EVAS-a practical tool to assess the sustainability of small wastewater treatment systems in low and lower-middle-income countries

• Presents a sustainability assessment tool for small wastewater treatment systems • Indicators covering five sustainability dimensions provide a holistic approach. • Uses contextualized sustainability indicators measured by specific targets • Practically assesses availability of local resources for optimal management. • Suggests action to raise the sustainability of the wastewater treatment systems. ⁎ Corresponding author at: Department of Architecture E-mail addresses: claudia.cossio@outlook.com (C. Coss https://doi.org/10.1016/j.scitotenv.2020.140938 0048-9697/© 2020 The Authors. Published by Elsevier B.V a b s t r a c t a r t i c l e i n f o


Background
Rapid urbanization in low and lower-middle income countries, especially in small towns where the existing infrastructure and management capacity are usually limited, accentuates the challenge of providing a fully functioning wastewater treatment service (Caplan and Harvey, 2010;Singh et al., 2015;Sundaravadivel and Vigneswaran, 2003). A small town is defined as a settlement with a mix of urban and rural characteristics (Caplan and Harvey, 2010), which in Bolivia translates into a settlement with a population of 2000-10,000 inhabitants (O'Hare and Rivas, 2007;The World Bank, 2017). The lack of financial resources to operate and maintain small wastewater treatment plants (WWTPs) in low and lower-middle-income countries has been identified as a key reason for the failure of these systems (Brunner et al., 2018;Cossio et al., 2017;Usha Rani and Vasumathi, 2013), alongside inappropriate design or selection of technology (Bdour, 2007;Brissaud, 2007;Mara, 2003;Massoud et al., 2009), poor operation and maintenance (Noyola et al., 2012;Singhirunnusorn and Stenstrom, 2009), lack of technical expertise (Cossio et al., 2019;Noyola et al., 2012;Ujang and Buckley, 2002;Zurita et al., 2012) and lack of monitoring (Cossio et al., 2017;Massoud et al., 2009). According to the United Nations, the achievement of the sustainable development goals (SDGs) is highly dependent on successful management of urban growth, especially in low and lower-middle income countries, which are expected to report the highest growth rate (UN, 2018a). Progress towards achieving Target 6.3 of SDG6 (Clean water and sanitation) is measured using two global indicators: Proportion of wastewater safely treated (6.3.1) and Proportion of bodies of water with good ambient water quality (6.3.2) (UN, 2016;UN, 2018b). Effective and sustainable wastewater treatment in small towns is vital for the successful fulfilment of the criteria set out in these indicators.
The Sustainable Sanitation Alliance (SuSanA) states that a sustainable wastewater management system should be economically viable, socially acceptable, and technically and institutionally appropriate, and it should be designed to safeguard human health, the environment, and natural resources (Andersson et al., 2016). A sustainable small WWTS should meet a broad range of locally and context-dependent needs (e.g. (Benavides et al., 2019;Cossio et al., 2020;Massoud et al., 2009;Nansubuga et al., 2016;Schweitzer and Mihelcic, 2012). Small and decentralized wastewater treatment systems (WWTSs) are also important for effective local recovery of water, sludge and nutrients (Brunner et al., 2018;Massoud et al., 2009), particularly in areas in which water is scarce (Nansubuga et al., 2016), but also as part of a circular economy that is critical to achieving sustainability (Brunner et al., 2018;Molina-Moreno et al., 2017;Molina-Sánchez et al., 2018;Salguero-Puerta et al., 2019;Verbyla et al., 2013). Developing sustainable small WWTSs in low and lower-middle income countries is thus a complex task and accordingly there is a need to support decision-making among local stakeholders when planning and improving the sustainability of these systems (Brunner et al., 2018;Clarkson et al., 2010;Cossio et al., 2020;Dhinadhayalan and Nema, 2012).
Developing a tool to assess sustainability is a challenge (Benavides et al., 2019), as it involves the need to bridge the gap between theoretical concepts and practical measurements (Domínguez et al., 2019;White, 2013). To facilitate the adoption of a holistic approach, such tool should ideally include all the locally relevant aspects the services are required to meet from a sustainability point of view (Sala et al., 2015). Consequently, a tool should typically take into account a variety of different criteria that need to be measured using different units (Molinos-Senante et al., 2014;Villeneuve et al., 2017). Data requirements for a holistic assessment are generally extensive, which in low and lower-middle income countries means further adaptation of such tool as in this context lack of data is common (Benavides et al., 2019). Almost all tools implicitly or explicitly include weighting of criteria, which is often subjective, sometimes using input from local experts or stakeholders (Ahi et al., 2018;Arroyo and Molinos-Senante, 2018;Molinos-Senante et al., 2014;Muga and Mihelcic, 2008;Singh et al., 2012). If a tool is to be adapted to the local context, it must not only be able to take account of relevant local aspects and encourage the participation of local stakeholders (Hoffmann et al., 2000;Venkatesh et al., 2017), but also encourage decision-makers and policy-makers to put plans or actions into effect at a higher level with the ultimate aim of achieving sustainability (Pope et al., 2004;Pope et al., 2017).
There are several tools that have been developed to support decisions regarding wastewater, each with a specific focus. Examples include the Burden to Capacity Sustainability Assessment (B2C SA) tool for a wastewater treatment infrastructure applied in China (Murray et al., 2009); a Performance Measurement System applied in Italy for improving the efficiency and profitability of larger WWTPs from an internal and managerial perspective (Guerrini et al., 2016); a tool for combining environmental and economic aspects when designing WWTPs for small communities, demonstrated in nine scenarios with different characteristics (Molinos-Senante et al., 2012); a target plot with 13 indicators to evaluate the sustainability of decentralized domestic wastewater treatment technologies (Mena-Ulecia and Hernández-Hernández, 2015); and the decision support tool FitWater, developed to support fit-for-purpose wastewater treatment trains by assessing different ways of reusing wastewater (Chhipi-Shrestha et al., 2017). In Latin America, Benavides et al. (2019) developed a comprehensive and regionally relevant sustainability assessment framework where sustainability was measured by adopting a distance-to-target approach using the traffic light method.
In general, the existing sustainability assessment tools examine specific dimensions, e.g. assessing the economic dimension in relation to the technical or environmental dimensions, using extensive data as well as methods that are generally applied by experts. However, there is often a gap between expert studies on sustainability and practical recommendations that local operators and decision-makers can implement. Development of a tool that uses local data regarding actual performance of the WWTS in operation, current management practices and other aspects relevant to the context of low and lower-middle income countries, including technical, environmental, economic, social and institutional aspects, and can be easily applied by practitioners is expected to overcome this gap.

