A transition state physiochemical model predicting nitrification rates in soil-water systems

Abstract

Transition state theory was applied to the nitrification process in soil-water systems, and a computerized, theoretical rate model was developed to include NH₄⁺ and 0₂ concentrations, pH, temperature, moisture content, and local differences in nitrifying capacities of Nitrosomonas bacteria. The model was restricted to enriched calcareous soils thus simplifying the application of basic physicochemical principles. Experimental rate data from an agricultural and a native desert soil provided verification of a zero order reaction for nitrification with respect to NH₄⁺ concentrations above a certain saturation level, as previously reported. The saturation concentration in soils was found to be about 1.0 to 5.0 ppm. A theoretical linear relationship between activation energy and ionic strength was confirmed by application of the above data. However, each local population of nitrifiers tended to display different values for the slope and intercept of the linear relationship. The structure of the activated complex for NH₄⁺ oxidation to NO₂⁻ was determined to be more like NH2OH or NH₄⁺ than NO₂⁻. As a first approximation, the NH₂OH activated complex was included in the rate model. The equation form for the equilibrium between the reactants and the activated complex was found to differ from the stoichiometric reaction between NH₄⁺ and O₂ to form NH₂OH. The equilibrium expression was found to be more closely approximated by the relationship, 2 NH₄⁺ + O₂ ≶ (ACTIVATED COMPLEX) + + H⁺. A method was developed to compute soil pH values as a function of moisture content. Verification was obtained by using data obtained from the agricultural and native desert soils, including cases where samples were acidified. The calculated pH values were used in the nitrification rate model. Further verification of the model was obtained using data from the literature for two soils from the Northern Great Plains. Data pairing of observed and predicted rates for these soils yielded R values of 0.944 and 0.940. The rate model was programmed in FORTRAN IV computer language and designed to operate in conjunction with existing computer models. Thus, this relatively sophisticated model may be applied to field simulation studies with a minimum of adaptive procedures. The model should aid in obtaining reliable predictions of NO₃⁻ formation and movement under a wide range of field conditions.