Anthropogenic nitrate attenuation versus nitrous oxide release from a woodchip bioreactor

Nitrogen loss via overland flow from agricultural land use is a global threat to waterways. On-farm denitrifying woodchip bioreactors can mitigate NO₃^- exports by increasing denitrification capacity. However, denitrification in sub-optimal conditions releases the greenhouse gas nitrous oxide (N₂O), swapping the pollution from aquatic to atmospheric reservoirs. Here, we assess NO₃^--N removal and N₂O emissions from a new edge-of-field surface-flow bioreactor during ten rain events on intensive farming land. Nitrate removal rates (NRR) varied between 5.4 and 76.2 g NO₃^--N m^-3 wetted woodchip d^-1 with a mean of 30.3 ± 7.3 g NO₃^--N m^-3. The nitrate removal efficiency (NRE) was ∼73% in ideal hydrological conditions and ∼18% in non-ideal conditions. The fraction of NO₃^--N converted to N₂O (rN₂O) in the bioreactor was ∼3.3 fold lower than the expected 0.75% IPCC emission factor. We update the global bioreactor estimated Q₁₀ (NRR increase every 10 °C) from a recent meta-analysis with previously unavailable data to >20 °C, yielding a new global Q₁₀ factor of 3.1. Mean N₂O CO₂-eq emissions (431.9 ± 125.4 g CO₂-eq emissions day^-1) indicate that the bioreactor was not significantly swapping aquatic NO₃^- for N₂O pollution. Our estimated NO₃^--N removal from the bioreactor (9.9 kg NO₃^--N ha^-1 yr^-1) costs US$13.14 per kg NO₃^--N removed and represents ∼30% NO₃^--N removal when incorporating all flow and overflow events. Overall, edge-of-field surface-flow bioreactors seem to be a cost-effective solution to reduce NO₃^--N runoff with minor pollution swapping to N₂O.