Optimisation of Noosa BNR plant to improve performance and reduce operating costs

Abstract

Noosa WWTP is publicly owned and privately operated by Australian Water Services. The process includes primary sedimentation, raw sludge fermentation, biological nutrient removal (BNR), sand filtration and ultraviolet (UV) disinfection. An innovative feature of the plant is the supplementary carbon dosing facility to avoid the use of metal salts (alum or ferric) for phosphorus removal. The average flow treated during 2000 was 9.0 ML/d. The annual 50 percentile effluent quality requirements for nutrients are total N < 5 mg/L and total P < 1 mg/L. The objectives of this project were to: determine the cause of variability in phosphorus removal; develop a strategy to control the variability in phosphorus removal; and minimise the operating cost of supplementary carbon dosing while achieving the effluent quality requirements. An investigation of chemical and microbiological parameters was implemented and it was concluded that there were several factors causing variability in phosphorus removal, rather than a single cause. The following four major causes were identified, and the control strategies that were adopted resulted in the plant achieving annual 50 percentile effluent total P = 0.37 mg/L and total N = 3.0 mg/L during 2001. First, phosphorus removal was limited by the available VFA supply due to consumption of VFA by other organisms competing with phosphate accumulating organisms (PAO), and due to diurnal variations in the sewage VFA and phosphate concentrations. Therefore, supplementary carbon dosing was essential to make allowance for competing reactions. Second, increasing the fermenter VFA yield via supplementary carbon dosing with molasses was found to be an effective and economic way of ensuring reliable phosphorus removal. Third, nitrate in the RAS resulted in consumption of VFA by denitrifying bacteria, particularly with process configurations where the RAS was recycled directly into the anaerobic zone. Incorporating a RAS denitrification zone into the process rectified this problem. Finally, glycogen accumulating organisms (GAO) were observed in BNR sludge samples, and consumption of VFA by GAO appeared to cause decreased phosphorus removal. Better phosphorus removal was obtained using VFA derived from the fermenter than dosing an equivalent amount of acetic acid. It was hypothesized that GAO have a competitive advantage to use acetate and PAO have a competitive advantage to use propionate, butyrate or some other soluble COD compound in the fermenter effluent. Contrary to popular belief, acetate may not be the optimum VFA for biological phosphorus removal. The competition between PAO and GAO for different VFA species under anaerobic conditions requires further investigation in order to control the growth of GAO and thereby improve reliability of biological phosphorus removal processes.