The influence of public perception on acceptance of potable water reclamation schemes. The case of Diepsloot township, Johannesburg Open Access

The availability of clean water supplies in many parts of the world is increasingly limited by factors such as physical water scarcity, uneven distribution, pollution, poor management of water facilities, and over-extraction (Mekonnen & Hoekstra 2016; Van Rensburg 2016). With the increasing demand for clean water supplies due to population growth and other natural and anthropogenic factors, some water utilities in South Africa have considered incorporating reclaimed water into drinking water supply systems in urban areas to augment their water supplies (Leong 2016; Van Rensburg 2016; Lahnsteiner et al. 2018; Nappier et al. 2018; Redman et al. 2019). In this article, we define water reclamation/reuse/recycling as the process of treating wastewater to turn it into clean water that can be used for various purposes such as agriculture, irrigation, potable water supply for domestic use, groundwater replenishment, industrial processes, and environmental restoration (Tricas et al. 2017). It may be viewed as a reliable source of water for augmenting existing supply systems and thus partly presents one of the viable long-term solutions to water scarcity (Daigger & Crawford 2007; Redman et al. 2019).

Many scholars agree that the practice of water reclamation offers several benefits. These benefits include reduced demand for freshwater, more sustainable water management and supply (Garcia-Cuerva et al. 2016; Leong 2016; Redman et al. 2019), reduced wastewater pollution, and a cheaper unconventional water supply option compared to other options such as desalination and construction of new dams (Kucera 2014; Tchobanoglous et al. 2015). Research evidence has shown that reclaimed water can be treated to acceptable water quality levels that meet the international drinking water standards (Furlong et al. 2019; Mukherjee & Jensen 2020). Municipal authorities in cities such as Windhoek in Namibia, Changi in Singapore, Orange County in California in the United States, Perth in Australia, and Beaufort West in South Africa have already incorporated reclaimed water into their potable water supply systems (Drewes & Horstmeyer 2016; Van Rensburg 2016; Lahnsteiner et al. 2018; Nappier et al. 2018). However, the existence of evidence demonstrating the safety of reclaimed potable water and the successful implementation of such schemes in several parts of the world has not enabled public perceptions to change, and indeed, opposition to water reclamation and reuse in South Africa and other countries remains generally high in most communities (Ross et al. 2014; Fielding et al. 2019; Scruggs et al. 2019).

Considering the critical importance of public acceptance and perceptions towards successful implementation of potable water reuse schemes, a considerable amount of research has been done to understand the major driving factors for acceptance or rejection as well as strategies for promoting more significant support mechanisms for water reclamation efforts. For instance, a detailed study by Muanda et al. (2017) revealed that in South Africa, public acceptance of water reuse based on reclaimed water remains highly controversial; and proposals for potable water reuse in Cape Town and Durban were rejected by members of the public in favour of seawater desalination plants even though desalination is known to be much more expensive than reclamation. In several countries, proposed water reclamation schemes have also been abandoned or halted due to public opposition (Po et al. 2003; Massoud et al. 2018; Fielding et al. 2019). Well-known case studies include San Diego and Tampa in the United States; Toowoomba and Queensland in Australia; and Cape Town and Durban in South Africa (Po et al. 2003; Ross et al. 2014; Hurlimann & Dolnicar 2016; Mukherjee & Jensen 2020).

These experiences suggest that municipalities and other water supply agencies that promote water reclamation may not have sufficient knowledge on how to change public perceptions and increase acceptance levels for potable water reclamation schemes. Frameworks to address social factors influencing rejection or acceptance of reclaimed potable water are still limited. This study explored the perceptions of communities regarding the feasibility of using treated wastewater for potable reuse in the Township of Diepsloot. The main question that the study sought to answer is: What are the levels of acceptance for the use of reclaimed water by the community of Dieplsoot township? In addition, what are the main underlying factors that influence community acceptance or rejection of reclaimed water? The study is intended to inform the agendas of policymakers and practitioners who seek to promote the use of wastewater reclamation schemes.

Diepsloot township is a densely populated, low-income settlement located north of Johannesburg in South Africa and covering about 12 km² (Stats 2011). Current estimates suggest that Diepsloot is home to about 350,000 people (Sobantu & Nel 2019). The population is predominantly black (98.05%) and of low socioeconomic class (Stats 2011). Even though Diepsloot has a mixture of informal and formal settlements, the majority (65.5%) of the houses are informal. Essential services such as water, waste removal, and electricity exist in the formal areas of the township, while in informal settlements, there are limited connections to electricity, water, and sanitation (Williams et al. 2016).

Diepsloot has experienced rapid population growth and expansion since its establishment in 1996, leading to inequality in access to basic services such as water (Bénit 2002) Thus, the township was selected as a typical example of one of the fastest-growing, newest, and largest settlements in Johannesburg. About 18.4% of the households have access to tap water inside their houses, and 74% use a flush toilet connected to the main sewerage lines (Stats 2011). Diepsloot is situated close to the City of Johannesburg's largest wastewater treatment plant, named the Northern Wastewater Treatment Works, which has a capacity of 405 megalitres a day. The plant discharges treated effluent into the Jukskei River, which passes through Diepsloot township.