Objectives of the study
The overall aim of this study was to develop a sustainability assessment tool to support the decision-making process at existing small WWTSs in low and lower-middle-income countries by increasing the capability and knowledge of WWTS managers, enabling them to take action or plan for long-term improvements in an effort to enhance overall performance and sustainability. In light of the problems typically experienced at small WWTSs, an important aspect was to develop a tool that is possible to use despite that some input data may be missing, and which can support managers and stakeholders to collect relevant data from a sustainability perspective. The specific objectives of this paper are i) to develop a sustainability assessment tool for small WWTSs in low and lower-middle-income countries, and ii) to test the sustainability assessment tool in practice using two case study sites in order to verify the practicality and utility of the tool.

Method
The assessment tool, named EVAS (EVAluación de Sostenibilidad: EVAluation of Sustainability), presented in this study is designed to help managers of small WWTSs improve the operational status and sustainability of their systems. The tool is designed to assess systems already in operation. Development and initial testing of the tool are described in Sections 2.1 and 2.2. The interface is in Excel, and an overview of the components that make up the tool is presented in Fig. 1. The tool is available at https://research.chalmers.se/en/publication/516714.

Development of the suggested sustainability assessment tool
Development of the sustainability assessment tool was based on five sustainability dimensions (technical, environmental, social, economic, institutional) and 21 sustainability indicators. The 21 sustainability indicators applied in the tool were extracted from a list of 27 indicatorsidentified in a previous studyfor small WWTPs in low and lowermiddle-income countries (Cossio et al., 2020). During the development of this tool, the original list of 27 indicators was revised to minimize overlapping indicators and focus on indicators relevant to the operational phase. The changes made in the original list of 27 indicators are listed below. Sub-indicators were added to the 21 indicators as necessary in order to specify particular performance aspects or management requirements (Table 1).
• Technical dimension: The original Removal of BOD, TSS, TN, TP and FC 1 indicator was moved from the environmental dimension to the technical dimension and named Removal efficiency (T1). The indicator Complexity of construction and O&M was only evaluated with regard to Operation and maintenance (T2), as the tool is focused on the operational phase and not on choice of technology. • Environmental dimension: Potential recycling was specifically interpreted as Potential safe reuse (E1) to highlight the need to avoid unacceptable human health risks due to microbial contamination (Cossio et al., 2019;Qadir et al., 2010). The indicator for Quality of effluent and sludge was split: the quality of sludge and water in terms of FC and Helminth eggs was included under Potential safe reuse (E1), and water quality in terms of BOD and TSS was included under Effluent quality (E3 There is a slight overlapping of the technical indicator of Removal efficiency (BOD, TSS, TN, TP and FC) with the environmental indicators Potential safe reuse (FC and Helminth eggs), Eutrophication potential (TN, TP), and Effluent quality (BOD, TSS). However, these indicators focus on different aspectsthe technology and how well it works on one hand, and whether the effluent complies with safe reuse standards and discharge limits on the other hand.
To assess whether indicators are being fulfilled, a number of factors and corresponding assessment questions were identified for each indicator. For example, factors affecting the indicator Removal efficiency (T1) are Monitoring of removal efficiency (T1-I), Actual removal efficiency (T1-II), and Existence of a long-term plan for improving low performance (T1-III). Another example is the indicator Institutional capacity of the utility (I1), which includes four factors: Roles and responsibilities (I1-I), Procedures (I1-II), Goals of the organization (I1-III), and Long-term goals (I1-IV). Each factor has a multiple-choice assessment question and the answers are scored according to a simplified scale (Table 2), using the logic employed in the traffic light method (Benavides et al., 2019). The details of all 64 factors, how the questions are asked, and the predefined scoring scales for the answers are presented in Supplementary Material 1.
A normalised weighted score is calculated for each sub-indicator, indicator, and dimension, and for the system as a whole. The tool reports the results for factors, sub-indicators and indicators, with only one significant number, whereas the results for dimensions and the overall score are presented with one decimal number. All results are scored according to the ranges detailed in Table 2. The tool allows users to change weightings for dimensions, indicators, sub-indicators, and factors if required, although it is pre-set with a standardized weighting system with a weighting of 1 for all, except the factors that can directly affect measured performance at the WWTS (i.e. removal efficiency, coverage of the sewer network, design of the sewer network, hydraulic loading to the sewer network, design of the WWTP, hydraulic loading to the WWTP, meeting safe reuse limits, eutrophication potential, discharge limits, exposure to public health risks and nuisance level), where the weighting is pre-set at 2. All equations for the calculations of weights and weighted scores are specified in Appendix A. An overview of all sustainability dimensions, indicators, sub-indicators, and factors, as well as Welcome to EVAS A sustainability assessment tool for small wastewater treatment systems (ver. 1.0, 2020)

Instructions:
1. ADMINISTRATIVE DETAILS: Fill out the general information regarding the WWTS that is being assessed in this sheet.
2. OVERVIEW and WEIGHTING: in the sheet OVERVIEW, the weights assigned to indicators, sub-indicators and factors can be found. If you consider that another weighting scheme applies to your case, you can change these weights (in the cells marked with a blue frame).

SCORING:
Enter the answers to the questions on the TEC, ENV, SOC, ECO, and INS sheets by choosing the option that best applies to your case from the drop-down list available for each question. If required for your records, make short notes on the reasoning behind your choice. Scores are calculated automatically.  their respective standard weighting, is presented in Table B1, Appendix B. The tool aims to assess the current status of the WWTS and how far it is from an "ideal" situation, i.e. to measure the distance-to-target. Consequently, all defined scoring scales are global (Monat, 2009) with scores ranging from 0 to 4 (with 4 as the best). Achieving the maximum score also means achieving the defined target.

