Use of duckweed-based constructed wetlands for nutrient recovery and pollutant reduction from dairy wastewater

•Duckweed wetlands reduce pollutants in dairy waste while producing valuable biomass.•First-order removal to a background best described treatment in duckweed wetlands.•Duckweed wetlands are best suited for primary treatment of dairy wastewaters.•Optimal nitrogen for duckweed growth higher for dairy waste than synthetic waste.•E. coli reduction from dairy wastewater by duckweed wetlands quantified. Over the last few decades, constructed wetlands have increasingly been designed and implemented to treat agricultural wastewaters. However, while treatment of manure-containing wastewaters protects water quality, it also eliminates a valuable source of nutrient-laden soil amendments. The goal of this research is to investigate the ability of wetlands to simultaneously treat high-strength manure-containing wastewaters while recovering nutrients from the manures in plant biomass. Diluted raw dairy waste was fed to a combination of duckweed-based surface flow and subsurface flow wetlands. Nutrient recovery through duckweed harvesting and waste treatment, characterized by chemical oxygen demand (COD), total nitrogen (TN), total phosphorus (TP) and E. coli removal, were assessed under steady-state conditions for three strengths of waste. Duckweed-based wetlands used for primary treatment had higher removal rates of COD and TN than those used for secondary treatment; however, no significant difference was observed for TP removal rates. COD removal ranged from 3 to 81% in primary duckweed wetlands and from −35% to 38% in secondary duckweed wetlands. Areal removal rates for nutrients in primary duckweed wetlands were 194.9±18.9gTN/m2/yr and 13.0±3.0gTP/m2/yr, while removal rates in secondary duckweed wetlands were 104.1±13.1gTN/m2/yr and 9.3±2.1gTP/m2/yr. Removal of COD, TN, and TP more closely followed first-order removal to a background concentration than first-order removal or a previously published model for duckweed ponds. Mean log E. coli reduction of 0.30 obtained in this experiment was within the range reported in literature. Duckweed production in the wetlands was satisfactorily described as a second-order polynomial function of influent TN concentration. More N and P was recovered from primary wetlands as compared to secondary wetlands. Average N and P recovered by harvesting duckweed across all the wetlands were 22.4gN/m2/yr and 7.4P/m2/yr, respectively.