Comparison of the wastewater treatment performance of continuously and discontinuously mixed high-rate algal ponds at Kingston on Murray Open Access

High-rate algal ponds (HRAPs) incorporate in their design both a shallow raceway and a paddlewheel that continuously mixes the wastewater to both eliminate thermal stratification and create a homogenous environment (Buchanan et al. 2018). HRAPs can either be single-looped raceway or adopt a serpentine configuration. This design promotes organic waste and nutrient removal through the promotion of algal and bacterial growth. The shallow design and use of continuous mixing allow for the pathogens to be inactivated through the disinfection by sunlight (Buchanan et al. 2018). The design, operation and performance of HRAPs for wastewater were recently reviewed by Young et al. (2017).

In South Australia, community wastewater management schemes (CWMS) are implemented when both environmental and public health issues occur in relation to disposal of effluent from septic tanks. In 2008, the first HRAP was built at Kingston on Murray (Plate 1) (Fallowfield et al. 2018). There are two ponds located at Kingston on Murray each of 250 m², which treat effluent from a population of 300 people at a rate of 12 m³ day⁻¹. The design of the HRAP at Kingston on Murray and its success brought about the design guidelines for the construction and operation HRAPs within South Australia (SA LGA 2020).

An issue that occurs with any operation of an HRAP is the risk of a failure of the paddlewheel, which continuously mixes the wastewater. Cessation of mixing of an HRAP can occur due to, mechanical failure, power grid failure or, for off-grid solar photovoltaic systems, when there are continuous cloudy days and insufficient sunlight. Nutrients are removed from the wastewater by algal production, which may be limited by the assimilation of nutrients into the algal cell from the wastewater (Shariff & Chakraborty 2017). The removal of nutrients from wastewater is reliant on the factors that determine the algal growth within the system. Such factors include temperature, light intensity, pond depth and the retention time which is typically 10 days in an HRAP (Evans et al. 2005). The process of continuous mixing enhances nutrient uptake by reducing the boundary layer around the algal cell. Mixing, especially continuously, increases cell light exposure, promoting algal photosynthesis (Ali et al. 2018). At high surface light intensities, mixing may also reduce the chance of photoinhibition occurring (Mehrabadi et al. 2015), potentially inducing dark recovery by enhancing light: dark cycling (Ratchford & Fallowfield 2003). Furthermore, in an unmixed environment, algae with a similar density to water may become more buoyant resulting in the algae remaining at the surface, experiencing photoinhibition, attenuating the light and reducing penetration through depth (Bosca et al. 1991). The loss of the homogenous environment on cessation of mixing also creates thermal stratification of the wastewater, which has been shown to impact the assimilation of nutrients.

A component of the treatment process within HRAPs is the exposure of pathogenic microorganisms to solar ultraviolet light (UVA and UVB), which plays a crucial role in the disinfection of wastewater. The faecal indicator organisms, E. coli and FRNA coliphage are often used as surrogates to determine potential pathogen inactivation rates of bacteria and viruses within the wastewater. The disinfecting UVB and UVA wavelengths are rapidly attenuated through depth in turbid wastewaters. Paddlewheel mixing in the HRAP circulates the wastewater increasing exposure to UVB and UVA (Buchanan et al. 2018). This influences the log₁₀ reduction value (LRV), the difference in concentration (log₁₀ 100 mL⁻¹) between wastewater inlet and outlet values of the surrogate organisms, E. coli or FRNA ‘phage, or pathogens of concern (Young et al. 2016).

The HRAP guidelines require continuous mixing of the wastewater (SA LGA 2020). Adoption of the technology in rural and remote communities, where connection to the grid is expensive or impossible will require adoption of solar photovoltaic systems with battery storage. Cloudy days, when there is insufficient light available both to run the paddlewheel and charge the batteries would result in cessation of mixing. This infringes the operating guideline. Installation of back-up diesel generators is an often-suggested solution, which increases capital and operating costs providing a disincentive to adoption of HRAP technology.

