Evaluating the effectiveness of the phosphorus sorption index for estimating maximum phosphorus sorption capacity

Authors: Bolster, Carl; MCGRATH, JOSH - University Of Kentucky; ROSSO, EMILEIGH - University Of Maryland; BLOMBACK, KARIN - Swedish University Of Agricultural Sciences

Submitted to: Soil Science Society of America Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/6/2020
Publication Date: 6/2/2020
Citation: Bolster, C.H., Mcgrath, J.M., Rosso, E., Blomback, K. 2020. Evaluating the effectiveness of the phosphorus sorption index for estimating maximum phosphorus sorption capacity. Soil Science Society of America Journal. 84(3):994-1005. https://doi.org/10.1002/saj2.20078.
DOI: https://doi.org/10.1002/saj2.20078

Interpretive Summary: Phosphorus sorption to soils is a primary determinant of both P availability to plants and the amount of P vulnerable to loss through runoff, erosion, and leaching. The sorption of P to soils is often determined in the laboratory using sorption isotherms where P is added to soil samples at various concentrations and allowed to equilibrate with the soil for a specified amount of time. One of the important sorption parameters affecting P loss risk is the maximum P sorption capacity of the soil. As soil P concentration approaches its maximum sorption capacity the soil is less able to sorb P; thus, when P is added to soil, the likelihood that this P is retained in the soil decreases thereby increasing the risk of P loss during runoff or leaching events. While various methods exist for measuring maximum sorption capacity, a commonly used approach is to fit the Langmuir equation to sorption isotherm data. While this approach can provide useful information on the P sorption capacity of soil, it can be a time-consuming approach for quantifying P sorption behavior. An alternative approach is to estimate maximum sorption capacity using the single point P sorption index (PSI). While studies have found good correlations between PSI and P sorption capacity, a thorough analysis of this correlation has not been conducted. Our objective was to further investigate the correlation between PSI and soil P sorption capacity. For our analysis we used both simulated and measured P sorption data. Our results show the potential limitations of using PSI for estimating maximum P sorption capacity of a soil, especially at low concentrations and for soils with low binding constants.

Technical Abstract: One of the important soil parameters affecting both P availability to plants and the amount of P vulnerable to loss through runoff, erosion, and leaching P loss risk is the maximum P sorption capacity of the soil (Smax). Several methods exist for estimating soil Smax, including the use of time-consuming sorption isotherms. Alternatively, the single point P sorption index (PSI) can be used to estimate maximum sorption capacity. While studies have found good correlations between PSI and Smax, a thorough analysis of this correlation has not been conducted. Our objective was to further investigate the correlation between PSI and Smax as determined by fitting the Langmuir equation to P isotherm data. For our analysis we used both simulated and measured P sorption data. Using numerical simulations, we tested the correlation between Smax and PSI using multiple initial and equilibrium P concentrations. Results of the simulations show that while PSI was strongly correlated with fitted Langmuir Smax values, a significant amount of variability exists in these relationships. In particular, the strength of the correlation between Smax and PSI was found to be dependent on the soil P concentration and Langmuir binding coefficient (K) – at low P concentrations and low K values the correlation is weak. Results from correlations between calculated PSI and fitted Smax values for soils collected in Maryland and Sweden confirmed our results from the numerical simulations. Our results show the potential limitations of using PSI for estimating Smax, especially at low concentrations and for soils with low binding constants.