Input Parameters in Phosphorus Loss Predictive Models As Applied to Spodosol and Ultisol Horizons.

A simple empirical equation for determining the Langmuir strength of P bonding, K_L, the Freundlich adsorption coefficient, K_F, or the linear adsorption coefficient, K_D, obtained from soil analysis would be valuable when such values are needed as input in any mechanistic model for predicting P release from soils on a site-specific basis. Generally, in such predictive models, the linear K_D is an input parameter for soils with low P concentration, whereas the K_L or K_F has been identified as an alternate input parameter for soils with higher solution P concentrations. The objective of this study is to assess P retention and potential transport in two contrasting soil types: the A, E and Bt horizons of Ultisols and the A, E, and Bh horizons of Spodosols. The compositional differences between soils as well as the horizons within the same soils are responsible for the mode of P transport in these soils. Isotherm parameters such as K_d, K_L and K_F as well as the equilibrium P concentration (EPC₀) and the P in the solid phase (S₀) are related to soil components such as Ca, Mg, Fe and Al. Backward stepwise and multiple regression analyses showed that isotherm parameters (K_L, K_F or K_D) are dependent on the threshold P saturation ratio (PSR) for all the three horizons in two soils. The “K” values are variable below the threshold PSR, i.e., the molar ratio of P to [Fe+Al] at which P release from the soil increases abruptly, but tend to zero as the threshold is reached. The “K” values are also related to the soil P storage capacity (SPSC), a parameter calculated based on the threshold PSR, and which reflects the additional amount of P that can be stored prior to the soil being an environmental P risk. Magnesium concentrations are related to P release beyond the threshold PSR. The S₀ of both soil types are related to Mg and the SPSC of the soils. Equations developed relating isotherm parameters to the PSR and SPSC are independent of the soil composition, suggesting that such equations might be applicable across a wide range of soils.