There are currently limited studies to investigate the effect of cessation of mixing on HRAP treatment performance or wastewater disinfection. The research reported here compares the wastewater treatment performance of two HRAPs, one mixed continuously and the second, where mixing was ceased for one theoretical hydraulic retention time (THRT, 10 days) over a 90-day period.

The research was performed using 2, 250 m² HRAPs each operated at 0.3 m depth at Kingston on Murray, located in South Australia (34.242816 ° S, 140.330197 ° E). The incoming wastewater (12 m³ day⁻¹), pre-treated in on-site septic tanks with hydraulic retention time (HRT) of 24 h, was pumped in six aliquots per day each of approximately 2 m³ and divided equally between the two HRAPs. Each pond had an eight bladed stainless-steel paddlewheel capable of providing a mean surface water velocity of 0.2 ms⁻¹. One HRAP was discontinuously mixed, where the paddlewheel was operated for a minimum of 10 days, then mixing was ceased for 10 days, equivalent to one HRT, followed by a further minimum mixing period of 10 days before the 30 day cycle was repeated. The second HRAP was continuously mixed. The treatment performance of the two HRAPs was compared in triplicate experiments.

Wastewater sampling began on the 12th of January 2021 and finished on the 12th of April, 91 days.

Wastewater samples (10 mL) were filtered through a 90 mm GF/C Whatman Filter and stored at 20 °C for subsequent analysis. Nutrient analysis (NH₄-N, NO₂-N, NO₃-N, PO₄-P) was conducted using a Skalar San + +analyser and American Public Health Association's Standard Methods for the Analysis of Water and Wastewater.

The BOD₅ was determined, following incubation in the dark at 20 °C for 5 days using the WTW Oxitop^® (Merck, Germany) system, according to the instructions of the manufacturer.

Wastewater samples (10 mL) was filtered through a Whatman 90 mm GFC filter pad which was pre-dried at 105 °C for 24 h and weighed. The filtered pads were then placed in the oven at 105 °C for 24 h (105 °C and was weighed. The suspended solids concentration (mg L⁻¹) was calculated from the difference between the final and initial filter weights.

Following wastewater filtration through a 90 mm GF/C Whatman Filter the chlorophyll a on the filter was extracted overnight into 90% (v/v) acetone: water in the dark at 4 °C and determined spectrophotometrically.

E. coli was quantified using a single Coliert Quanti-tray most probable number (MPN) method, (IDEXX Laboratories Westbrook, ME, USA) according to the manufacturer's instructions. The concentration of E. coli values were recorded as the MPN 100 mL⁻¹.

The independent samples T-test for both the equality of mean and Levene's test for the equality of variances were determined using SPSS statistical software. Statistical significance accepted at P = <0.05.

The composition of the influent wastewater from the on-site septic tanks is shown in Table 1. As expected, NH₄-N predominated as the source of nitrogen.

The temporal change in NH₄-N concentrations in the continuously and discontinuously mixed HRAPs for all three experiments is shown in Figure 2. Frequently, the discontinuously mixed pond had higher NH₄-N concentrations in the effluent. However, only Experiment 2 conducted in February–March showed statistically significant difference (P < 0.05) in NH₄-N concentration, where the concentration was significantly higher in the discontinuously mixed HRAP (Table 2).

The temporal changes in PO₄-P concentrations in the HRAP-treated effluent for the continuously and discontinuously mixed HRAPs are shown in Figure 2. Generally, the concentration was higher in the effluent from the discontinuously mixed HRAP. The difference in concentration, however, was statistically significant (P < 0.05) only in Experiments 2 and 3 (Table 2).

The NO₂-N concentrations were always higher in the discontinuously mixed pond (Figure 2). The difference between the two ponds was statistically significant within all experiments (P < 0.05; Table 2). Similarly, the NO₃-N concentration in the discontinuously mixed HRAP effluent was significantly greater than that of the continuously mixed HRAP (Figure 2, P < 0.05; Table 2) for all three experiments.

