A Simplified Soil and Plant Phosphorus Model: I. Documentation

Abstract

The structure of a simple soil and plant P model designed for use in the Erosion‐Productivity Impact Calculator (EPIC) agriculture management model is described. The model maintains pools of stable, active, and labile inorganic P; fresh organic and stable organic P; and grain, stover and root P. The model has a daily time step, simulates P uptake and transformations in up to ten soil layers of variable thickness, and is sensitive to soil chemical and physical properties, crop P requirements, tillage practice, fertilizer rate, soil temperature, and soil water content. Fertilizer P is added to the labile inorganic P pool which is in rapid equilibrium with the active inorganic P pool. The relative sizes of the labile and active inorganic pools are soil specific and are based on soil classification, texture, and chemical properties. Movement of active inorganic P to the stable inorganic P pool simulates slow adsorption of inorganic P. Crop P uptake from a soil layer is sensitive to crop P demand and the amounts of labile P, soil water, and roots in the layer. Stover and root P are added to the fresh organic P pool upon their death and/or incorporation into the soil. Decomposition of fresh and stable organic matter may result in net immobilization of labile P or net mineralization of organic P.

Obtaining the parameters required to model labile phosphorus for South African soils

“…This work can potentially be adapted for local modelling purposes. Local research, similar to the work done by Jones et al (1984) is ultimately required to develop P modelling algorithms more suited to South African soils.…”

“…ACRU-NP and SWAT have simple crop routines and were developed to be run at the catchment scale, while SWB-Sci and APSIM were developed to be run on the field scale and are more reflective of management practice interventions. The P modelling routines of all four these models can be traced back to work done by Jones et al (1984) and Sharpley et al (1984) to develop the model EPIC (Erosion Productivity Impact Calculator) (Williams et al, 1983).…”

“…Originally, Jones et al (1984) and Sharpley et al (1984) used 78 continental USA and Puerto Rican soils to develop their plant and soil P model. Calcareous and non-calcareous soils which have undergone different degrees of weathering can be expected to undergo greatly differing soil-P reactions (Sharpley et al, 1989), and Sharpley et al (1984) observed that the most accurate estimation of Labile P, was achieved when soils were divided into calcareous, slightly weathered or highly weathered groups based on the presence of CaCO 3 and degree of weathering.…”

“…Sharpley et al (1984) originally used weathering and soil taxonomic information to group soils, and although the United States Department of Agriculture mostly uses subsoil parameters to determine classification, for South African soils we suggest that the properties of the top horizon only should be considered for categorization as this is the diagnostic horizon used in the South African Classification system (Soil Classification Working Group, 1991). Furthermore, only surface samples (0-10 cm) were used by Jones et al (1984) and Sharpley et al (1984) to develop the various algorithms used.…”

“…According to the approach of Jones et al (1984), the initial size of the Active P pool is calculated using a P Availability Index (PAI), with equation (7): (7) ACRU-NP and SWB-Sci are also able to estimate the size of the Active and Stable P pools by subtracting organic P and Labile P from total soil P, if these values have been provided by the user. Initial Stable P is assumed to be four times larger than Active P.…”