A household survey instrument was developed and administered to household representatives who were mostly household heads. If the household head was not present, any household member available on the day of the survey who is older than 18 years would be interviewed as the household representative. Households to be surveyed were randomly selected using ArcGIS, covering all the township zones. Based on the household survey results, informal interviews were also conducted to obtain community perspectives on reclaimed water. Diepsloot has approximately 62,882 households (Stats 2011). The number of households required for this study was estimated using an equation presented by Daniel (2000). The minimum sample size for this study was determined to be 384 households. The sample size was increased to 500 to account for non-responses and incomplete surveys.

A structured pre-tested questionnaire was administered to a household representative by trained data collectors in respondent's language. Before the interview, the respondents were made aware of the study objectives, and participants gave either verbal or signed informed consent. The survey instruments comprised sections on sociodemographics, access to water, attitudes toward water reclamation, and levels of satisfaction regarding the use of reclaimed water. Ethical approval for the study was obtained from the Human Sciences Research Council Ethics Board (No REC 8/20/09/17).

In the context of this study, AIC represented by the aforementioned equation was used to estimate prediction error, bringing forth the relative quality of the statistical model for community acceptance of water reclamation for potable reuse.

The study results show that more males participated in the study than females (Table 1). Most of the respondents had secondary education, while less than 10% of the respondents had tertiary education. The results further show that most of the respondents were unemployed. The predominant water sources were taps inside the yard and communal taps. The third most common source of drinking water was a tap inside the yard. Almost all respondents use tap water as a drinking water source. Of all the households with access to tap water, the water was supplied by the local municipality. When asked if the household paid for water services, most of the respondents indicated that they were not paying. To assess their awareness of the utility of water recycling, the respondents were asked whether they recycle water at home, for example, using bath water to water the lawn or flush the toilet. Almost a quarter of the respondents indicated that they recycle water. They further explained that they usually recycle water during periods of water supply interruptions.

The results show that water supply services are unreliable; this is demonstrated by a higher proportion of respondents who have experienced water interruptions (Table 1). A large proportion of the respondents indicated that these interruptions typically last for more than 2 days. Regarding the frequency, approximately 9 of 10 respondents experienced water interruptions one to three times a month. During these interruptions, the predominant water source was stored water, followed by other areas in the township not experiencing interruptions.

Table 2 summarizes the responses on acceptance of the use of water reclamation. The results of the study show a high acceptance rate for water reclamation. The majority of the respondents expressed that they did not have any problems with drinking reclaimed water. Some of the respondents mentioned that they heard rumours that the water they currently use for domestic purposes was already mixed with treated wastewater, and therefore, it did not make a difference. The respondents were further asked about their acceptance of reclaimed water for four specific uses: drinking, bathing, washing clothes, and irrigation. Bathing had the highest acceptance rate, followed by washing clothes. Irrigation had the lowest acceptance rate among the four uses (Table 2).

A Chi-square was undertaken to assess if gender, education, drinking water sources, payment of water, water treatment, water recycling, and community satisfaction with water quality and quantity influenced community acceptance of reclaimed water (Table 3). The results indicate a significant relationship between education and water recycling. Acceptance of reclaimed water was higher among respondents having tertiary education than those with primary and secondary education. A greater percentage of respondents who recycle water at home were more willing to accept reclaimed water than those who do not recycle water. Gender, water treatment, payment for water services, water interruptions, and drinking water sources did not significantly influence water reclamation acceptance.

A logistic regression model was developed to determine the effects of education, recycling water at home, and government assurance of water safety on water reclamation acceptance (Table 4). The model was statistically significant at X² = 179.770, p < 0.001, and the AIC was 28.487. The model accurately predicted 84.8% of the cases and explained 12.3% of the variance in water reclamation acceptance.

All variables were significant predictors of water reclamation acceptance. Respondents from households who recycle water were 2.9 times more likely to accept reclaimed water. The government's assurance of water safety also positively and significantly influenced the acceptance of reclaimed water. Having not gone to school at all had a statistically negative influence on water reclamation acceptance. Respondents who never went to school were 0.2 times more likely to reject water reclamation.

The study showed that even though Diepsloot residents have access to clean water, the supply is not regular, and thus, they resort to using alternative water sources such as rivers and stored water, which may not be safe. This is indicated by the number of protests in the area demanding basic service delivery, including housing, sanitation, and water. Although there is enough water to meet the current demand, the rapid population growth witnessed over the years in Diepsloot suggests that in the foreseeable future, water supply might not be sufficient to meet the increased demand. Given the lack of conventional water sources near Diepsloot, the township might have to diversify its water supply portfolio in the future.

Diepsloot township's proximity to the largest wastewater treatment plant in Johannesburg makes water reclamation a potentially viable and sustainable solution for predicted shortages. The relatively high levels of acceptance of reclaimed water for domestic purposes (85%) demonstrated in this study also support this assumption. It is also important to note that the acceptance rate for reclaimed water in this study is higher than what was witnessed in some studies undertaken elsewhere. For instance, a similar study conducted using the same instrument and methods to investigate community attitudes and perceptions towards a planned water reclamation plant in Chambishi Township in Zambia reported an acceptance rate of 68% (Siame et al. 2020).