The discontinuously mixed pond showed cycling of the nitrification and denitrification processes.

With nitrification taking place during the ‘on’ phases of the paddlewheel and denitrification taking place in the ‘off’ phases of the paddlewheel. In most aquatic environments, there exists a relationship between nitrification and denitrification processes (Xia et al. 2017). The denitrifying bacteria rely on nitrate from external sources (Gao et al. 2012). Algal blooms in water bodies are followed by the decay of the algal biomass, which then provides organic carbon required by the heterotrophic denitrifying bacteria. The decay also decreases the dissolved oxygen concentration in the sediments, thus creating the anaerobic environment needed for denitrification to occur. When the paddlewheel is turned off the absence of mixing would likely decrease the rate at which the atmospheric diffusion of oxygen takes place. This decrease in diffused oxygen would create a favourable environment for denitrification to take place. It was demonstrated that large abundances of algal biomass significantly impacted the symbiotic relationship of denitrification and nitrification processes (Zhu et al. 2020). Interestingly, it was also noted that when the anoxic environment, caused by the algal biomass ceased, there was a decrease in ammonia and a rise in nitrate concentrations. This was presumed to be a result of the aerobic conditions within the system being a more favourable environment for nitrification processes to occur (Zhu et al. 2020). Within the discontinuously mixed HRAP turning off the paddlewheel, reduced atmospheric oxygen diffusion into the wastewater caused the algal biomass to settle, reducing photosynthetic oxygen production and likely causing the algae to degrade. These conditions favoured the reduction of nitrate via denitrification. After 10 days, when mixing resumed, atmospheric oxygen diffusion increased, the algal biomass was resuspended and oxygen was produced via photosynthesis, promoting nitrification as demonstrated by the increase in nitrate.

These results suggest that the rate of nitrification was higher in the discontinuously mixed HRAP. Nitrifying populations attached to particles may sediment when mixing ceases causing greater retention within the HRAP, the higher biomass increasing nitrification in the discontinuously mixed HRAP. Nutrient removal is important in the process of wastewater treatment as ammonia can be toxic to plant life and animals. High levels of nitrate can lead to eutrophication (Sutherland et al. 2014). However, no guidelines exist on the maximum level of nitrate levels acceptable for land-based discharge.

The TSS and the chlorophyll a concentration are shown in Table 3. There was no significant difference in biosolids (TSS) or algal production between the continuously mixed and discontinuously mixed HRAPs, except for the chlorophyll a concentration in Experiment 3 in March–April, which was significantly higher (P < 0.05) in the discontinuously mixed HRAP (Table 3). Algal biomass has been observed to spontaneously create flocs and settle in a mixed environment (Garcia & Hernandez-Marine 2000). The lack of mixing in the discontinuously mixed pond promoted sedimentation. The chlorophyll concentrations, however, were found to be higher in the discontinuously mixed pond. This result was not expected as cessation of mixing would be expected to reduce algal photosynthesis by limiting access to light. Alternatively, settling of solids may reduce light attenuation throughout the depth of the pond enabling the settled algal biomass to continue to grow.

BOD₅ is an important parameter in the determination of wastewater treatment performance. It represents the amount of dissolved oxygen consumed by aerobic microorganisms. Comparison of the discontinuously and continuously mixed HRAPs showed there were no significant differences (P > 0.05) in the BOD₅ content of the treated effluent between the continuously and discontinuously mixed HRAPs (Table 3). Notwithstanding, the effluent from discontinuously mixed HRAP in Experiment 1 and 2, however, marginally exceeded the SA LGA (2020) guideline value of <20 mg BOD₅ L⁻¹.

