Search

Abstract: In the design of wood-based, enhanced-denitrification bioreactors to treat nitrate in agricultural drainage, the consideration of the highly variable flow rates and nitrate concentrations inherent to many drainage systems is important. For optimized mitigation of these nitrate loads, it may be best to contain drainage water prior to treatment in order to facilitate longer, more constant retention times rather than to allow cycles of flushing and dry periods in the denitrification bioreactor. Simulated containment prior to bioreactor treatment compared to passing drainage directly through a bioreactor was investigated with the use of six pilot-scale denitrification bioreactors constructed with plywood and filled with Pinus radiata woodchips at Massey University No. 4 Dairy Farm (Palmerston North, New Zealand). Initial bromide tracer tests were followed with a series of five simulated drainage events each at successively declining inflow nitrate concentrations. During each drainage event, three pilot bioreactors received a simulated hydrograph lasting 1.5 days (Non-Containment treatment) and three pilot bioreactors received the same total drainage volume treated over 4 days at a constant flow rate (i.e. constant retention time; Containment treatment). Results showed significantly different total mass removal efficiencies of 14.0% vs. 36.9% and significantly different removal rates of 2.1gNm−3 day−1 vs. 6.7gNm−3 day−1 for the Non-Containment and Containment treatments, respectively, which indicated that treating drainage at constant retention times provided more optimized nitrate removal. While this work was done to evaluate treatment under New Zealand drainage conditions, it also provides valuable information for optimizing agricultural drainage denitrification bioreactor performance in general. [Copyright &y& Elsevier]