Comparing measures for determination of phosphorus saturation as a method to estimate dissolved P in soil solution

Karin Blombäck,Carl H Bolster,Anders Lindsjö,Kathrin Hesse,Helena Linefur,Mohammed Masud Parvage

doi:10.1016/j.geoderma.2020.114708

Karin Blombäck, Carl H Bolster + Show 4 more

Open Access

https://doi.org/10.1016/j.geoderma.2020.114708

Copy DOI

Abstract

In response to concerns over the translocation of P from soils to P-sensitive water bodies, there is high demand for developing simple indicators for evaluating a soil’s risk of releasing P into solution. Many studies have shown that the degree of soil phosphorus saturation (DPS), calculated as the ratio of soil P concentration to soil P sorption capacity (PSC), is good predictor of a soil’s risk of releasing P in solution. In this study we compared four different DPS indices in how well they predicted dissolved P following extraction with either a 0.01 M CaCl2 (PCaCl2) solution or deionized water (PW). The first two indices were calculated from the ratio of extractable P to extractable Al and Fe using either acid ammonium oxalate (AlOX + FeOX) or ammonium lactate solutions (AlAL + FeAL). The second two DPS indices were calculated from the ratio of either Olsen-extractable P or AL-extractable P with sorption capacity estimated from the single point P sorption index (PSI). On a subset of 11 soils, we compared the different methods for estimating PSC with fitted Langmuir sorption maximum (Smax) using data from complete sorption isotherms. Both (AlOX + FeOX) and PSI were well correlated with Smax and hence regarded as good estimates for P sorption capacity. Conversely, (AlAL + FeAL) was not significantly correlated with Smax. P saturation calculated from PSI together with PAL or POls predicted PCaCl2 and PW best, whereas P saturation calculated from ammonium oxalate predicted PCaCl2 and PW the least. We did not find notable improvements in the regression models when we added a second explanatory variable (clay content, pH or total carbon) to the models. Our results show that multiple measures of P saturation provide similar predictions of a soiĺs potential for releasing dissolved P into soil solution. This provides flexibility in how P saturation indices are calculated to identify leaching prone hotspots.

Highlights

Eutrophication caused by phosphorus (P) losses from land to surface waters is a worldwide problem (Sharpley et al, 2015)
Since P sorption index (PSI) was very well correlated with both sorption maximum (Smax) (r = 0.91) and the sum of AlOX and FeOX (r = 0,67), our results indicate that PSI maybe could be a more suitable predictor for the P sorption capacity
Our results suggest that multiple degree of soil phosphorus saturation (DPS) metrics provide similar predictions for evaluating a soil’s potential for releasing dissolved P into solution

Summary

Introduction

Eutrophication caused by phosphorus (P) losses from land to surface waters is a worldwide problem (Sharpley et al, 2015). Data on leaching losses of P from more than 30 observational sites in Sweden (from catchments and drained fields) show that the ratio of the load of dissolved P to total P in drain water (kg ha−1 year−1) ranges between 0 and 100%, with a mean of approximately 30% (Linefur et al, 2018a,b). The degree of phosphorus saturation (DPS) takes into account both the soil P concentration and the P sorption capacity (PSC) of the soil and has been shown to improve predictions of a soil’s risk of releasing P in solution during runoff or leaching events compared to soil P test (Sharpley, 1995; Maguire and Sims, 2002b; McDowell et al, 2002; Nair et al, 2004; Nair, 2014; Wang et al, 2012; Kleinman, 2017)

Objectives

Methods

Discussion

Conclusion