Nitrogen and phosphorus removal from agricultural drainage water by a modular bioreactor.

Agricultural runoff losses of nitrate-nitrogen (NO3-N) can be treated by woodchip denitrifying bioreactors, but phosphorus (P) removal by this practice has been inconsistent. This study was conducted to test the NO3-N and P removal performance of a novel, modular, portable, downflow bioreactor design. One-meter cubic containers were packed for saturated downflow; influent passed through a layer of P-sorbing media followed by corncobs and woodchips. Three treatments for P removal included steel slag, crushed recycled concrete, and fragmented limestone. Potassium acetate (KCH3CO2) was used as an external carbon source to enhance denitrification. The field experiment continued for two seasons: May-December 2016 and March-July 2017. Mean hydraulic retention time (HRT) was 5.6 h during 2016 and 2.9 h in 2017. The mean annual N removal rate (NRR) among the three P-treatment materials was 2.84 g N m-3 d-1 in 2016 and 1.79 g N m-3 d-1 in 2017. During 2016 and 2017, when acetate was added to the bioreactors, NO3-N load reduction averaged 22.3 % and 14.1 %, respectively. Mean P removal rate (PRR) among the three P-treatment materials was 230 mg P m-3 d-1 in 2016 and 21.9 mg P m-3 d-1 in 2017. During some months bioreactors acted as a P source exhibiting a negative PRR and percent P loss. Overall, P treatment by bioreactors reduced P loss by 28.9 % in 2016 and increased P loss by 14.1 % in 2017 (mean P loss of 12.4 % across years). Results show that this bioreactor design, which may be installed directly under a drainage outlet, can be used as an alternative to classical denitrifying bioreactor beds for subsurface drainage water treatment of N and P, but the P sorbing materials would require removal and replacement on an annual basis.