The comparison of the discontinuously mixed and continuously mixed ponds included an assessment of faecal contamination using E. coli as an indicator of potential bacterial pathogen removal. Table 3 shows the changes in E.coli count (MPN 100 mL⁻¹) for the duration of all three experiments. The E. coli count was significantly different (P < 0.05) between the continuously and discontinuously mixed HRAP in Experiment 1 with the continuously mixed HRAP recording a higher mean concentration in the effluent (Table 3). Interestingly, there were no statistically significant differences in the E. coli LRVs between ponds over the whole experimental period (Table 3). Importantly, however, the treated effluent from both HRAPs complied with the SA LGA (2020) effluent guideline of <10⁴E. coli MPN 100 mL⁻¹.

The mean LRVs showed no significant difference over each of the three 30-day trials. This indicated that intermittent mixing did not have a significant impact on the removal of E. coli. The E. coli concentration and LRVs in the treated effluent during the 10-day period when the paddlewheel mixing ceased was compared over the same period of 10 days for the continuously mixed HRAP. There was no statistical difference in E. coli concentration or LRV between HRAPs for any 10-day period when mixing ceased. Of particular significance was that the discontinuously mixed HRAP complied with the guideline treated effluent value of <E. coli 10⁴ MPN 100 mL⁻¹ over the 10 days when there was no mixing. Comparison on LRVs was made on all three of the 30-day trials between the discontinuously and continuously mixed ponds. The results from ‘30 day’ comparison showed that there were no significant differences in LRVs. To further a comparison of LRVs between the discontinuously and continuously mixed HRAPs was also performed on each 10-day period when mixing ceased, which also found there were no differences in LRV.

There was no statistical difference in effluent BOD₅ between the discontinuously and continuously mixed ponds (Table 3). The SA LGA Guidelines (2020) require a treated effluent BOD₅ < 20 mg L⁻¹. The BOD₅ of treated effluent from the continuously mixed HRAP complied with the guideline, whereas that of the discontinuously mixed effluent complied with the guideline in Experiment 1 but marginally exceeded the guideline in subsequent experiments.

The discontinuously and continuously mixed ponds performed similarly. The results showed that the cessation of mixing on an HRAP for 10 days has little effect on the wastewater treatment performance. The results of this study showed that the efficiency of wastewater treatment was not compromised by cessation of mixing, potentially caused by either mechanical or power failure, over a whole retention time (10 days). This finding has important implications for the capital cost of installing HRAPs as components of Community Wastewater Management Scheme (CWMS) since there is no imperative to install costly back-up power systems to manage the risk of mechanical, solar or power grid failure.

The study compared the wastewater treatment performance of two HRAPs at Kingston on Murray, one pond being continuously mixed and the other with mixing discontinued for 10 days, equivalent to one THRT, every 30 days over a 90-day period. The findings of the research showed that over the 90-day period of the study, both the discontinuously and continuously mixed ponds showed no statistically significant difference in relation to their wastewater treatment performance in terms of discharge values of both BOD₅ and E. coli; however, the discontinuously mixed HRAP marginally exceeded BOD₅ discharge guidelines in two out of three experiments. Although not relevant to the guidelines differences in microbial processes were observed, especially in nitrification–denitrification. The discontinuously mixed HRAP had significantly higher concentrations of NO₂-N and NO₃-N indicating higher rates of nitrification than the continuously mixed pond. This may be explained by sedimentation retaining nitrifying bacteria within the discontinuously mixed HRAP elevating nitrification rates. This study showed that intermittent mixing had no effect overall on the wastewater treatment performance. The research demonstrated that treated effluent quality was not compromised when paddlewheel mixing ceased for one retention time (10 days). These findings have important implications for the adoption of such technology, eliminating the requirement for the inclusion of expensive back-up power options, including battery storage, in the event of solar or power grid failure, reducing the capital costs for the construction and implementation of HRAPs.

The authors acknowledge Raj Indela for technical assistance, Loxton Waikerie District Council for access to the wastewater treatment plant and the South Australian Local Government Association for financial support.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.