Application of the sustainability assessment tool
The sustainability assessment tool was tested in two case studies involving small WWTSs located in Cochabamba, Bolivia. The testing was carried out by a local assessor and a researcher jointly entering data into the sustainability assessment tool, i.e. the assessor provided the input data and the researcher guided the process. The researcher also noted the comments provided by the assessor regarding the relevance of the questions, difficulty answering them, and discussions regarding potential to improve data collection or to modify the questions in the tool.

Case study 1
The first wastewater treatment system (WWTS-1) is located in a small town in the Cochabamba Valley and provides services to 7980 population equivalents. WWTS-1 includes pre-treatment, a storage tank, a mechanical screen, and five treatment trains in parallel. Each  Table 2 Overview of the traffic light scoring system used in the EVAS tool and an example of how a factor is scored in relation to the corresponding multiple-choice questions and answers.
Integrated scores are classified according to the following ranges: score 0 = 0; score 1: N0 and b1.5; score 2: ≥1.5 and b2.5; score 3: ≥2.5 and b3.5; score 4: ≥3.5. treatment train has a grease chamber, an upflow anaerobic sludge blanket (UASB) reactor, two biofilters in parallel, and a sludge drying bed. In total, the WWTP has five grease chambers, five UASB reactors, and 10 biofilters to treat the wastewater. The construction of the WWTS-1 was funded through a combination of external support, municipal funding, and user contributions. The municipality is responsible for the financial management of the WWTS-1. The operation and maintenance of the sewer network and the WWTP is carried out by the foundation, which was also responsible for constructing the WWTP at the request of the municipality. The staff responsible for the management of WWTS-1 (WWTP and sewer network) include a full-time operator as well as a manager, engineer, communicator, and accountant, all of whom work part-time. The assessor in this case was the manager, who provided all the input data for the assessment and contributed to the testing of the tool.

Case study 2
The second wastewater treatment system (WWTS-2) is located in another small town in the Cochabamba Valley and provides services to 3344 population equivalents. The treatment processes include pretreatment, a UASB reactor, two biofilters, a maturation pond, and a sludge drying bed. The treatment plant was funded by the national government. Once construction was completed, responsibility for the WWTS-2 was transferred to the local water association. The water association was already responsible for the drinking water services before taking on responsibility for the management of the sewer network and wastewater treatment. For this case study, the tool was tested and validated using monitoring data and information collected in a previous study (Cossio et al., 2019). The assessment in this case was made by two local researchers (Cossio and Mercado) with the support of an undergraduate student who at the time was carrying out fieldwork at the WWTP.

Sensitivity analysis
A simplified sensitivity analysis was carried out for the case study assessments to investigate the effects of applying different weights to indicators and dimensions using scenario analysis. Three scenarios were investigated: i) all weights in the tool being equal to 1 (see Appendix B), ii) a ranking of sustainability dimensions according to technical and social experts from the study by Cossio et al. (2020) where the institutional dimension was given a weight of 3, the social, economic and technical dimensions were each given a weight of 2 and the environmental dimension was given a weight of 1, and iii) each of the 21 sustainability indicators was weighted equal (4.76% each), thus compensating for the hierarchical structure which results in lower weights for indicators in dimensions with more indicators than for indicators in dimensions with fewer indicators.

Sustainability assessment of case study 1
The overall results obtained from the sustainability assessment of WWTS-1 are summarized in Table 3. The sustainability assessment of WWTS-1 resulted in an overall sustainability score of 2.9, i.e. medium sustainability level. At the indicator level, most indicators (11 out of 21) received a score of 3. Five indicators in the assessment received a high sustainability score (4): Reliability of the WWTP, Potential safe reuse, Public acceptance, Institutional capacity at the utility, and Employee satisfaction. Only five indicators received low scores (1 or 2) at the indicator level. The indicators Sewer network functionality, Affordability, and Institutional capacity on a higher level received a score of 2, whereas Energy and Global warming potential received a score of 1.
Each indicator score is a weighted sum of the individual scores under each sub-indicator and factor (equations in Appendix A), and Fig. 2 presents the percentages of the individual scores under each dimension. There are some low scores in each dimension, although the social dimension stands out in the assessment with no red scores and only 7% yellow scores.
Details of all low scores (0, 1 and 2) in each dimension in the assessment are presented in Appendix C, Table C1. This table can be viewed as a summarized to-do list that could support the identification and prioritization of actions. For example, the indicator T3 Sewer network functionality scored 1 for factor T3-II Design and factor T3-IV Hydraulic loading, which are on top of the to-do list as their factor weightings are 2; and factor T3-III Breakdown, factor T3-V Clogging and factor T3-VI Long-term plan SEW are next on the to-do list as their factor weightings are 1. In other words, a prioritized activity in Case study 1 regarding indicator T3 Sewer network functionality would be to implement a long-term plan to improve the design of the sewer network to ensure better management of failures due to breakdown, clogging and hydraulic loading. The detailed results of the assessment of Case study 1 are presented in a locked version of the tool itselfin an Excel fileas Supplementary material 2. In the following sections all the results are presented according to each dimension.

Technical dimension
The indicator scores for Removal efficiency (T1) and Operation and maintenance (T2) are 3, indicating that in general the WWTP monitors Table 3 Summary of the EVAS assessment results for Case study 1.
1/red = requirement not fulfilled, 2/yellow = requirement partly fulfilled, 3/light green = requirement fulfilled relatively well, 4/ dark green = requirement fulfilled. and meets effluent requirements most of the time and that most of the treatment unit processes at the WWTP are working well. Removal efficiency for TSS and FC received a score of 2, as well as Long-term maintenance of all treatment units (pre-treatment, UASB reactor and Biofilter). As mentioned in the example in the previous section, Sewer network functionality (T3) received a low score (2) in this case study. However, Coverage of the sewer network (T3-I) is satisfactory and received the highest score (4). Reliability of the WWTP (T4) received a score of 4 for all factors included in the indicator. Actions directed at improving the sewer network should thus be prioritized in the technical dimension, as well as following up the low removal rates of TSS and FC and improving the long-term maintenance of the treatment units at the WWTP.

