Overall performance evaluation of an urban water supply system: a case study of Debre Tabor Town in Ethiopia

Population growth, urbanization, economic development, and lifestyle changes in the urban sectors have pressed the demand for water high in terms of both quantity and quality (Domene & Sauri 2006). Climate change and environmental pollution put additional pressure by affecting the accessibility of water resources to meet the growing demand (Danilenko et al. 2010). Thus, available water sources in any part of the world are becoming depleted and/or degraded (He et al. 2021). Water shortage, poor drinking water quality, and frequent interruption of supply in urban areas of developing countries are especially common (Kumpel & Nelson 2016; Salehi 2022). The main reasons for these are a lack of resources, low GDP, polluting industries, an insufficient response from institutions, and unsatisfactory planning and management (Talat 2021). Generally, scarcity of traditional water sources, along with poor service delivery, is increasingly threatening the security of water needs in urban centers of developing countries.

Thus, proper management of the available source, the introduction of alternative sources and surveillance of water quality, and then upgrading the performance level of the water supply system are needed. Performance assessment is a tool that can help decision- and policy-makers adopt what actions they should and should not take in an attempt to make an optimal contribution to sustainable development (Waas et al. 2014). The aims of sustainable development in the urban water sector are to have access to safe, adequate, and clean water to satisfy potable and non-potable needs of consumers at all times, to achieve a reasonable and lasting balance between water supply and demand, to reduce adverse impacts caused by the system to prevent permanent environmental damage, to optimize the allocation of financial resources to make water services profitable and strengthen household affordability (Motevallian et al. 2011).

A sustainability index (SI) is used to indicate the performance of a water supply system concerning predetermined thresholds of a satisfactory state. Thus, sustainability criteria and their indicators are used to quantify the different aspects of utility services and to assess the trend of sustainable development. Generally, performance indicators of urban water supply systems can be grouped as water resources, physical, operational, quality of service, and economic performances. Water resource availability is the main water resource performance indicator. The capacity of storage, quality of the transmission and distribution lines, and the density of the metered customers are taken as physical performance; whereas, loss management, operation, and maintenance, as well as the quality of water supplied, are indicators of operational performance. Water supply pressure adequacy, continuity, water quality and other complaints, and population coverage are quality of service performance indicators (Hashimoto et al. 1982; Kanakoudis et al. 2011; Motevallian et al. 2011; Sandoval-Solis et al. 2011; Rathnayaka et al. 2016).

Hydraulic performance is about the ability to provide a reliable water supply at an acceptable level of service; that is meeting all demands placed upon the system with provisions for adequate pressure and reliability of uninterrupted supply. The index related to hydraulic performance can be described as separate or weighted combinations of reliability, resilience, and vulnerability measures of various economic, environmental, technical, ecological, and social criteria (Loucks 1997). The relative sustainability of the system concerning each of these criteria is higher the greater the reliability and resilience, and the smaller the vulnerability. Numerous investigators have applied these performance indices differently. Hashimoto et al. (1982) analyzed the three measures to evaluate the possible performance of water resource systems. On the other hand, Tabesh & Saber (2011) proposed a model using hydraulic (pressure and velocity performances, and reliability indices), physical, and empirical indices for the rehabilitation of water distribution networks.

Similarly, the water quality index (WQI) is used to evaluate the overall water quality of the system. The physical performance of the water supply system is the ability of the distribution system to act as a physical barrier that prevents external contamination from affecting the water quality. Water quality itself is also an indicator to be used for performance assessment of urban water supply systems, in reference to a set of standards against which compliance can be assessed. On the other hand, producing water and ensuring it gets to consumers require an input of expertise, labor, energy, and chemicals, all of which cost money. Thus, the water supply system should aim to recover this expenditure and gather funds for maintenance, building up of financial reserves, and expansion of the system when the need arises. Full cost recovery ensures the sustainability of the water scheme. Thus, the cost recovery performance of the system should also be assessed. Moreover, for a complete evaluation, it would be ideal to consider customers with their preferences, expectations, and requirements and then explore the ways of meeting their expectations considering the basic water supply service parameters (WHO 2011).

