Measuring the environmental performance of urban water systems: a systematic review

In recent decades, environmental issues have been more comprehensively discussed to identify actions and strategies for the monitoring and reduction of environmental impacts from anthropological activities. Moreover, governments, researchers, entrepreneurs, and consumers participate in these discussions. Water supply and sanitary sewage services are essential for economic activities, the preservation of the environment, and the well-being and health of the public. However, utility companies continue to face challenges that complicate the sustainability of their operations, such as extreme weather events, ageing infrastructure, population growth, increasing coverage, and the need for improved quality (Alegre et al. 2006; EPA 2018; Xue et al. 2019).

Performance evaluations can be used to address some of these challenges. A system of indicators can be adopted to measure the quality, efficiency, and effectiveness of management operations; this can serve as a tool for planning and expanding services (Alegre et al. 2006; Vilanova et al. 2015).

In addition to measuring operational indicators, urban water systems (UWSs) can be evaluated using multiple environmental and sustainability parameters, thus allowing for the provision of water and sanitation services that can promote the environmental, economic, and social health of communities (Behzadian & Kapelan 2015a; Xue et al. 2019; Landa-Cansigno et al. 2020). The growing stakeholder interest in environmental performance has caused numerous organizations to establish procedures for monitoring and measuring the potential environmental impact of their activities, as guided by environmental management systems (ISO 2015; Galant & Cvek 2021).

Indicators are used in environmental performance assessment to determine the performance of an organization based on its environmental benchmarks over different time periods (ISO 2013). Recent studies show that investments in environmental performance contribute to improved financial performance (Lee et al. 2014; Galant & Cvek 2021) and increased organizational performance (Huynh 2020) and positively affect the competitiveness of companies (Martínez & Poveda 2022). Moreover, the identification and quantification of environmental impacts related to UWSs provide significant information to the formulators of management instruments and public policies related to the sector (Loubet et al. 2014).

The environmental performance of UWSs has been reviewed in the past; however, these studies selected more specific subjects, such as those related to the water sector in South Africa (Buckley et al. 2011), the environmental burden of wastewater treatment systems (Corominas et al. 2013), or the environmental impact of sewage treatment plant sludge (Yoshida et al. 2013). Some reviews have also considered all stages associated with the water system; however, they reviewed the application of a specific environmental impact quantification methodology, life cycle assessment (LCA) (Loubet et al. 2014; Byrne et al. 2017). No review has identified and analyzed publications related to measuring the environmental performance of UWS broadly.

Therefore, this systematic review aimed to elucidate the different methods of measuring the environmental performance of UWSs and to identify the main environmental impacts of UWS operations and their respective sources. In this review, UWSs refer to an integrated set of stages comprising water collection, water treatment, water distribution, sewage collection, and sewage treatment.

A comprehensive literature review was conducted based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. This method contributes toward transparent communication regarding the motivations for conducting a systematic review, the process by which it was performed and the results (Page et al. 2021). The PRISMA checklist is given in the Supplementary Material. The Scopus and Web of Science databases were selected as the sources of the articles. Scopus was selected for its broad coverage of management and engineering journals (Ahi & Searcy 2013), and Web of Science was selected for its focus on management (Shepherd & Günter 2006).

Our research questions were as follows:

• (1)

  How is the environmental performance of UWSs measured?

• (2)

  What are the main environmental impacts analyzed and at which stages of the system do they occur?

The terms used in the database searches were based on the questions above and are presented in Table 1.

Articles were screened based on the eligibility criteria developed for this study as presented in Table 2.

Information related to the general data (year of publication, journal, and country of origin) and specific characteristics (methodological/practical contribution, methods used, impacts analyzed, and characteristics of the systems studied) were extracted for each publication. The record of the extracted data, as well as their organization, summarization, and creation of graphics, was performed in Microsoft Excel 365.

The initial studies related to the measurement of the environmental performance of UWSs included methodological contributions, where frameworks and models were presented and environmental performance indicators were suggested. From 2010 onwards, the publication of case studies that quantified the environmental impacts of these systems began.

