Identifying dissolved reactive phosphorus sources in agricultural runoff and leachate using phosphate oxygen isotopes

Agricultural phosphorus (P) losses may result from either recently applied fertilizers or from P accumulated in soil and sediment. While both P sources pose an environmental risk to freshwater systems, differentiating between sources is crucial for identifying and implementing management practices to decrease loss. In this study, laboratory rainfall simulations were completed on runoff boxes and undisturbed soil columns before and after fertilizer application. The oxygen-18 signature of phosphate (δ¹⁸O_PO4) in fertilizer, surface runoff, subsurface leachate, and soil were analyzed (n = 107 samples) to quantify new (recently applied) and old (soil) P losses in runoff and leachate. Results showed that dissolved reactive P (DRP) concentration in runoff and leachate substantially increased during the rainfall simulation immediately after fertilizer application, with runoff and leachate δ¹⁸O_PO4 similar to fertilizer δ¹⁸O_PO4 signatures. Greater than 90 % of the DRP load during this event could be attributed to direct loss of P from fertilizer using δ¹⁸O_PO4. Beyond the first rainfall event after fertilizer application, DRP concentration decreased and leachate δ¹⁸O_PO4 values differed from the fertilizer values. Interpretation of isotope results was challenging because both abiotic (isotope fractionation during transport) and biotic (P cycling) processes may have influenced δ¹⁸O_PO4 signatures during these subsequent events. While abiotic effects on δ¹⁸O_PO4 appear more probable given the experimental conditions in the current study (high soil test P concentration, short duration between rainfall simulations, and strong relationship between event water and δ¹⁸O_PO4 signature), tracing or separating P sources remains highly uncertain during these events post-fertilizer application. Findings highlight both potential opportunities and challenges of using δ¹⁸O_PO4 to trace sources of P through the landscape.