Nitrate removal and environmental side-effects controlled by hydraulic residence time in woodchip bioreactors treating cold agricultural drainage water

Denitrifying woodchip bioreactors (WBRs) remove nitrate (${\mathrm{NO}}_3^{-}$) from agricultural drainage water at field-scale, but their efficacy at cold temperatures remains uncertain. This study shows how hydraulic residence time (HRT) controls ${\mathrm{NO}}_3^{-}$ removal and environmental side-effects of WBRs at low water temperature under pilot-scale conditions with controlled operation of nine WBRs (94 dm³). Hydraulic properties were assessed by a bromide tracer test, and ${\mathrm{NO}}_3^{-}$ removal, emissions of nitrous oxide (N₂O) and methane (CH₄), and losses of dissolved organic carbon (DOC) were measured at HRTs of 5-30 h. Inlet ${\mathrm{NO}}_3^{-}$ concentrations were increasingly reduced at higher HRTs. The relationship between HRT and the efficiency (%) of ${\mathrm{NO}}_3^{-}$ removal was linear ($R_{adj}^2$ = 0.94), while the relationship between HRT and ${\mathrm{NO}}_3^{-}$ reduction rates (NRR) was logistic ($R_{adj}^2$ = 0.88). Gaseous emissions of N₂O were equally low at HRTs of 10-30 h, but higher at 5 h (P < 0.05). Methane fluxes were small, but with consistent emissions at HRTs of 20-30 h and uptake at 5-15 h. HRT had limited effect on effluent DOC concentrations, but strong effect on mass losses that were five-fold higher (320 mg L^-1) at the HRT of 5 h than at 30 h. In summary, at cold temperatures HRTs of ≤ 20 h resulted in suboptimal NRR, accelerating DOC losses, and increased risk of N₂O losses at least below a threshold HRT of 5-10 h. HRTs of 20-30 h gave maximal NRR, smallest losses of DOC and N₂O, but an increased risk of CH₄ emissions.