The Journal of Cleaner Production had the highest number of publications in this field with four publications, followed by Water Research, Science of the Total Environment, and Resources, Conservation and Recycling with three publications each. However, the Journal of Cleaner Production and Science of the Total Environment published only case studies, whereas Resources, Conservation and Recycling and Water Research published both studies with methodological contributions as well as case studies. The number of publications in the other journals was lower and comprised a single study each.

Among the continents, Europe had the largest number of publications (11 publications), followed by North America, Latin America, Asia, and Oceania with three publications each and Africa with two publications. Developed countries made the greatest contribution to the total number of publications (17 publications), while underdeveloped or developing countries contributed eight publications in total.

Table 3 summarizes the 25 publications that were reviewed, along with a brief description of their objectives and main results. In this review, the studies were classified as ‘studies with methodological contributions’ or ‘case studies based on the type of contribution’.

Hellström et al. (2000) proposed the first specific framework for measuring the environmental performance of UWSs to analyze and compare their sustainability criteria. This study is based on the separation of the water system into modular blocks that can be combined in different ways, making it possible to obtain different types of water management systems. The study analyzed multidimensional criteria involving economic, environmental, social, cultural, technical, and health aspects. For each of these aspects, the study identified quantifiable indicators that allow for analysis and comparison between the systems (Hellström et al. 2000).

Subsequent studies aimed to assess the UWS using the LCA framework. These frameworks were proposed for the evaluation of the environmental sustainability of UWSs (Lundin & Morrison 2002) or, in a broader sense, of any urban infrastructure (Sahely et al. 2005). According to these analyses, the most relevant environmental sustainability indicators for UWSs included electricity consumption, chemicals, greenhouse gas (GHG) emissions from wastewater treatment, nutrient discharge, sludge disposal, and nutrient and energy recycling. In addition, other studies proposed specific models for the simulation and analysis of UWSs. These models were based on the modeling of dynamic systems (Fagan et al. 2010), mass and energy balances (Behzadian & Kapelan 2015b), and object-oriented programming (Loubet et al. 2016b) to provide performance metrics, aid in the analysis of UWSs, or reduce their complexity.

The framework for modeling dynamic systems proposed by Fagan et al. (2010) was developed and linked to an LCA tool. This can provide a set of metrics that integrate all subsystems of the water cycle. The dynamic balance between the materials and energy, thermodynamics, and kinetics associated with an LCA and the economy of the process enables the analyzed variables to be calculated simultaneously throughout the analyzed system. The model was tested using a case study that compared the use of the framework at three scales of infrastructure. Its dynamic nature allows the water system to be designed for extreme events and assists in searching for alternatives that reduce its environmental impacts. The model's application also determined that the GHG emissions from the operation of the UWS were equal to or greater than those of the infrastructure; therefore, actions to improve the environmental performance of the UWS must first focus on the operation phase.

Behzadian & Kapelan (2015b) proposed a conceptual model based on mass balance called WaterMet2. This is a dynamic model of mass flow analysis that allows the simulation of the flows related to the UWS with daily intervals and an analysis of the environmental impacts over a long-term duration. This model proved to be useful to measure the performance of the UWS considering different system configurations in the long term. Its application demonstrated that wastewater treatment plants were identified as the main sources of GHG emissions and acidification potential; but the source of eutrophication was shared between the wastewater treatment plant and the wastewater collection network.

The WaLA model proposed by Loubet et al. (2016b) used generic components to represent the UWS units, their users, and associated water flows and impacts. The model allowed the construction of analysis scenarios that connected technologies and users in a modular and integrated manner, and this enabled the model to calculate the impacts of the life cycle of the analyzed system. The model was tested through a virtual case study and demonstrated that systems can easily be built using the modular approach. The results indicated a significant contribution of wastewater treatment plants to local environmental impacts, such as eutrophication; but supporting activities, such as electricity, chemicals, and infrastructure, contributed toward global environmental impacts.

Table 4 classifies the case studies according to the methods used and the characteristics of their application.

