Denitrifying bioreactors and dissolved phosphorus: Net source or sink?

Understanding the world through a lens of phosphorus (P), as Dr. Andrew Sharpley aimed to do, adds a deeper dimension for water quality work in the heavily tile-drained US Midwest where nitrate is often the nutrient of biggest concern. Denitrifying woodchip bioreactors reduce nitrate pollution in drainage water, but dissolved phosphorus leached from the organic fill is a possible pollution tradeoff. Recent work by Dr. Sharpley and others defined such tradeoffs as strategic decisions in which a negative outcome is accepted with prior knowledge of the risk. In this vein, we assessed 23 site-years from full-size bioreactors in Illinois to determine if bioreactors were a net dissolved reactive phosphorus (DRP) source and, if so, to determine flow-related correlation agents (1904 sample events; 10 bioreactors). DRP was removed across the bioreactors in 15 of 23 site-years. The 23 site-years provided a median annual DRP removal efficiency of 12% and a median annual DRP removal rate of 7.1 mg DRP/m³ bioreactor per day, but the ranges of all removal metrics overlapped zero. The highest daily bioreactor DRP removal rates occurred with high inflow concentrations and under low hydraulic retention times (i.e., under higher loading). Dr. Sharpley was one of the first to explore losses of DRP in subsurface drainage and performed decades of useful applied studies that inspired approaches to management of P loss on both drained and undrained land. We seek to honor this legacy with this practical study of the DRP benefits and tradeoffs of denitrifying bioreactors.