Deriving sorption indices for the prediction of potential phosphorus loss from calcareous soils

The aim of this study was to develop techniques to evaluate soil phosphorus (P) sorption capacity (PSC) and determine critical soil P levels to predict P loss potential for calcareous soils. Seventy-five soils mostly from Northern China were analyzed for soil P using four extraction methods (water, Pw; carbonate, POls; ammonium oxalate, Pox; and Mehlich 3, PM3) as well as PSC derived from single-point (PSC150) and multipoint sorption (S t) isotherms. Strong correlation was found between PSC150 and S t (r (2)=0.89, p<0.001). The sum of αCaM3 and βMgM3 as an index of PSC (PSC(CaM3 + MgM3)) was most closely related to the maximum amount of P sorbed (S max) as given by the sum of S t and soil initial P setting α=0.039 and β=0.462 (r (2)=0.80, p<0.001). The degree of P saturation (DPS) was thereafter calculated from PSC(CaM3 + MgM3) (DPS(CaM3 + MgM3)), to which Olsen P (POls) was significantly correlated (r (2)=0.82, p<0.001). In a split-line regression from Pw against DPS(CaM3 + MgM3) (r (2)=0.87, p<0.05), a change point was identified at 28.1% DPS(CaM3 + MgM3), which was equivalent to 49.2 mg kg(-1) POls and corresponded to a Pw concentration of 8.8 mg kg(-1). After the change point, a sharp increase in Pw was observed. Our results reveal a new approach to approximating DPS from CaM3 and MgM3 for calcareous soils without the need to generate a S max. We conclude that in the absence of an environmental soil test criteria for P, the DPS(CaM3 + MgM3) and POls could be used to predict P loss potential from calcareous soils.