The studies aimed to analyze and identify the environmental impacts of UWS (Friedrich et al. 2009; Amores et al. 2013; Lemos et al. 2013; Slagstad & Brattebø 2014; Alanbari et al. 2015; Jeong et al. 2015; Loubet et al. 2016a; Sambito & Freni 2017; Boldrin et al. 2022). In addition, certain studies assessed the environmental impacts to improve the environmental performance (Mahgoub et al. 2010) by comparing two urban-scale water systems (Lane et al. 2015) and analyzing the environmental impacts of novel strategies and technologies (Liu et al. 2016; Hsien et al. 2019; Landa-Cansigno et al. 2020).

Certain studies have complemented the assessment of the environmental impacts of UWSs by including assessments of the social (García-Sánchez & Güereca 2019) and economic (Behzadian & Kapelan 2015a; Stanchev & Ribarova 2016; Xue et al. 2019) dimensions; these studies provide a more comprehensive analysis of UWS performance.

The quantification of the environmental impacts of UWSs was conducted using the LCA and mass balance methodologies.

Some publications applied the WaterMet2 model to analyze UWSs (Behzadian & Kapelan 2015a; Landa-Cansigno et al. 2020). Issues related to the strategic planning of UWSs were analyzed by Behzadian & Kapelan (2015a); they found that conventional and sustainability indicators are necessary for the strategic planning of UWSs. In addition, Landa-Cansigno et al. (2020) considered a range of centralized and decentralized water reuse strategies. They identified that decentralized water reuse strategies yield more optimized performance with respect to water conservation, GHG emissions, and eutrophication indicators, while energy savings are negligible. They also concluded that centralized strategies could achieve more optimized energy-saving performance.

The LCA can assess material input and output flows throughout the life cycle of a given product and estimate its potential environmental impacts (ISO 2006). It offers a holistic approach for the evaluation of UWSs and its analysis should include the construction, operation, and decommissioning stages of the facilities (Loubet et al. 2014). However, this is not always possible due to the lack of actual data. In addition, the use of databases is limited, as in some cases, researchers use data from other regions.

Regarding the operational process of UWSs, the water withdrawal, water treatment, water distribution, sewage collection, and sewage treatment stages were considered in all case studies except that performed by Friedrich et al. (2009), which did not include the water withdrawal stage.

Some system stages, such as water use by consumers (Lundie et al. 2004; Loubet et al. 2016a; García-Sánchez & Güereca 2019), rainwater (Jeong et al. 2015; Loubet et al. 2016a; Xue et al. 2019), and water recycling processes (Friedrich et al. 2009; Hsien et al. 2019), were relatively rare in case studies. Moreover, the environmental impacts related to UWS infrastructure were not included in all studies. However, infrastructure can be considered in its entirety (Lundie et al. 2004; Friedrich et al. 2009; Jeong et al. 2015; Lane et al. 2015; Loubet et al. 2016a; Xue et al. 2019) or partially (Amores et al. 2013; Lemos et al. 2013; Slagstad & Brattebø 2014; Alanbari et al. 2015; Boldrin et al. 2022).

The analysis of UWSs using the LCA must include a volume metric as a functional unit (Loubet et al. 2014; Byrne et al. 2017). In the analyzed publications, the functional units varied among the supply or distribution of 1 m³ of drinking water (53% of all publications), the supply of 1 m³ of drinking water or the treatment of 1 m³ of sewage (16% of the publications), one year of water supply and collection, the transport and treatment of wastewater (16% of the publications), and the number of households served (5%). Two publications did not report the functional units (11% of the publications).

In addition, several publications were included in the scenario analysis considered in the present review, which aimed to elucidate the environmental impacts resulting from changes in system operation or infrastructure. The alternative scenarios were proposed considering mainly changes related to the reduction of water losses during distribution, seawater desalination processes, water recycling processes, adoption of new technologies for sewage treatment, changes in the mix of electric power used by the systems, and an increase in service demand and provision (Lundie et al. 2004; Mahgoub et al. 2010; Amores et al. 2013; Lemos et al. 2013; Lane et al. 2015; Liu et al. 2016; Loubet et al. 2016a; Hsien et al. 2019; Boldrin et al. 2022).