Environmental dimension
The reuse of both treated wastewater and sludge is practised and for the indicator Potential safe reuse (E1), the lowest factor score (1) was due to a lack of regular monitoring of the sludge (E1b-II). However, all monitoring samples that were collected and analysed for water (once every three months) and sludge (once a year) complied with safe reuse limits in relation to the type of crops that were being irrigated (in this case corn). As regards eutrophication potential, the subindicators TN and TP scored 3 for Monitoring -Eutrophication Potential (E2-I) and Compliance with discharge limits to avoid eutrophication (E2-II). All sub-indicators and factors under the indicator Effluent quality (E3) relating to BOD and TSS received scores of 3 and 4. The WWTP always complies with the limits for BOD removal (E3a-II), but only in 70% or more of the samples for TSS removal (E3b-II). The frequency of Monitoring-Effluent quality (E3-I) and compliance with limits for TSS (E3b-II) could be further improved to obtain the maximum score of 4 for both factors. The lowest score (1) in the environmental dimension was received for Energy (E4) and Global warming potential (E5), and it is clear that energy saving, use of renewable energy sources, and reducing emissions have not yet been prioritized in this system.

Social dimension
In general, the assessment of the social dimension resulted in high scores. All but one scored either 3 or 4, which means that wastewater treatment practice is accepted locally. The only low score (2) was related to Claims (S1-II), i.e. there are claims from the public at least once a year related to people or animals being affected negatively by discharges from the WWTP. This is linked to the quality of the effluent discharged and it is probable that improvements in the technical performance will reduce the number of claims. In the social dimension, the level of Public awareness (S2) of the benefits of the WWTP could be increased and action should be taken at the WWTP to reduce the smell (S3a-II). Despite occasion public claims and odour problems, the high score of 4 for the Public acceptance (S4) indicator confirms that the local population values the benefits the WWTP brings to the town, specially to farmers that reuse the effluent.

Economic dimension
The assessment of the Tariffs (EC1) indicator showed a very good Tariff structure (EC1-I, score 4) and very good Documentation of financial management with monthly and annual financial reports available for customers (EC1-II, score 4). However, Follow-up (EC1-III) to ensure the tariff structure is correct over time requires improvement (score 1). As regards Operation & maintenance costs (EC2), the economic requirements for Optimal operation and maintenance (EC2-I) and Long-term investment (EC2-II) received scores of 3. Affordability (EC3) scored 3, indicating that a large proportion of the customers pay their tariffs on time (EC3-I), but scored 2 for factor EC3-II, in turn indicating that the payment capacity of customers is not very well known and the municipality does not have very effective strategies in place to raise the affordability level among customers (EC3-III). Efforts to acquire a better understanding of the true affordability level should be implemented.

Institutional dimension
The indicator Institutional capacity (I1) at the utility level scored the highest (4) in all the factors included in the assessment. The foundation in charge of the management of the WWTS has experience in implementing and managing other WWTPs in the study area. The indicator Institutional capacity at a higher level (I2) scored 2, meaning that better-structured support is required from the institutions at higher levels. Although the roles of the municipal technical services department and the Ministry of Environment and Water are well-defined (I2a-I and I2c\ \I), the actual level of fulfilment of their duties is low. External support from the municipality (I2a-II) is low (2) and support from the Ministry (I2c-II) is very low (1). At the Regional level (I2b), roles and duties are neither well-defined nor fulfilled, both scoring 1. As regards Staff requirements (I3), an operator and an engineer with adequate skills scored 4 and 3, and they are generally available and receive regular training (score 4). A communicator, accountant, and manager are available part-time, and although their level of expertise meets the requirements (score 3), they do not receive regular training, and consequently the maximum score of 4 could not be obtained. The manager indicated that due to the size of the WWTP it is not financially viable to have all the staff employed full-time. As they also manage other WWTPs, staff salaries are covered by more than one WWTP. As regards the Employee satisfaction (I4) indicator, the score for working conditions was 3 due to a lack of documentation of safety procedures, although it scored 4 for salary level. As regards the Communication (I5) indicator, communication by managers to customers about Use of the sewer network (I5\ \I) scored the maximum of 4. However, for information regarding the performance of the WWTP (I5-II), it received a score of 2 as the information is not distributed to customers. Although the staff inform users when there is a risk of exposure (I5-III) due to a discharge of poorly treated wastewater, no standard procedure has been introduced and consequently the score was 3.

Sustainability assessment of case study 2
The results obtained from the sustainability assessment of WWTS-2 are summarized in Table 4. The sustainability assessment of WWTS-2 resulted in an overall sustainability score of 1.6, i.e. a Low level of sustainability. On the indicator level, almost half of the indicators (9 out of 21) received a score of 1, resulting in at least one unsustainable indicator in each dimension. Ten indicators in the assessment scored 2 and only two indicators received a score of 3 (the technical indicator T4 -Reliability and the social indicator S3 -Aesthetics). No indicator received the highest score (4) in the assessment.
The assessment resulted in a majority with scores of 1 and 2, with only the technical dimension and social dimension receiving a low proportion of 4 scores (Fig. 3). The economic and institutional dimensions have the highest proportion of indicators that scored 3: 33% and 20%, respectively. From Fig. 3, it can also be seen that there are three factors that could not be assessed due to a lack of information, as indicated by the 0 score in the environmental and economic dimensions (i.e. Compliance with reuse limits for sludge (E1b-III), Payment capacity (EC3-II) and Strategy for Affordability (EC3-III)). Detailed results of all low scores (0, 1 and 2) in each dimension in the assessment are presented in Appendix C, Table C2. Supplementary material 3 presents the detailed results of the assessment of Case study 2 in a locked version of the EVAS tool as an Excel file. In the following sections, the results of Case study 2 are summarized according to dimension.