A more recent study by Redman et al. (2019) in Nevada, USA, reported that 35% of the respondents were willing to drink reclaimed water. Hurlimann & Dolnicar (2016) reported an acceptance rate of 13% in Israel. The higher acceptance level in Diepsloot compared to Chambishi and other studies could be explained by Diepsloot township's proximity to a wastewater treatment plant. In addition, some residents believe that the water they drink comes from the nearby wastewater treatment plant in Diepsloot. However, most of the extant literature do not show a direct link between levels of acceptance and community proximity to wastewater treatment plants. The difference with our study could be explained by the study area settings and the methodology used. Although Redman et al. (2019) and Gu et al. (2015) used large metropolitan areas with heterogeneous settlement classes as their case studies, our study only used a low-income settlement within a metropolitan area.

In our study, payment for water stood out as a major driver for acceptance or rejection of reclaimed water, with higher prices acting as disincentives for acceptance and vice-versa. More detailed studies will be needed to articulate water payment readiness to enhance acceptance levels for reclaimed water in low-resource settings such as Diepsloot. Regarding the intended uses of reclaimed water, the study had four options for the respondents, that is, drinking, bathing, washing clothes, and irrigation. Surprisingly, bathing was found to have the highest preference followed by washing clothes. Although the preference was higher than 80%, irrigation and drinking had the lowest preferences. This finding is contrary to previous studies which showed that as human contact with reclaimed water increases, the acceptance rate decreases (Fielding et al. 2019). This inconsistency may be due to the number of options that were provided in this study that excluded toilet flushing and cleaning, which were included in other studies.

Assurance by the government that reclaimed water is safe for human consumption increased acceptance levels to 97%. This further demonstrates the crucial role of trust in water utilities for the acceptance of reclaimed water. The responses highlight, for example, that the respondents were prepared to drink reclaimed water only if they were assured of its safety for human consumption and also if they saw government officials drinking the water. This finding resonates with other studies that identified trust as partly determining public acceptance of reclaimed water (Price et al. 2012; Ormerod & Scott 2013). To increase the trust in water utilities, South Africa could adopt some of the strategies used in other countries that have successfully implemented water reclamation schemes. For instance, Singapore's NEWater project was successful mainly because of effective communication strategies that were used to educate and gather public support (Hartley et al. 2019). However, in contrast, South Africa has not implemented any communication strategy for water reclamation schemes despite the existence of the National Water Reuse Strategy.

Findings from this study also show that education, recycling, price, and assurance of water safety were linked to increased levels of acceptance of reclaimed water. Of all sociodemographic characteristics, education had a significant influence on community acceptance of reclaimed water. Willingness to accept reclaimed water was higher in respondents with formal education training compared to those who do not have any formal education. The practice of reusing water at the household level, such as using bath water to irrigate and flush toilets during water interruption, also significantly influenced the acceptance of reclaimed water. Research has shown that one of the most efficient ways to influence public opinion on reclaimed water is by raising public awareness about water scarcity and water reclamation. However, the public should be provided with factual information instead of persuasive information. Indeed, research has already shown that providing factual information about reclaimed water increases knowledge and acceptance of reclaimed water (Fielding et al. 2019).

Considering the key issues highlighted in this study, it is recommended that a communication strategy on water reuse be developed as recommended in the National Strategy on Water Reuse for South Africa to address negative public perceptions and increase awareness of the utility of water reclamation in augmenting current supplies. The involvement of water users and communities in the planning of the schemes is vital as it enhances public trust in the decisions of water utilities and municipalities. Future research should assess the impact of different public education programs on community acceptance of water reclamation. While the study acknowledges that the public should be more involved in water policy-making, an in-depth analysis of the link between understanding water processes, public participation in policy-making, and support for water reuse is still necessary.

We set out to identify the main driving factors for acceptance or rejection of water reclamation initiatives and schemes. From the study, it is clear that community's willingness to accept reclaimed water as an alternative source of potable water supply is influenced by the level of education, the current cost of the conventional water supply, and government assurance on the safety of reclaimed water. Among the factors mentioned earlier, recycling water at home is a key driver for public acceptance of reclaimed water among vulnerable urban communities. The success of water reclamation projects for potable reuse in Diepsloot and other low-resource settings of South Africa requires a strong focus on raising awareness, educating communities, and increasing the communities' trust in water utilities that they will provide safe, clean, and affordable water. This approach stands a greater chance of raising acceptance levels for water reclamation schemes. The study provides a valuable reference for other cities of South Africa and developing countries grappling with the development of reclaimed water as an option for addressing water scarcity challenges.

This work is based on the research supported by the Leading Integrated Research Agenda for 2030 in Africa Programme, which is implemented by the Network of African Science Academies (NASAC) and the International Science Council (ISC), with support from the Swedish International Development Cooperation Agency (Sida). The authors wish to acknowledge Diepsloot township residents for taking part in the study.

The study was funded by the National Research Foundation of South Africa (Grant Number: MND210901636792)

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.