The Ethiopian urban water supply sector operates under several challenges, like geographical location difficulty, rapid population growth and urbanization, and lack of resources for building new water supply facilities. These affect the attempt to provide adequate water of good quality for urban residents in the country. The challenges mainly result in intermittent piped water supply systems in almost all urban centers of the country, and thus Debre Tabor Town is not different (Abel 2020). The utility delivers water to the town once per week and sometimes once per fortnight which lasts for 5–7 h (Abel 2020). This supply system in turn creates a risk to public health due to possible water contamination.

Hence, the preliminary assessment triggered the researchers for a detailed investigation into the extent of the accessibility of sufficient and safe drinking water in the area. In another talk, the present study tried to assess the performance level of the town's water supply system. For this, the hydraulic efficiency of the water supply system was assessed to check if it guarantees the basic requirement. The water quality was also evaluated if it meets the minimum standard. Besides, the cost recovery rate of the utility was assessed as it contributes to the sustainability of the existing schemes and their future expansion. Finally, the consumer satisfaction level was investigated. The results of the study are helpful to understand existing performance levels and then making an informed decision to upgrade the system.

As a result of the 5-year capacity-building project of WaterAid Ethiopia, Debre Tabor Town's domestic connection increased the number of customers from 3,871 to 11,067. Thus, the water supply coverage is raised from 18.75 to 53.6%; the change is impressive. However, only about half of the population is served (WaterAid 2020). This shows that, in addition to the intermittent supply, the supply coverage is limited.

Primary data were collected from sample households using a questionnaire-based assessment of water supply, face-to-face interviews, field measurement and simulation for the distribution system and based on representative samples for water quality analysis. The questionnaire-based data were obtained from households and technical staff of the utility; whereas key informant interviews were conducted with individuals purposively selected from community members and the town water supply office. All the basic service parameters, which are water quality and quantity, accessibility (distance and time to fetch water), water availability, affordability to pay, and the level of continuity of water supply were considered for the survey (WHO 2011). The key informant interview was used in gathering information about administrative, technical, and other attempts made to address issues related to the services of the utility. For the water quality analysis, samples were collected from water sources, reservoirs, and customer taps using polyethylene sampling bottles. The bottles were washed well with boiled water to avoid contamination and labeled carefully. The pressure was measured by a pressure gauge installed at the household taps for calibration. The cost of water production covered by the customers was determined based on the current price of water and the revenue collection efficiency of the utility.

The town administration units were purposefully selected by considering spatial variation (mainly topography) and distance from the service reservoir. Random sampling was preferred to select sample households since household representatives can give relevant information for the study from their experience. Thus, three admin units, namely, units 1, 3 and 6 were purposively selected out of six of the town. Detail of the questionnaires, existing water supply system components and the sampling points is given in the Supplementary Material.

For the assessment of water quality, samples were taken three times from each point in different months (from March to May 2021). Five samples from water sources, three samples from storage tanks, and 20 samples from private taps were collected (Figure 1). The most important tests for water quality surveillance of small communities, namely total coliform, turbidity, residual chlorine, and pH, were used for this study (WHO 1997). Then, laboratory analysis was conducted on the stated physicochemical and microbiological water quality parameters.

In analyzing the hydraulic performance of the water distribution system, AWWA (1999) recommends minimum criteria of 3–6% of nodes for hydraulic network model calibration. There are 144 total numbers of junctions in the network. Thus, using a sample size of 5% of the total junction, seven representative measurements were taken for calibration. To calibrate the model, junctions which were accessible and available for pressure measurement were needed. However, in this case study, the pressure was measured on the faucet which was nearest to the junction as the junction is lined with asphalt and concrete and thus measuring the pressure is difficult without excavating the road. According to Walski et al. (2003), the calibration process was performed by adjusting sensitive parameters, such as pipe type (thus its roughness coefficient), water demand, and pattern of demand until it became within the acceptable limit of 85% of the field test measurements. It should be within 0.5 m or 5% of the maximum head loss across the system.

On the other hand, the model validation involved comparing model results with numerical data collected independently from experiments or observations. The coefficient of determination (R²), which describes the degree of co-linearity between simulated and measured data, was used to manually validate the model. Typically, R² > 0.5 is considered acceptable. To update the system, demand from the current population must be considered; thus, this study used a number of base populations from Debre Tabor Town.

