Denitrifying bioreactor response during storm events

Abstract

Denitrifying bioreactors can be an effective management strategy to address excess nitrate from tile-drained fields. However, there has been minimal research on how bioreactors function during storm events, which can account for large portions of annual nitrate load. Storm events cause reduced hydraulic retention time and potentially temperature and flow path fluctuations that may disturb denitrifying microbial communities within the bioreactor. In this study, nitrate concentration and flow rate were monitored during baseflow and storm-induced flow in four denitrifying bioreactors treating agricultural tile drainage to compare removal rate (RR) and removal efficiency (RE) between flow conditions. Removal rate was higher in stormflow than baseflow for all the bioreactors, though this was only significant for two bioreactors. However, RR was lower on average following events compared with pre-storm rates. Lag time, the period between a parcel of water entering the bioreactor and exiting, ranged between 2 and 24 h for the observed storms. Calculations of RR and RE that account for lag time were significantly lower than standard calculations. The model of RR that included a variable of flow variation (coefficient of variation in flow over a 24-hour period) was most effective at describing observed data. This, along with differences in RR between baseflow and stormflow, suggests that bioreactor function in storms should be considered separately from baseflow. Decreased RR following storms also suggests that bioreactors may require additional design considerations to reduce disturbance, especially for fields where much of the annual nitrate runoff load is concentrated in storm-induced events.