Multiyear changes in snowmelt phosphorus runoff with soil P drawdown or application of struvite to an organic forage crop

In this study, changes in P runoff losses from three field-scale small watersheds in the Northern Great Plains region (Manitoba, Canada) under organic management were evaluated over a 7-year period (2017-2023). A perennial forage crop was grown through this time period, either with biennial addition of a slow-release fertilizer to maintain soil fertility (struvite, 4 site-years) or without P amendment (drawdown, 8 site-years). Annually measured soil P (Olsen P) strongly correlated with mass balance of P added as struvite or removed by crops and surface runoff. As a result, P limitation of forage yield was observed in the later years of the drawdown treatment. P loads and flow-weighted mean concentrations (FWMCs) measured in snowmelt runoff after struvite application were compared to predicted values using published regional models developed with 82 site-years of data from field-scale small watersheds, with Olsen P, water yield, and percent surface cover as predictors. To evaluate response to struvite over a wider range of conditions, P runoff was measured in two additional watersheds under conventional annual grain production following struvite application (2 site-years). Higher risk of P loss was observed with higher Olsen P. However, following struvite applications, the risk of P loss to snowmelt was consistently lower than predicted (between -0.13 and -0.85 mg L^-1 for FWMC of total dissolved P) suggesting that the solubility of struvite in runoff water may be lower than for other forms of soil P also extractable as Olsen P.