In the current study, the SI assessment using reliability, resilience and vulnerability performance criteria was applied, which are described as:

Table 1 shows the degree of sustainability, which ranges from 0 (lowest degree of sustainability) to 1 (highest degree of sustainability).

For water quality analysis, the weightage arithmetic water quality index (WAWQI) approach was used (Akter et al. 2016). WAWQI method has an advantage over the other methods as it incorporates multiple water quality parameters into a mathematical equation that rates the suitability of water for human consumption and/or also it describes the level of contamination. It is a widely acknowledged method for calculating the WQI. The steps used to determine WQI using WAWQI method are:

Step 1: Data of various physicochemical water quality parameters are collected from lab analysis.

Step 4: The unit weight for each water quality parameter is determined by W_i=

Step 5: WQI is determined, using ⁠. After determining WQI values, the rating of the water quality was stated in reference to Table 2.

GPS was used to obtain the coordinates of points and WaterCAD V8i software was used to simulate the velocity and nodal pressure of the water distribution system. To assess the pressure head of water users and the water flow rate, field measurements were carried out, by using a pressure gauge and from known volume and stopwatch for the time of filling, respectively. To analyze water quality, using the APHA standard method (Baird & Laura 2017), the Debre Tabor Town water supply service lab was used. Sampling bottle, test tubes, beakers, distilled water, taste tubes, spoon, boiling water, incubation machine, turbid meter, pH meter, volumetric flask, pad, filter pad, Petri dish, forceps, microscope coliform counter, and filter membrane were used for the water quality analysis. On the other hand, the Statistical Program for Social Sciences (SPSS) was used to organize and analyze household survey questionnaire data and then to evaluate the overall performance of the urban water supply system. The findings were analyzed in descriptive statistics.

Velocity in a pipe is another important parameter to evaluate hydraulic performance. The country's guideline states velocity to be above 0.6 m/s to prevent sedimentation and bacterial growth in the pipe, and it should be limited to 2 m/s (MoWR 2006). The velocity of Debre Tabor Town's WSD system during peak hour consumption is given in Table 3. The full nodal pressures and line velocities are given in the Supplementary Material.

Based on equations given under the methods section, the performance criteria of the distribution system were determined and shown in Table 4. From the values shown in the table, the sustainability indices for pressure and velocity are 0.614 and 0.132, which are acceptable and unacceptable, respectively. In this study, the overall sustainability score of the water distribution system (average of the two) is 0.373 and it indicates that the system is just moderately sustainable in terms of the variables considered.

After the WaterAid Ethiopia capacity building was provided to Debre Tabor Town, the NRW is reduced from 19.5 to 13.85%, the maintenance cost is reduced to 9.3% of the total operating cost and the revenue collection efficiency reached 98.5%. Currently, the utility sells water to customers in a rising block tariff type on average at 14 birr/m³ and the operating cost of water production is 31.68 birr/m³ (birr is Ethiopian currency, 1 birr = 0.02 USD). Thus, considering the revenue collection efficiency and the NRW, the income collected and the total water production costs are 11.88 and 34.63 birr per m³ of water. This shows that only 34.31% of water production cost is collected from customers.

The majority (94.5%) of the respondents use piped water supply, 2.5% use spring water, and 3% use hand-dug wells. However, the piped water flows in an on-and-off manner for a few hours that customers store water in containers to use until the water comes again. Improper storage may result in contamination (Gizachew et al. 2020). To overcome the existing water supply shortage, the community uses untreated alternative sources of water, such as surface water, spring, and rainwater (in the rainy season), and 76.9% of them hand-dug wells. Thus, due to the intermittent supply of piped water, people are forced to use water from unimproved sources that may be contaminated with infectious microorganisms and may be subject to waterborne diseases (Kumpel & Nelson 2016). Table 6 gives the reasons for limited access to water and equity of distribution.

