Long-term water quality response to increased hydraulic loadings in a field-scale free water surface constructed wetland treating domestic effluent
Danielle J Allen,Mark Farrell,Jianyin Huang,Chris Reynolds,Madhawa Rupasinghe,Luke M Mosley
DOI: https://doi.org/10.1016/j.jenvman.2022.114858
2022-03-11
Abstract:There is limited understanding of how constructed wetland (CW) water quality may change over time in response to increased wastewater nutrient and hydraulic loadings. We evaluated long-term water quality trends and drivers for a full-scale (8.19 ha) free water surface CW that was developed in 2001 for the treatment of increasing amounts of pre-treated domestic wastewater from the township of Mount Barker, South Australia. Water quality parameter concentrations and loads, hydraulic loadings rates, trend direction assessments (TDAs), and water quality parameter removal efficiencies were analysed over the study period. The wetland received an annual average loading rate of 947, 19644, 31039, 18140, 2985, and 807 kg year-1 for BOD5, TN, NH4-N, TKN-N, NOx-N, and TP respectively and removed on average 8%, 72%, 73%, 78%, 12% and -246% of these loadings respectively. The average influent concentrations for the study period were 2.6, 42.3, 40.6, 35.9, 9.0, and 1.9 mg L-1 for BOD5, TN, NH4-N, TKN-N, NOx-N, and TP respectively. Average concentration removal rates over the study period were 50%, 39%, 40%, 15%, -216% and -600.5% for TN, NH4-N, TKN-N, NOx-N, BOD5 and TP respectively, suggesting that nitrogen was only partly assimilated by the wetland and it was a source of organic material and phosphorus. Using seasonally and inflow rate adjusted data, TDAs predicted virtually certain increases in TN, NH4-N, and TKN-N influent concentrations over time, a decline in NOx-N, no trend in BOD5, and a possible decreasing trend in TP. The inflow explained variance accounted for approximately 50% of the variation in TN, NH4-N and TKN-N effluent concentrations. Annual removal efficiencies of N declined with increasing hydraulic loads, and hydraulic loading rates varied with management practices. Seasonal analysis showed that N removal was greater during summer and lower in winter. Due to local population growth and various management practices, hydraulic loading is variable and has often exceeded design targets. Our findings indicate the long-term performance of CWs need to be closely monitored, as water quality can deteriorate due to increased hydraulic loadings.