Nutrient Removal from Agricultural Subsurface Drainage Using Denitrification Bioreactors and Phosphate Adsorbents

Agricultural subsurface drainage is widely used to remove excess water from the soil. Nitrate and phosphorus contained in subsurface drainage can negatively affect water quality. Denitrification bioreactors have been used to remove nitrate from subsurface drainage. However, bioreactors are not designed to remove phosphorus. Steel byproducts and other materials have been used as adsorption media to remove phosphorus from water. The objective of this project is to develop a new woodchip bioreactor system using steel media as post-treatment to simultaneously remove nitrate and phosphate from simulated agricultural subsurface drainage. Short-term batch and long-term column experiments are conducted to evaluate the nitrate and phosphate removal efficiency of the proposed bioreactor system. Batch experiments were conducted to determine the phosphorus removal capacity of selected steel byproducts. Maximum adsorption capacities for fine steel chips, carbon steel medium chips, small and large turnings were 5.81, 2.00, 1.47, and 1.30 mg phosphorus/g steel, respectively. Based on phosphate removal efficiency and particle size, medium carbon steel chips were selected for column experiments. Two column reactors were built to simulate a woodchip bioreactor followed by the steel media as post-treatment. With influent concentrations of 20 mg N/L and 50 mg N/L, the woodchips achieved denitrification rates of 10.0 and 14.4 g N/m3/d. Reducing the HRT to 12 and 6 hours with an influent nitrate concentration of 20 mg N/L, a nitrate removal of 92% and 54% was observed through the woodchips, which yielded denitrification rates of 15.1 and 15.4 g N/m3/d, respectively. The woodchips achieved phosphorus removal rates of 0.28, 0.56, and 0.88 g P/m3/d with influent concentrations of 1 mg P/L for hydraulic retention times of 24, 12, and 6 hours, and 1.0 g P/m3/d with an influent concentration of 10 mg P/L. With influent concentrations of 1 mg P/L, the steel media removed 100% of the phosphorus. With an influent concentration of 10 mg P/L, the steel media had 100% removal for 14 days followed by 76% removal from days 32 to 130. A decrease in nitrate removal efficiency and an accumulation of nitrite was observed under consecutive short wet and dry cycles. The steel media completely removed any phosphorus entering from the woodchips during the wet and dry cycles. A breakthrough test found the adsorption capacity of the steel media used in column experiments to be 3.12 mg phosphorus/g steel. The results of this study suggest steel byproducts can be used in conjunction with woodchip bioreactors to remove nitrate and phosphorus from subsurface drainage. This new technology has great potential to reduce the nutrient loading from agricultural subsurface drainage to surface water bodies