Methods used in the life cycle impact assessment (LCIA) stage may consist of a single impact category analysis, such as cumulative energy demand, water footprint, carbon footprint, freshwater ecosystem impact (FEI), and water stress index (WSI). The cumulative energy demand, carbon footprint, and water footprint methods quantify the direct and indirect amounts of primary energy, carbon equivalent, and water that are used in certain activities, respectively (Jungbluth & Frischknecht 2007; Hoekstra et al. 2011). Water consumption is also evaluated using the FEI method, which assesses the ecological consequences of water use in a region, and the WSI method, which assesses the environmental impacts of freshwater consumption, and considers damage to human health, ecosystem quality, and resources (Milà i Canals et al. 2009; Pfister et al. 2009).

Other methods have multiple impact categories that can be selected according to the available input data and purpose of the analysis, such as ReCiPe, CML, Eco-indicator 99, Impact2002 + , TRACI, and the Uniform System for the Evaluation of Substances for LCA Purposes (USE LCA 2.0). The assessment of potential environmental impacts may consider an approach at the midpoint level, when the characterization of the environmental impact is performed along the environmental mechanism or endpoint, and when it is considered as an aspect of the natural environment, such as human health, ecosystems, and resources (ISO 2006).

The CML and Eco-Indicator 99 methods were developed in the Netherlands, and while the CML presents a midpoint-level approach to assess environmental impacts, the Eco-Indicator 99 method sought to simplify the interpretation and weighting of LCA results through an approach endpoint (Guinée 2001; Joint Research Centre 2010; Mendes et al. 2006). The TRACI method consists of a method for assessing potential environmental impacts with a midpoint approach that considers the environmental conditions in the United States (Bare et al. 2003). The ReCiPe method was developed to integrate and harmonize the midpoint and endpoint approaches and consists of a continuation of the CML and Eco-Indicator 99 methods (Goedkoop et al. 2009; Joint Research Centre 2010). Methods such as Impact 2002+ and USES LCA 2.0 also integrate midpoint- and endpoint-level approaches into their impact categories. However, while Impact 2002+ allows for the analysis of 14 impact categories, the USES LCA 2.0 permits the quantification of environmental impacts related to ecotoxicity and human toxicity (Jolliet et al. 2003; van Zelm et al. 2009).

The only impact category analyzed in all publications is climate change, which is also referred to as global warming in some LCIAs. Analysis of eutrophication occured in 17 publications, while acidification and human toxicity were analyzed in 15 and 14 publications, respectively. Ozone depletion, photochemical oxidant/ozone formation, and ecotoxicity were analyzed in 13 publications. Fossil depletion and particulate matter were analyzed in 11 and eight publications, respectively. The other categories were analyzed less frequently as shown in the figure.

The most commonly identified sources of the environmental impacts of UWSs were electricity consumption, nutrient discharge into the environment, and chemicals used in the UWS.

The water collection and distribution stages, sewage collection and treatment, and seawater desalination process consume most of the required electricity (Friedrich et al. 2009; Mahgoub et al. 2010; Amores et al. 2013; Lemos et al. 2013; Alanbari et al. 2015; Lane et al. 2015; Loubet et al. 2016a; Stanchev & Ribarova 2016; Sambito & Freni 2017; Hsien et al. 2019; Xue et al. 2019; Boldrin et al. 2022). In addition, the use of water from underground springs or the pumping of water to cover the extensive distances between the points of collection and consumption increases the electricity consumption of the systems (Lundie et al. 2004; Amores et al. 2013; Alanbari et al. 2015; Jeong et al. 2015; Sambito & Freni 2017; García-Sánchez & Güereca 2019; Xue et al. 2019; Boldrin et al. 2022).