Technical dimension
The Removal efficiency indicator (T1) is assessed to be unsustainable. The factor Monitoring for removal efficiency (T1-I) scored 1 for all subindicators (BOD, TSS, TN, TP, FC), corresponding to a monitoring frequency of once a year. The actual Removal efficiency (T1-II) is b50% (score 1) for all, while for TSS the removal efficiency is between ≥80% and b95% (score 3). A Long-term plan (T1-III) to improve the situation for all sub-indicators that scored 1 for Removal efficiency is only partially in place. As regards Operation and Maintenance (T2), the situation is almost as unsustainable with a score of 2. There are no O&M manuals (T2-I) for any of the treatment levels, i.e. pre-treatment, and primary, secondary, and tertiary treatment (score 1), and although Short-term O&M (T2-II) is performed in part (score 2), Long-term maintenance is lacking for all treatment steps (score 1) except for pre-treatment (T2a-III: score 2). As regards Sewer network functionality (T3), scores of 1 were given to the factors Design (T3-II), Hydraulic loading (T3-IV), Clogging (T3-V), and Long-term plan for the sewer network (T3-VI). Coverage (T3-I) scored 2 as the coverage of the sewer network is between ≥50% and b75%, but since Breakdowns due to material failure in the network (T3-III) are rare, it scored 4. As regards the Reliability of the WWTP (T4), the Design of the WWTP (T4-I) only includes a bypass and due to the relatively high frequency (once a month) of Hydraulic loading (T4-III) reaching the WWTP, both these factors scored 2. The remainder of the factors under this indicator scored 4.

Environmental dimension
In the environmental dimension, the assessment of the indicator Potential safe reuse (E1) showed that water is sometimes reused (E1a-I), resulting in a score of 2, and that sludge is always reused (E1b-I) resulting in a score of 3. However, there is no Monitoring for a safe reuse (E1-II) of either water or sludge for reuse (score 1), the Safe reuse limits (E1-III) are not being complied with for water (score 1), and no information is available regarding sludge (score 0). Furthermore, there is no Long-term plan (E1-IV) for meeting safe reuse limits (score 1). The Eutrophication potential (E2) scored 1 for both subindicators (TN and TP) in all the factors: Monitoring for eutrophication potential (E2-I), Eutrophication potential (E2-II), and Long-term plan to reduce eutrophication potential (E2-III). Effluent quality (E3) received a score of 1 for Monitoring (E3-I) and Compliance with Discharge limits (E3-II) for BOD (E3a-II), but 2 for TSS (E3b-II). In the case of TSS and BOD, Long-term plans (E3-III) to improve the quality of the effluent are only partly in place, hence the score of 2. Similar to Case study 1, the environmental indicators Energy (E4) and Global warming potential (E5) only received a score of 1. 1/red = requirement not fulfilled, 2/yellow = requirement partly fulfilled, 3/light green = requirement fulfilled relatively well, 4/ dark green = requirement fulfilled.

Social dimension
The problems experienced in the technical and environmental dimensions regarding the discharge of poorly treated wastewater are also reflected in the Public health (S1) indicator, where all the factors scored 1, i.e. Exposure (S1-I), Claims (S1-II), and Long-term plan to protect public health (S1-III). The quality of the effluent needs to be improved to ensure public health is safeguarded. The Level of public awareness (S2) for Public health concerns (S2-II) scored 3, i.e. the users acknowledge the importance of the WWTP in safeguarding public health. However, as public awareness in relation to Ecological concerns (S2-I) and proper Use of services (S2-III) was low, and there is no Long-term plan to raise Public awareness (S2-IV), they all scored 1. As regards the indicator Aesthetics (S3), only Smell was identified as an issue with regard to Nuisance level (S3a-II), scoring 2, but there was no nuisance with regard to Noise (S3b-II) or Visual impact (S3c-II), each scoring 4. There was no longterm plan for reducing smell (S3a-III). Public acceptance (S4) of the WWTP is also quite low. The public is opposed to Reuse practice for water and sludge from the WWTP (S4-II: score 1), and partly opposed to the Location of the WWTP (S4-I: score 2). There is no Long-term plan to increase Public acceptance (S4-III), resulting in a score of 1.

Economic dimension
For the indicators Tariffs (EC1), Tariff structure (EC1-I), and Documentation on financial management (EC1-II) the score was 3, but 1 for not having a Follow-up system in place to ensure an appropriate tariff over time (EC1-III). The result of the assessment of the Operation and maintenance costs indicator (EC2) clearly shows there is insufficient funding to implement optimal O&M. The Economic requirements, EC2-I factor, i.e. the tariffs collected from users on a monthly basis, do not cover the regular O&M costs, thus scoring 1. In contrast, funding for Long-term investment is partly assured (EC2-II: score 2) and a Long-term plan (EC2-III: score 2) to solve these financial issues is partly in place. Affordability (EC3) was in part good with regard to Payment of tariffs (EC3-I: score 3), as N70% of the customers pay on time. However, as regards Payment capacity of the users (EC3-II) and the existence of a Strategy to make the service affordable for everyone (EC3-III), no information was available for the assessment.

Institutional dimension
In the assessment of the Institutional capacity at the utility level (I1), Roles and responsibilities of the staff (I1-I) were quite clear (score 3) and certain Procedures were documented (I1-II: score 2) to support the staff. However, there were no clear Goals of the organization (I1-III) or Long-term goals (I1-IV) (score 1). The assessment of the Institutional capacity at higher levels (I2) revealed certain results that were different compared to Case study 1. The key institutions are the same on the regional level and the national level, i.e. the Regional Environmental Authority in Cochabamba and the Bolivian Ministry of Environment and Water, and at the municipal level it is the technical services department in the area in which WWTS-2 is situated. In this case study, the clarity of Roles and responsibilities (I2-I) and External support (I2-II) was very low (score 1), both at the municipal and national level. However, at the regional level, Roles were clear (I2b-I) and External support was good (I2b-II). As the Regional Environmental Authority requested monitoring of the results of the WWTS-2 to check whether the discharge requirements were being fulfilled, both factors scored 3. This is in contrast to WWTS-1, which never received such a request from the Regional Environmental Authority. The Staff requirements (I3) indicator in terms of Availability only scored 3 for Operator (I3a-I). The Level of expertise required was met for the Engineer (I3b-II) and the Accountant (I3d-II) and they were both assigned a score of 3. There was no Long-term plan to improve the situation for any of the staff categories (I3-III: score 1). Employee satisfaction (I4) in terms of Working conditions (I4-I) and Salary level (I4-II) was low (score 2). As regards Communication (I5), customers are informed about proper Use of the sewer network (I5-I) on certain occasions (score 2). However, no open information is provided to the customers regarding Wastewater utility performance (I5-II) and there were no Warnings about exposure risks to customers (I5-III) when there is a discharge of poorly treated or untreated wastewater, and consequently both scored 1. Indicators in the dimensions are interconnected and influence each other, and all dimensions are therefore important if wastewater is to be treated successfully.