Households were also asked about the quality of water they use for domestic purposes. The majority of the respondents had good awareness toward their water quality. More than 50% of them feel safe that the quality of water is good for domestic purposes. Some households stated, ‘sometimes there will be sludge at the bottom of water holders when the water is finished’.

The affordability of water has also a significant impact on the use of water and the selection of water sources. Only 68.6% of surveyed householders replied that they pay for the piped water supply service. The payment is entirely in the form of a bill, and the majority of respondents pay between 30 and 200 birrs per month per household. The households were also asked about their judgments on the town's piped water tariff. 13.8, 34.2, 15.6, and 5% of the respondents stated the piped water supply tariff to be cheap, fair, expensive and very expensive, respectively. However, according to the interview with the utility officials of the town, the piped water supply tariff is low to cover the maintenance and operation costs. This agrees with the comparison made between water production and revenue collection. The results of the household survey showed that the overall satisfaction level of the existing water supply in the town is low, as 62.6% of them were not satisfied, and 35.9% are at intermediate satisfaction level and only 1.5% stated that they are satisfied by the town's water supply service.

In agreement with the results of this study, Ebsa et al. (2022) showed only 39.9% of nodes of Jaldu town of Ethiopia have a pressure range specified by the standard. As pressure lower than the specified standard can result in a reduction in the amount of water supplied to the consumer as well as the introduction of contaminants or self-deterioration of water quality within the network itself, causing harm to public health, attention should be given to the related nodes. The velocity assessment also showed that only a limited number of the pipes were within the standard, as 86.7% of the velocity is not within acceptable limits, which may affect the water quality. This service delivery can be considered a failure as it is below 50% to call it at minimum standard service.

The system needs solutions such as pump operational changes, providing pressure control valves, and/or adjustment of network components such as a booster pump installation and replacement of old pipes with new ones of proper diameter to operate within required pressure and velocity and then to upgrade the service level. By using a similar approach to this study, Ebsa et al. (2022) also found an overall SI of 0.399, based on the values of the technical performance indicators pressure and velocity. Moreover, the study by Adank et al. (2016) on water and sanitation service levels of 16 small and medium towns in four regions of Ethiopia showed that a small minority of households receive services, with characteristics such as reliability, quality, quantity, and accessibility of water meeting the standards set by the government.

On the other hand, the water quality-related finding indicated that most of the town's domestic tap waters did not have proper chlorination systems. This result agrees with the general national drinking water quality (CSA 2017). Thus, the drinking water may develop waterborne diseases such as amoeba and giardia, which are a matter of public concern. The study suggests the addition of adequate chlorine, say by setting new re-chlorination points on the distribution system, to kill pathogens and ensure public health. The majority of the water samples had high counts of total coliforms. Similar to these results, the drinking water supply of Kobo town, which is from a borehole through pipes, showed coliform counts not meeting the national and WHO standards (Sitotaw & Nigus 2021). The zero or low value of the residual chlorine effect, the age of the pipes, and the intermittent water supply system are possible causes of the contamination (Kumpel & Nelson 2016; CSA 2017). According to verbal information gathered from the health center of the town, the top three diseases in the area are water-related and waterborne. Gizachew et al. (2020) showed in their water quality study of the Wolaita zone (Ethiopia) that 44% of water source samples and 91% of household water samples were positive for coliform. The water sources had fecal coliform means, and noncompliance with the WHO water quality guideline and they recommended capacity building for the protection and management of water sources and safe water handling practices at the end of use.

Out of the 20 tap water sampled, more than 50% are either poor or totally unfit for drinking purposes. Generally, the results of the WQI indicated that the water quality gets deteriorated from the source to the reservoir and then to the tap, signifying the entry of pollutants on the way. The finding agrees with a similar study conducted in Bangladesh, in which the majority (67%) had poor-drinking water quality (WQI value > 100) and the quality of the drinking water was found to be unsuitable for drinking (Akter et al. 2016). Similarly, Addisie (2022) studied the water quality of the Sidama district, which is close to the current study area, and found WQI values for both improved and unimproved sources in the rank of unsuitable for drinking purposes.

