Effects of hydraulic retention time, soluble carbon source and substrate media on nitrate removal efficiency of column denitrification bioreactors : a thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Agricultural Science at Massey University, Palmerston North, New Zealand
Loading...
Date
2024
DOI
Open Access Location
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Massey University
Rights
The author
Abstract
In New Zealand, dairy farming is a major contributor to nitrogen (N) contamination of the aquatic environment, due mainly to nitrate leaching from cow urine patches and intensive cropping. This enrichment of the aquatic environment affects groundwater and surface water quality and their dependent freshwater ecosystems by promoting eutrophication. It is therefore important to better understand and mitigate critical flows of nutrient losses from agricultural lands to receiving waterways. In response to this environmental challenge, denitrifying bioreactors have emerged a novel edge-of-field practice, which can reduce nitrate levels in agricultural drainage waters by facilitating the conversion of dissolved nitrate to nitrogen gases using a carbon source in microbial denitrification process under anaerobic conditions. Woodchips, a common carbon media of bioreactors, are favoured for their simplicity, cost-effectiveness, and efficiency in nitrate removal. Their use is growing worldwide, particularly at the edges of agricultural fields to minimise farming disruption and effectively mitigate nitrate losses in drainage waters. However, as woodchips age, the availability of dissolved organic carbon (DOC) decreases, potentially limiting denitrification rates over time in woodchip bioreactor. However, soluble carbon dosing is proposed as a solution to enhance nitrate removal by maintaining a consistent supply of organic carbon. Although carbon dosing has predominantly been studied in woodchip bioreactors, it has also shown success in non-woodchip bioreactors that utilise a range of substrates including soil, sediment, sand, pumice stone, and vermiculite, supplemented with ethanol and methanol as soluble carbon sources. These systems have proven effective in reducing nitrate levels in aquaculture and underground water systems. However, the use of ethanol as an external carbon source in non-woodchip bioreactors for agricultural drainage systems has not been well researched. This study investigated the use of soluble carbon dosing in denitrifying bioreactor to enhance nitrate removal, comparing its effectiveness across woodchip, pumice, and sedimentary rock gravel a substrate media. The study used a set of nine (9) small-scale replicated column bioreactors. The specific research objectives were to investigate the effect of varying hydraulic retention time (HRT), soluble carbon source (methanol and ethanol) and compare different low-cost bioreactor media (woodchips, sedimentary gravel, and pumice gravel). The first experiment assessed the impact of different HRTs on the nitrate removal performance in woodchip column bioreactors. Three HRTs (6.6, 10, and 20 hours) and average inflow nitrate concentration of 19.6 mg N L -¹ were used to measure effects of HRT’s on nitrate removal efficiency (% reduction in nitrate concentration) and nitrate removal rate (quantified as g NO₃ - -N removed per m³ of woodchips per day). The inflow water temperature varied from 19.6 ᵒC to 20.4ᵒC, with an average of 20oC. However, extending the HRT from 6.6 to 20 hours, increased nitrate removal efficiency from an average of 35% to 71%, but decreased nitrate removal rate from an average 13.6 to 9.4 g NO₃ - -N m-³ day-¹, respectively. The second set of experiments (two) evaluated the effects of methanol and ethanol dosing on nitrate removal efficiency in woodchip column bioreactors, operated with an HRT of 6.6 hours, an average inflow water nitrate concentration at 19.5 mg N L -¹, and average water temperature of 16.4 ᵒC from both experimental sets. A soluble carbon dosing at a C:N ratio of 1:1 significantly enhanced nitrate removal efficiency. The ethanol-dosed treatment achieved a nitrate removal efficiency of 66-68%, compared to 55-57% for the methanol-dosed treatment and 33-38% for the control (non-dosed) treatment. In the third experiment, the efficacy of ethanol dosing for nitrate removal was assessed across different column bioreactor media (woodchips, pumice, and sedimentary rock gravel). The experiment used ethanol dosing at a 1:1 C:N ratio, a 6.6-hour HRT, an average inflow water nitrate concentration at 19.6 mg N L-¹, and average water temperature of 13.4 ᵒC. The ethanol dosing was effective particularly achieving an average of 97% reduction in the inflow NO₃⁻ concentration in the woodchip bioreactors columns, compared to 72% for pumice and 75% for gravel used as substrate media. However, the increased outflow total organic carbon levels in woodchip bioreactors suggest a higher release of organic carbon from both fresh woodchips and added ethanol, providing a higher carbon source for enhanced denitrification in the woodchip bioreactor columns. This was also evidenced by consistently low outflow NO₃⁻-N concentrations in the ethanol-dosed woodchip bioreactor columns. The bioreactor columns experiment results in terms of measured nitrate removal efficiency were finally applied to construct a comparative economic-opportunity cost analysis of ethanol dosed (C:N ratio of 1:1) woodchip and gravel bioreactors (assuming 200 m³ bioreactor size) for informing further development of practical cost-effective bioreactor design for agricultural drains. A lifespan of either 10 or 15 years was sued for the woodchip bioreactor and of 30 years for the gravel bioreactor. When both media (woodchips and gravel) were assumed to have a porosity of 50%, the cost-effectiveness of nitrate removal using woodchip bioreactor was calculated at NZ $5 .90 per kg N removed for a 10-year lifespan scenario, and NZ $4 .30 per kg N removed for 15 years lifespan scenario, while for the gravel bioreactor it was NZ $8 .80 per kg N removed for 30 years lifespan scenario. A gravel bioreactor, despite its higher annual costs, due to their longer lifespan without the need for media replacement offers practical advantage but requires active management of external carbon-dosing system. In summary, this thesis has developed a comprehensive dataset that enhances our understanding the potential of ethanol dosing to improve the performance of denitrification bioreactors, while also exploring alternative media such as gravel and pumice compared to woodchips. The accompanying cost analysis reveals that although woodchip bioreactor appears more cost-effective in the short term, due to lower initial and maintenance costs, gravel bioreactor may have potential to be cost effective and practical in the long-term. However, longer term field testing is needed to assess the actual relative performance of these two media over time to better assess their cost-benefits and practicality in their potential applications in real-world conditions.