Submitted to: Water Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/11/2017
Publication Date: 5/11/2017
Citation: Christianson, L.E., Lepine, C., Sibrell, P.L., Penn, C.J., Summerfelt, S.T. 2017. Denitrifying woodchip bioreactor and phosphorus filter pairing to minimize pollution swapping. Water Research. 121:129-139.

Interpretive Summary: In response to increasing eutrophication from high nitrogen (N) and phosphorus (P) inputs, nutrient loss reduction goals have been established in areas including the Mississippi River basin and Chesapeake Bay watershed. Both point source aquaculture effluents and diffuse agronomy nutrient sources are suitable for innovative dual nutrient removal technologies. Denitrifying bioreactors, a well-established N-mitigation practice, are carbon filled trenches where woodchips fuel heterotrophic denitrification. Phosphorus filters are a new flow-through technology where media remove dissolved P out of solution by adsorption (i.e., surface adhesion) and/or precipitation processes. Lab-scale experimentation demonstrated innovative compatibility by combining these two nutrient removal systems. Comparison between two P removal media types indicated acid mine drainage residuals combined with upstream woodchip denitrification resulted in the optimal design for maximizing P removal while minimizing pollution swapping between systems. Effective paired nutrient removal technologies will help meet established water quality goals in impaired watersheds.

Technical Abstract: Pairing denitrifying woodchip bioreactors and phosphorus-sorbing filters provides a unique, engineered approach for dual nutrient removal from waters impaired with both nitrogen (N) and phosphorus (P). This column study aimed to test placement of two P-filter media (acid mine drainage treatment residuals and steel slag) relative to a denitrifying system to maximize N and P removal and minimize pollution swapping under varying flow conditions (i.e., woodchip column hydraulic retention times (HRTs) of 7.2, 18, and 51 h; P-filter HRTs of 7.6-59 min). Woodchip denitrification columns were placed either upstream or downstream of P-filters filled with either medium. The configuration with woodchip denitrifying systems placed upstream of the P-filters generally provided optimized dissolved P removal efficiencies and removal rates. The P-filters placed upstream of the woodchip columns exhibited better P removal than downstream-placed P-filters only under overly long (i.e., N-limited) retention times when highly reduced effluent exited the woodchip bioreactors. The paired configurations using mine drainage residuals provided significantly greater P removal than the steel slag P-filters (e.g., 25-133 versus 8.8-48 g P removed m-3 filter media d-1, respectively), but there were no significant differences in N removal between treatments (removal rates: 8.0-18 g N removed m-3 woodchips d-1; N removal efficiencies: 18-95% across all HRTs). The range of HRTs tested here resulted in various undesirable pollution swapping by-products from the denitrifying bioreactors: nitrite production when nitrate removal was not complete and sulfate reduction, chemical oxygen demand production and decreased pH during overly long retention times. The downstream P-filter placement provided a polishing step for removal of chemical oxygen demand and nitrite.