Results of the sensitivity analysis
The results of the sensitivity analysis with two decimal numbers are summarized in Table 5. The different weighting scenarios do not produce a significant difference in outcome of the sustainability assessment compared to the base case.  4. Discussion

Case studies and sensitivity results
The assessment results for Case studies 1 and 2 were quite different in terms of sustainability level. The performance of the system in Case study 1 is quite good although there are still certain issues that need to be addressed, whereas the results for Case study 2 show a WWTS with a low sustainability level and multiple issues that need to be addressed. In Case study 1, it is possible to identify actions that need to be taken in each dimension, but for Case study 2, where low scores are so numerous, it is difficult to prioritize the actions that need to be taken. Many of the issues for Case study 2 are linked to problems in the technical and environmental dimensions, i.e. lack of monitoring of effluents, low removal efficiencies, and non-compliance with discharge or safe reuse limits. The primary function of a WWTP is to protect human health and the environment downstream and both technical and environmental indicators reflects how well the WWTP achieves this. Even if all other dimensions score high and the overall system would get a relatively high integrated score, it is important to also achieve sustainability within each dimension as e.g. the economic aspects included in the tool cannot directly compensate for e.g. negative environmental effects downstream the WWTP. For Case study 2, it is important to communicate the present issues to higher-level institutions and possibly ask for support.
Both cases received a low score (overall 2) for the indicator Sewer network functionality, which is in line with results from the previous indicator study where the sewer network was highlighted as important and an often weak link (Cossio et al., 2020). It is indeed important as any wastewater that does not reach the WWTP will not be treated and any extra water that enters the collection system as a result of infiltration and rainwater inflow will decrease the removal efficiency of the WWTP. In both case studies, the environmental indicators Energy reuse and Global warming potential received very low scores. It is understandable that these types of global environmental targets have lower priorities in the light of the more direct operational problems these types of WWTSs are facing. In the long run however, there is a need to promote awareness of these environmental concerns also in low and lower-middle income countries and identify opportunities for improvement. Also, a town where infrastructure investments are still in the future has unique possibilities to avoid the unsustainable solutions of those who chose their wastewater systems 50 or 100 years ago.
In Case study 2, data was lacking for the assessment of three factors (Compliance of sludge quality with Safe reuse limits, Payment capacity of users, and a Strategy for raising affordability among users), as reported in the to-do list results (see Appendix C, Table C2). In low and lower-middle income countries in general, the lack of proper recordkeeping in small WWTSs is a common issue. In the EVAs tool, a score of 0 is assigned to a factor when there is no information available, which will decrease the overall sustainability score. For Case study 2, if all three scores of 0 would be replaced by the highest score of 4, the overall score would change from 1.6 to 1.8, and the economic dimension would shift from 1.7 (yellow) to 2.6 (light green). This would not however, change the overall impression that there are several areas to improve regarding WWTS-2. For Case study 1 on the other hand, if all factors that involves monitoring of chemical data (T1-I Monitoring of removal efficiency; E1-II Monitoring of safe reuse; E2-I Monitoring of eutrophication potential and; E3-I Monitoring of effluent quality) were scored 0 instead of 3, the overall sustainability score would decrease from 2.9 to 2.4 and would turn yellow instead of light green. In theory, there may thus be systems that work rather well, but still receive an overall low score in the EVAS tool if there is no chemical data available. However, we argue that an important part of a long-term sustainable system is that managers are aware of the actual status of the system by regularly collecting relevant data, and that it is difficult to manage a system optimally in the long run if there is no knowledge of the actual status of the effluent. One of the objectives of the tool is to build up a body of knowledge among managers and stakeholders about the WWTS, and by clearly indicating that there is insufficient data, they can become more aware and motivated to systematically collect data.
The simplified sensitivity analysis used to investigate different weighting schemes indicates that the tool is not very sensitive to nonextreme changes in weights. This is because the tool includes several factors and sub-indicators in the assessment, and thus the overall result is an integrated number not easily affected by quick fixes, such as slightly adjusting weights. Instead, systematic improvements of the WWTS on multiple indicators will have a greater impact on the overall results of the sustainability assessment.

Potential improvements to the tool
Although the general feedback from the application of the tool in the case studies was very positive, and some improvements were carried out as an immediate result of the testing, potential further improvements to the tool were identified. The application of the tool showed in general that the theoretical description of the indicators could to some degree be misinterpreted by the assessor. In contrast, the scoring based on quantitative data, mainly involving the technical, environmental and economic dimensions, was easier to assess in comparison with the qualitative data, e.g. the social dimension. Constructed scoring scales for the assessment of qualitative data can lead to vagueness and misinterpretation if the linguistic terms used are not well defined (Kamble et al., 2017) and there is a need for further development regarding social indicators and the methods to assess them, in order to reduce this type of subjectivity (Padilla-Rivera et al., 2016;Popovic and Kraslawski, 2018). As the personal judgement of the assessors potentially could influence interpretation and score assessment, a further improvement to the tool would be to provide more detailed descriptions of the factors specified in the tool, especially for those with qualitative character. Further specific comments on indicators in each sustainability dimension, except the social dimension, are discussed below.

