting major ion chemistry in the soil zone and in recharge to groundwater. Soils that contain CaCO, are prevalent in arid and semiarid regions, as well as in humid and temperate regions that have been glaciated or contain carbonate bedrock. Under these conditions, carbonate-solution reactions and ion exchange are the dominant chemical processes. In this model we couple one-dimensional unsaturated water and solute transport with a major ion chemistry routine and plant water uptake. The model has several unique features, including expressions relating reductions in hydraulic conductivity to chemical factors, prediction of CO; partial pressure in the root zone based on a CO, productionmultiphase transport submodel, kinetic expressions for silicate weathering, calcite precipitation-dissolution, and dolomite dissolution, representation of B adsorption using the constant capacitance model, a new method for predicting cation-exchange selectivity, the option to use Fitzer ion interaction expressions for high ionic strength, and a plant growth submodel that includes water, salinity, and O2 stress. The chemical submodel considers equilibrium ion exchange, as well as various equilibrium and kinetic expressions for precipitation and dissolution of soil minerals, including gypsum, Mg carbonates, and sepiolite. The use of a predictive submodel for CO, production and transport allows for the calculation of CO, concentrations with depth and time. This enables us to avoid the assumption of constant CO, distribution or constant pH required by previous models. Use of kinetic expressions for carbonate chemistry allows a more realistic simulation of soil and groundwater solution composition as well as simulations of carbonate redistribution and climatic change with time. M THE TRANSPORT and concentration of major soluble ions in and below the root zone is a requisite for predicting groundwater quality as well as for managing irrigation and fertilization practices. Many models have been developed in the past 20 yr to quantify the physical and chemical processes that affect the transport of major ions. The hydrological models, solute transport models, and aqueous chemical equilibrium models were developed independently and only later were these three kinds of models coupled. Among these, Jury et al. (1978) developed a model for steady-state water flow in the unsaturated zone that considered ion USDA-ARS, U.S. Salinity Lab., 450 W. Big Springs Rd., Riverside, CA 92507-4617. Received 2 Aug. 1996. *Corresponding author ([email protected]) Published in Soil Sci. Soc. Am. J. 61:1633-1646 (1997). exchange and calcite equilibrium. Schulz and Reardon (1983) presented a mixing cell analytical model for groundwater transport that considered steady-state water flow, and a simplified chemical model (no ion pairing or complexation) with ion exchange and calcite equilibria. Forster and Gerke (1988) proposed a multicomponent transport model based on steady-state water flux that considered a mobile-immobile water concept and carbonate equilibrium. Numerous other models have been developed for steady-state groundwater flow coupled with chemical equilibrium (Jennings et al., 1982; Walsh et al., 1984; Miller and Benson, 1983; Narasimhan et al., 1986; among others). Simulation of soil water flow and soil chemical processes requires consideration of water and solute transport under variable water content conditions. Modeling of unsaturated water flow coupled to equilibrium chemistry has been undertaken by only a few researchers. Robbins et al. (1980a,b) combined an unsaturated water flow model with an ion exchange and carbonate chemistry and gypsum submodel that assumed fixed soil pH. Subsequently, Wagenet and Hudson (1987) developed a similar model that input fixed CO2 at various depths instead of assuming fixed pH. Russo (1986) combined the solution chemistry of the Robbins et al. (1980a) model with the Bresler (1973) transport model. Yeh and Tripathi (1989, 1991) coupled a generalized chemical equilibrium model with unsaturated water flow. Application of the Yeh and Tripathi model to soil environments requires specification of total inorganic C and fixed pH. The model does not contain any provisions for plant water extraction. Simunek and Suarez (1994) developed a two-dimensional model with unsaturated water flow and major ion chemistry. Evaluation of these models under field conditions is very limited. Only Robbins et al. (1980b) tested their model by comparing its results with experimental data obtained from a lysimeter study. However, in a field study, Dudley et al. (1981) indicated that the Robbins et al. (1980b) model could predict salinity, but not specific ion concentrations. These difficulties may be related to heterogeneous water flow, as mentioned by the researchers, but may also result from the assumptions of chemical equilibrium and fixed pH and the manner in which the calcite equilibrium is solved. Most previous models assume chemical equilibrium 1634 SOIL SCI. SOC. AM. J., VOL. 61, NOVEMBER-DECEMBER 1997 between the solution and the solid phases and require input of either CO2 or pH with depth, without provision for time dependence or interaction with soil processes. The requirement that CO2 be input leads to serious limitations to the utility of the models, especially in near-surface environments where climatic factors such as rainfall and temperature have an important influence on soil CO2 concentrations. Several regression relations have been developed to predict average CO2 from temperature, rainfall, or evapotranspiration. In addition to these factors, there are other factors that affect soil CO2 concentrations, such^as soil hydraulic properties and porosity (Suarez and Simunek, 1993). Also, average values do not consider the important effect of seasonal and short-term cycles at each site. Buyanovsky and Wagner (1983) measured seasonal changes in soil CO2 concentration of 0.3 to 8% for cropped fields in Missouri. The CO2 model developed by Simunek and Suarez (1993) was able to predict CO2 production, transport, and distribution in the soil based on water inputs, soil hydraulic properties, potential evapotranspiration, and a production submodel. The generally utilized assumption of mineral equilibrium is also only a rough approximation for soil and shallow groundwater environments. Suarez (1977a) measured calcite supersaturation of groundwaters beneath two irrigated regions. In both regions, waters were on average threefold supersaturated. Suarez et al. (1992) also determined calcite supersaturation in soil water sampled from the (water) unsaturated zone. The utility of a kinetic approach to improve prediction of major ion chemistry was shown by Suarez (1985), who presented results of simulations using steady-state unsaturated water flow combined with calcite precipitation kinetics using the Plummer et al. (1978) rate equations. Existing models do not consider the effects of chemical properties on hydraulic conductivity. It is well documented, however, that low electrolyte concentration, high sodicity (McNeal, 1968; Frenkel et al., 1978), and elevated pH (Suarez et al., 1984) all adversely affect hydraulic conductivity. Consideration of these factors is essential to predicting water and solute movement in sodic soils. Our objective was to develop a one-dimensional unsaturated water flow and solute transport model for predicting major ion and B chemistry in field environments, with emphasis on arid-zone chemistry, including the processes of plant water uptake, root and plant growth, and without the need for specifying the CO2 or pH.