Coming to the financial issues, the 34.31% cost recovery of the utility agrees with ADB (2020), which concluded financial viability to be the weakest performance indicator in the region. If a water supply service could not recover its costs, this would restrict chances to develop and extend networks; particularly affecting the poor. For domestic users, the price is decided considering affordability, which varies depending on the local situation, but for water to be affordable; its price should not exceed 5% of the income of households. In small towns in developing countries, water has to be made accessible and affordable for them at the cost of the utilities not covering expenses. That is, water resource management policy requires social tariffs to be set up for poor communities, based on the recovery of operation and maintenance costs only. The principle behind such a policy is that each person should have the right to access a minimum level of water supply (MoWE 2013).

There are many reasons for tariffs to fail or at least, tariffs not to provide an adequate level of financial sustainability for a utility's operations. Among them is ‘the imperative to suit local needs’. Short- and long-term recommendations to overcome this challenge are to decide how the structure should meet local needs and to evolve the structure proportionately to socioeconomic development (Yates et al. 2020). Besides, for the economic feasibility of the system, in addition to lowering maintenance costs by both preventive and corrective actions and increasing revenue collection efficiency, the strategic business plan needs to be in place by utilities of developing countries. This plays a significant role to overcome the challenge of the limited willingness and capacity of consumers to cover the basic costs. For instance, Ambo town of Ethiopia has built a multi-story building, the ground floor of which is rented to a bank, and playing a huge role in making the municipal water supply sustainable (WaterAID 2020).

The household survey's result confirmed water scarcity in the area. Water is unequally distributed due to line extension issues, and the town's topography occasionally causes WSD failure, based on responses to both questionnaires and interview surveys. Steady and rapid spatial expansion of the built-up area crossing the existing municipal boundary of the town further impacted the pipeline extension management, which in turn influenced meter-connection. The most dominant factor affecting the water consumption of inhabitants of the town is the quantity of water as confirmed by both interview and questionnaires results. The actual production of piped water is significantly lower than the demand due to under-capacity production and frequent interruptions in well operation. Besides, the limited number of boreholes and a high percentage of water loss have further reduced the actual amount of water made available. The majority of the households depend on 5–20 L/capita/day, which is in the basic access service level. Even though distributing water through a house connection use is obviously the most convenient system of water supply for households, the related installation is costly, which most of the households in the community, especially the poorest of the poor, could not afford. Most parts of the peripheral built-up areas of the town are currently beyond the reach of pipelines; thus, those living in these areas suffer greatly from the complete lack of water supply around their homes. This confirms the limitation in equitable access to a potable water supply.

Generally, it is assumed that the potential for long-term performance improvement in urban water supply systems can be established through the application of all the indicators studied. Water quantity and quality should be together studied and managed to attain a better water supply service for urban areas. It should also be emphasized that cost recovery and financial capability and fairness and effectiveness (affordability) in drinking water supply need to be balanced in developing countries (Debel et al. 2020). Simply, trade-offs are expected.

The present study assessed the performance level of Debre Tabor Town's water supply system, based on the hydraulic efficiency of the distribution system, water quality, comparison of water production cost and revenue collection and customer satisfaction. The SI of the distribution system was on average 0.37 that the system is moderate and needs pump operational changes and additional network components to upgrade its hydraulic performance. The WQI results showed that most of the tap water was below standards for drinking purposes set by both WHO and Ethiopia. The intermittent water supply systems may contribute to poor tap water quality. Debre Tabor Town's per capita water consumption was also lower than the regional and country-level standards set by the country's ministry of water resources and WHO domestic water standards. On the other hand, only 34.31% of water production cost is covered by revenue collected from customers, showing that the system is currently far from being economically viable. Generally, the overall performance of the system is weak and improvement of the coverage, quality and thus satisfaction of the community needs due attention. Using additional alternative sources of clean water, reducing water loss, and expanding distribution networks in areas where it is not available are required to upgrade the system's performance. Developing an applicable business plan is also recommended to achieve economic capability. Debre Tabor Town is a fast-growing town and the distribution network expansion service needs to be widely available. The utility should also provide customer awareness programs to limit illegal connections and other sources of non-revenue water. Generally, system-based improved water management approaches in which the factors are considered in combination, are required to realize sound performance in the long-term.

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

The authors declare there is no conflict.