Technical dimension
The questions and scores could potentially identify seasonal events for the factors Hydraulic loading and Clogging in the Sewer network functionality indicator (T3). The same applies to the Hydraulic loading factor in the Reliability of the WWTP (T4) indicator, since overflow issues mostly occur during the rainy season. It is important that the EVAS assessment reflects how the system copes with the normal and reoccurring conditions and not only those of the week or month just prior to the assessment. As regards the technical factor Organic loading (T4-IV), a suggestion was to ask whether there are small industrial enterprises and, if so, do they pre-treat their wastewater before discharging it into the sewer network. Municipal authorities could potentially regulate this Table 5 Results of the sensitivity analysis showing the overall sustainability scores of the assessments with two decimal numbers for the base case scenario and the three additional weighting scenarios described in methods. The colours correspond to the explanations in Table 2. type of discharge to avoid overloading the WWTP. The assessor in Case study 1 suggested that the assessment of Renewable Energy in (E4), should take account of the size of the plant in order to create a more accurate scoring scale. The potential energy that can be generated from, e.g. anaerobic digestion, is not significant if the utility is small. This factor could also be expanded on by asking whether the plant primarily opts to source energy from renewable sources instead of simply focusing on whether the plant manages to reuse energy generated within its own operations.

Environmental dimension
For the Potential safe reuse (E1) indicator and the factor Safe reuse limits for water, the question should optimally depend on the type of crops for which the treated wastewater is used (i.e. restricted and non-restricted irrigation). In Case study 1 for example, the farmers use the treated wastewater to irrigate corn, which is a non-restricted irrigation crop, and there are thus no limits for safe reuse. However, if the tool is to be updated with a scoring scale for specific crops, the managers will also need to know what crops are irrigated using wastewater, implying more monitoring and more detailed, up-to-date knowledge of agricultural practices in the surroundings.
For Eutrophication potential (E2), it was suggested that this indicator is only important in those cases where the effluent is being constantly discharged into receiving water bodies and that questions and scores should be adjusted to fit site-specific conditions. In Case study 1, all the effluent is used for irrigation during the dry season and during the rainy season it is discharged into a stream. However, even though this first way of managing the effluent will reduce the eutrophication potential in the receiving water bodies, a site-specific assessment is needed to justify deviations from the set discharge standards. Further, in a previous study it was stated that practising wastewater irrigation could give rise to ecological risks associated with accumulation of nutrients in the soil (Cossio et al., 2019). The results suggested that the reuse of treated wastewater should be managed carefully to avoid eutrophication of water sources as a result of accumulation in the soil.

Economic dimension
The Affordability (EC3) indicator and the Payment capacity factor were regarded as being difficult to score. In Case study 1, for example, the municipality had set the charges for the services before the WWTP was constructed based on a prediction of whether the users could pay the tariffs. In addition, the manager of the WWTS-1 stated that in general the population of small towns that have drinking water and sewer network services can afford the tariffs for these services. Consequently, the question concerning if there was a strategy to handle users who cannot pay became redundant. Instead, it was suggested that willingness to pay should be considered and to ask whether the users are aware of the costs for achieving good WWTP performance. Willingness to pay is linked to the level of information among users regarding necessary costs (Birol and Das, 2012;Vásquez et al., 2009) and was suggested to be included in the tool. Affordability for customers is highlighted as an important sustainability issue in several studies (Balkema et al., 2002;Singhirunnusorn and Stenstrom, 2009), and is likely to affect how many people are connected to the drinking water and wastewater treatment systems. Differentiated tariffs, a good level of knowledge of the customers' payment capacity, good communication, and raising awareness among the general public about the benefits of a WWTS, are all interlinked. A study dealing with customers' willingness to pay is thus likely to be dependent on several factors in the assessment tool.

Institutional dimension
Excluding indicator I2 Institutional capacity (higher level) from the assessment was suggested, since the managers do not have control over improving and motivating external stakeholders to carry out their roles. It was stated that it would take a great deal of time and effort to achieve this. On the other hand, the assessment as it is currently being implemented, could help to communicate a lack of institutional capacity at higher levels to induce those concerned to search for strategies to achieve sustainability. In our previous study on sustainability indicators in the context of low and lower-middle-income countries (Cossio et al., 2020), the institutional dimension was highlighted as being of prime importance by local technical and social experts. Indeed, it is difficult to expect that a WWTS will function well without local and institutional support.
4.3. Application of the assessment tool in practice 4.3.1. How can the assessment results be used?
The manager of WWTS-1 stated that "the sustainability assessment tool was very interesting as it can provide us with a final score for the overall sustainability level of the system". On the other hand, the process used to perform the assessment is expected to be even more important than the results as it guides the assessor (s) through the targets needed to achieve sustainability. The assessment results can be used to understand the system and at the same time used to communicate performance and work at the utility, both to the general public and to other institutions that are expected to provide support or to monitor the utility's performance. Repeating the assessment over time is also a means of following up and communicating progress regarding improvements.
The results on the factor level (see Appendix C) provide the assessors with more detailed information about the steps that need to be taken to improve identified weaknesses. Certain actions could potentially be implemented in the short term, e.g. increasing the monitoring frequency or developing a manual of procedures for the O&M of specific technologies (assuming there are short-term resources available). Nonetheless, many of the actions may require a longer timeframe if they are to be achieved. The formulation and implementation of longterm plans for improving the sustainability level typically requires action to be taken in all the dimensions. A wider application of the EVAS tool in a region or in a municipality, could also help authorities to prioritize actions at these levels.

Who should perform the assessment?
Where more than one institution is responsible for the management of the WWTS, they should all participate in the assessment to obtain more accurate answers. In Case study 1, the municipality is responsible for financial management, and the manager was unable to provide complete reassurance that long-term maintenance plans could be implemented for all the treatment units as it would require investment in the infrastructure. The manager was therefore not in a position to take action to improve the situation, as responsibility would rest with external parties. In these cases, the tool and the assessment could function as a means of communicating the difficulties encountered by the WWTS on its path towards sustainability.

How should the assessment be carried out?
We suggest using this assessment within a broader participatory assessment framework (Fig. 4). The suggested framework includes four main stages: i. Identification of assessor or group of assessors. The assessor should preferably be the manager and the staff in charge of the main activities in the WWTS, although the participation from local stakeholders and external experts involved in the WWTS management might also be needed. Assessors should be able to provide information for the assessment and input on the weighting of the variables of the tool (i.e. dimensions, indicators, sub-indicators and factors). ii. Assessment of the current sustainability status of the WWTS using the EVAS tool. If necessary, weights can be modified at this stage, preferably by involving stakeholders. The weighting may be useful to prioritize actions in the resulting to-do list or to discuss the different perspectives among the stakeholders involved in the assessment. iii. Revision and communication of the results from the assessment by managers of the WWTS. Identification of aspects in the WWTS with a low level of sustainability support the prioritization of actions that can be implemented. To facilitate quality assurance, the assessment must be documented fully and preferably reviewed before being communicated. Effective communication of the results to the general public and decision-makers can motivate their support in the process of enhancing the sustainability of the systems. iv. Iteration. The suggested framework indicates an iterative working process as the assessment may require additional data collection or it can be updated as new data becomes available. The assessment can be updated and followed up over time to record improvements in the sustainability of the WWTS.