The environmental impacts arising from electricity consumption are a function of the energy mix used by each country or region. Therefore, the adoption of an energy matrix with higher participation of renewable energy can contribute considerably to reducing the environmental impacts of UWSs (Lemos et al. 2013; Sambito & Freni 2017; Boldrin et al. 2022). Furthermore, solutions such as the use of photovoltaic panels and energy recovery in wastewater treatment plants can also contribute to reducing these impacts (Friedrich et al. 2009; Boldrin et al. 2022). Another important factor that can help reduce energy consumption is the reduction of losses in distribution systems, as a considerable amount of energy is consumed to capture, treat, and distribute water that is not consumed (Lemos et al. 2013; Sambito & Freni 2017; Boldrin et al. 2022). According to Friedrich et al. (2009), the reduction of losses not only contributes to the advantages associated with the treatment and distribution of water but also prevents unnecessary wastage of water.

With respect to sewage treatment plants, the treated effluents are discharged into aquatic environments and excess nutrient discharge can result in the eutrophication of the aquatic environments (Lemos et al. 2013; Jeong et al. 2015). The application of treatment plant sludge to arable soils reduces the need for synthetic fertilizers, which leads to savings in energy, raw materials, and emissions from the production of these fertilizers. However, the application of sewage sludge increases the impact of potential eutrophication (Amores et al. 2013).

Unlike the other publications, García-Sanchez & Güereca (2019) reported that the sewage treatment plant presented the lowest system environmental impact owing to the low power consumption of the analyzed sewage treatment plant and the fact that the study used the Net Environmental Benefit (NEB) equation presented by Godin et al. (2012), which evaluates the impact avoided by the removal of pollutants in treated effluents. The study conducted by Boldrin et al. (2022) identified a significant contribution of the wastewater treatment plant in the impact categories of climate change and photochemical oxidant formation, resulting from methane emissions from the lagoon system used in the wastewater treatment plant.

Although electricity consumption and nutrient discharge contribute significantly to the environmental impacts of the analyzed systems, chemicals were notably responsible for significant environmental impacts (Stanchev & Ribarova 2016; Xue et al. 2019; Boldrin et al. 2022). Lundie et al. (2004) identified that chemicals contributed considerably to the total energy consumption indicators as well as to the categories of climate change and photochemical oxidant formation. Amores et al. (2013) found that chemicals contributed directly to the impact categories related to cumulative energy demand and stratospheric ozone destruction.

Regarding the infrastructure of UWSs, Lundie et al. (2004) concluded that the operational phase of the system is considerably more significant for environmental impacts than the infrastructure. However, subsequent studies have found that UWS infrastructure contributes significantly to the environmental impacts analyzed and therefore should be considered in the analyses whenever possible (Slagstad & Brattebø 2014; Jeong et al. 2015; Lane et al. 2015; Loubet et al. 2016a; Xue et al. 2019).

Significant variations related to the degree of contribution of each system step in environmental impacts may be associated with the characteristics of systems (the type of system or technology used) or the level of detail of input data used in the analysis. When actual data are unavailable, numerous researchers use simplifications or data from previous studies to compose the LCA inventory. Such simplifications limit the method, which may not correctly express the impacts related to the analyzed system.

Furthermore, studies related to measuring the environmental performance of UWSs are carried out individually to analyze a specific system and methodological variations in its application complicate comparative analysis between systems.

We conducted a systematic literature review to identify and analyze studies that measured the environmental performance of UWSs. Publications with two distinct contribution types such as methodological contributions and case studies were identified. The former included studies that contributed toward the development of frameworks and models, identifying relevant indicators to measure the environmental performance of UWSs. The case studies conducted an applied analysis of the environmental performance of the UWSs.

Our results identified models based on mass balance and LCA. However, the high usage of LCA in the analyzed publications suggests that it is the most accepted method for measuring the environmental performance of UWSs. The classification of the studies indicated that, within the application of LCAs, the most frequently used LCIA method was ReCiPe and the most frequently analyzed impact categories were climate change, eutrophication, acidification, human toxicity, photochemical oxidant/ozone formation, ecotoxicity, and fossil depletion.

The findings of the present review form the basis for future related studies that intend to develop new methodologies or tools to measure the environmental performance of the UWS. Moreover, they can aid in formulating plans and policies to improve the environmental performance of the UWS.

M.T.N.B. is thankful to the Coordination for the Improvement of Higher Education Personnel – Brazil (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – CAPES) for providing a doctoral scholarship grant.

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

The authors declare there is no conflict.