Prediction of phosphorus fractions in soils

Abstract

Abstract Organic and inorganic phosphorus fractions were determined in 20 soil samples representing different soil parent materials and cropping managements. The distribution of phosphorus (P) across different fractions and their relationships to soil chemical and physical data were used to develop models capable of predicting P forms. Correlation and multiple regression analyses of P fractions and chemical analyses confirmed that total inorganic P (TIP), occluded inorganic P (OCIP), non‐occluded inorganic P (NOCIP), and Ca‐bound inorganic P (Ca‐BIP) can be relatively accurately predicted from soluble CO2‐ 3 and HCO‐ 3while organic P (OP) can be predicted from soluble CO2‐ 3 and organic matter (OM). Organic matter and electrical conductivity (EC) explained 77% of the variability in the agronmically important P form (loosely bound inorganic P (LBIP)). In general, the predictive models illustrated in this article are useful in estimating P forms from data available in routine soil analysis.