The complexity of sustainable small WWTS in low and lower-middle income countries
Treating wastewater in a low and lower-middle income country is a service that differs from most other community services, such as water and electricity, and presents additional challenges. Wastewater treatment generally benefits or affects people living downstream of the town depending on the efficiency of the WWTS, whereas the burden of operating and financing the WWTP rests with those living in the town. The benefit of wastewater treatment is only partly apparent, because most of the contaminants are invisible, unless of course the water recipient has a low flow, in which case the contamination is evident. The main benefit to the downstream recipients is also invisible and can only be measured using advanced methods. It may also take time before improvements become apparent. On the other hand, every failure or local nuisance (e.g. smell) caused by the local WWTP will be noted immediately, especially when in a sceptical community. Local benefits, such as reuse of treated wastewater and sludge and the possibility of employment, may be recognized and are important to highlight. In a context where wastewater treatment is a new phenomenon, it could be a challenge to find competent staff for the WWTP and the community may not fully understand how their use of the service has an impact on the sustainable development of their town. It may also be difficult for local, regional and national regulators to understand how much support and regulation is needed to initiate efficient wastewater treatment. As wastewater treatment demands considerable financial investment that generates benefits over an extended period of time, it may be better to initially introduce a simple yet stable treatment process. This could give rise to important improvements and local benefits and provides the possibility of improving treatment over time, rather than expecting an immediate, large-scale investment in technology that cannot be handled correctly or be sustained. The EVAS tool aims to support managers in the complex decisionmaking process as a means of improving current small wastewater treatment services in low and lower-middle income countries sustainably.

Conclusions
In summary, the main conclusions from this study are: • The EVAS tool can provide managers of small WWTSs in low and lower-middle income countries with the means to identify weaknesses in the management of their systems, and to better understand the needs of the WWTS in their endeavour to achieve improved performance and sustainability. • The sustainability level of WWTS-1 was reported to be medium.
Suggested prioritized strategies to bring about further improvements include increasing the monitoring of effluents, improving the sewer network, and following up low levels of removal efficiency as well as the affordability for the wastewater treatment system. • The sustainability level of WWTS-2 is low. Suggested prioritized strategies include increasing the monitoring of effluents, improving the sewer network, and following up on low levels of removal efficiency. However, these must be accompanied by efforts in all sustainability dimensions, including improving the organization of the utility and communicating more effectively with users. • For WWTS-1 where the function as well as the environmental performance was better, also the social and institutional dimensions scored higher than for WWTS-2, indicating an interconnection between dimensions and the need to take a holistic view. • The sewer systems of both WWTSs scored low, and long-term plans to improve the status of the sewer system were not available, potentially causing increased future problems with the sewer systems, increasing bypass of wastewater and hydraulic overloading the WWTP. • The EVAS tool may be used to involve local stakeholders in the assessment and to communicate with decision-makers and external stakeholders to receive better support for identified strategies in order to raise the sustainability level of small WWTSs. • The EVAS tool is also useful for public or regulatory authorities, as they can apply the tool on WWTSs in their region, compare the status of several WWTPs, and based on these results formulate support and investment strategies.
CRediT authorship contribution statement

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements
This study was supported by the Swedish International Development Cooperation Agency (Sida -Styrelsen för Internationellt Utvecklingssamarbete) through the cooperation agreement with San Simon University -Centro de Aguas y Saneamiento Ambiental within the framework of the Integrated Water Resources Management Programme, under Sida Contribution No: 75000554-12. Two anonymous reviewers contributed with constructive comments to improve the manuscript.

Appendix A. Equations and calculations in the EVAS tool
Each factor (f) in the EVAS tool is a assigned a score (S f ) by the user between 0 and 4 according to the traffic light scale, and a weight (W f ) according to the importance of the factor. Standard weights in the EVAS tool are 1 or 2, but any number can be set by the user if motivated: the higher the number the higher importance of the factor (sub-indicator, indicator, or dimension). The normalised weight (w f,si ) of each factor under each sub-indicator (si) is calculated as: The sub-indicator scores (S si ) are then calculated as a weighted sum: Each sub-indicator is also given a weight (W si ) according to the relative importance of the sub-indicator and a normalised weight (w si,i ) is calculated for each sub-indicator included in an indicator (i) as: The indicator scores (S i ) are calculated as: According to the same principle, each indicator is given a weight (W i ), and the normalised weight (w i,d ) of each indicator under each dimension (d) is calculated as: The dimension scores (S d ) are calculated as: Each of the five dimensions is given a weight (W d ), and the normalised weight (w d ) of each dimension is calculated as: Finally, the overall score (S o ) of the EVAS tool is calculated as: Further, in the EVAS tool, there are two types of logical functions related to the calculations of the scores included, listed below.
• If there is no monitoring at all regarding any sub-indicator relating to factors T1-I, E1-II, E2-I, and E3-I, the resulting score of that sub-indicator becomes 0. • If targets are reached for a sub-indicator on a medium or high level, the last factor of that sub-indicator relating to implementation of long-term plans is regarded as not relevant (n.r.) and left out of the assessment. This relates to indicators T1, T3, T4, E1, E2, E3, S1, S2, S3, S4, EC2, I3, I5.
Appendix B. Overview of the content of the sustainability assessment tool    Detailed results from the sustainability assessment for Case study 2, factors that scored 0, 1 and 2 are presented in order, with lower scores and higher weightings first. SOCIAL DIMENSION S1-I 1 2 S1. Public health risk S1-I. Exposure S1-II 1 1 S1-II. Claims S1-III 1 1 S1-III. Long-term plan PH