Carbon Footprint Drivers in China’s Municipal Wastewater Treatment Plants and Mitigation Opportunities through Electricity and Chemical Efficiency

Shen Qu; Yuchen Hu; Renke Wei; Ke Yu; Zhouyi Liu; Qi Zhou; Chenchen Wang; Lujing Zhang

doi:10.1016/j.eng.2024.01.021

Engineering ›› 2025, Vol. 50 ›› Issue (7) :106 -116. DOI: 10.1016/j.eng.2024.01.021

Research

Article

Carbon Footprint Drivers in China’s Municipal Wastewater Treatment Plants and Mitigation Opportunities through Electricity and Chemical Efficiency

Shen Qu ^a^,^b^,^#^,^*
, Yuchen Hu ^a^,^b^,^#
, Renke Wei ^c
, Ke Yu ^a^,^b
, Zhouyi Liu ^a^,^b
, Qi Zhou ^a^,^b
, Chenchen Wang ^c
, Lujing Zhang ^d

Author information +

History +

PDF (2331KB)

Abstract

Reducing greenhouse gas (GHG) emissions to address climate change is a global consensus, and municipal wastewater treatment plants (MWWTPs) should lead the way in low-carbon sustainable development. However, achieving effluent discharge standards often requires considerable energy and chemical consumption during operation, resulting in significant carbon footprints. In this study, GHG emissions are systematically accounted for, and the driving factors of carbon footprint growth in China’s MWWTPs are explored. In 2020, a total of 41.9 million tonnes (Mt) of carbon dioxide equivalent (CO₂-eq) were released by the sector, with nearly two-thirds being indirect emissions resulting from energy and material usage. The intensity of electricity, carbon source, and phosphorus removing agent consumption increasingly influence carbon footprint growth over time. Through statistical inference, benchmarks for electricity and chemical consumption intensity are established across all MWWTPs under various operational conditions, and the potential for mitigation through more efficient energy and material utilization is calculated. The results suggest that many MWWTPs offer significant opportunities for emission reduction. Consequently, empirical decarbonization measures, including intelligent device control, optimization of aeration equipment, energy recovery initiatives, and other enhancements to improve operational and carbon efficiency, are recommended.

Graphical abstract

Keywords

Municipal wastewater treatment plants / Carbon footprint / Driving Factors / Mitigation opportunities

Cite this article

Download citation ▾

Shen Qu, Yuchen Hu, Renke Wei, Ke Yu, Zhouyi Liu, Qi Zhou, Chenchen Wang, Lujing Zhang. Carbon Footprint Drivers in China’s Municipal Wastewater Treatment Plants and Mitigation Opportunities through Electricity and Chemical Efficiency. Engineering, 2025, 50(7): 106-116 DOI:10.1016/j.eng.2024.01.021

登录浏览全文

4963

注册一个新账户忘记密码

1. Introduction

In a global society increasingly committed to achieving carbon neutrality [1], [2], [3], the water sector emerges as a foundational pillar of the infrastructural system. This sector requires substantial resource allocation to ensure operational stability. For instance, water and wastewater systems in the United States consume approximately 1%–4% of electricity [4], [5], [6], [7], [8], while in Beijing, China, the water supply chain accounts for 4%–6% of total municipal energy consumption [4], [9], [10]. The urban water sector operates as a closed loop encompassing abstraction, treatment, distribution, use, wastewater collection, wastewater treatment, and discharge [11]. Wastewater treatment plays a critical role in promoting the recycling of clean water resources [12], [13], [14], [15]. Contaminated water originates from various sources, demanding more energy and materials for its treatment compared to fresh water [11]. Wastewater treatment involves highly energy-intensive processes [16], [17], [18], With the treatment of domestic wastewater from daily use alone occupying 3% of global electricity consumption [19]. Furthermore, wastewater treatment ranks as the fourth and sixth highest contributor to atmospheric nitrous oxide (N₂O) and methane (CH₄), respectively [20], [21], with associated processes responsible for 9.6% of non-carbon dioxide (CO₂) greenhouse gas (GHG) emissions worldwide [22], [23]. At the national level, Denmark attributes 10% of its waste sector’s GHG emissions to wastewater treatment [24], [25], a figure that increases to 13% within the broader context of the European Union [24], [26]. Therefore, wastewater treatment will play a crucial role in global climate change mitigation.

In recent decades, China has developed the world’s largest and still expanding municipal wastewater infrastructure [27], [28], driven by rapid economic and technological growth. Since 2007, over 6200 municipal wastewater treatment plants (MWWTPs) have been put into operation across the country, recognized as a significant source of the national carbon footprint. From 2005 to 2014, CH₄ and N₂O emissions from MWWTPs in China increased nearly threefold [29], a trend likely to continue due to increasingly stringent effluent discharge requirements [30], [31]. Thus, urgent measures are needed to mitigate the global warming effects of MWWTPs [22], [32], [33]. In fact, municipal wastewater treatment holds tremendous potential for reducing GHG emissions [34], [35], [36], [37] and is expected to play a more positive role in facilitating carbon neutrality rather than serving as a source of GHG [38], [39], [40]. Understanding the emission patterns and potential for mitigation of China’s MWWTPs is a necessary step towards achieving this goal [41], [42].

While studies focusing on the carbon footprint of MWWTPs have emerged, there are still varying degrees of limitations. Primarily, conducting large-scale and detailed accounting of GHG emissions remains challenging. The most direct and accurate method to obtain the carbon footprint is on-site monitoring. However, GHG emissions are not included in the routine monitoring indicators of most MWWTPs, leading to additional monitoring work and higher costs. The open space and seasonal variations further limit the scope of on-site measurements to individual or a small number of cases [43], [44]. Another approach is theoretical modeling, which can be further divided into mechanistic modeling [45], [46] and indirect estimation [47], [48]. While mechanistic modeling is relatively precise [49], it is difficult to implement due to the complex biochemical reactions involved in the treatment process, thus limiting its research scale. Indirect emission estimations can expand the accounting boundary to a much higher level, but they rely on socioeconomic indicators like population. The resolution and accuracy of such estimations need further improvement.

Moreover, although some large-scale and detailed accountings of carbon footprint have been conducted by overcoming the aforementioned difficulties, the vast amount of information hidden in the data has not been fully extracted, leading to a lack of further in-depth research. Firstly, one of the features of nationwide firm-level GHG emissions accounting is its higher resolution, implying obvious regional heterogeneity caused by the combined effects of socioeconomic and natural geographical conditions. Coupling the distribution patterns of the carbon footprint of MWWTPs with these comprehensive factors will elevate the research perspective to a more macroscopic level, thereby better promoting the integrated development of the plant with its host city [50]. Secondly, benefitting from the extremely large volume of data, the identification of drivers of the carbon footprint of MWWTPs over time can be done more universally and accurately, rather than simply investigating the composition of GHG emissions. Finally, statistical inference based on machine learning has not been adequately applied in these nationwide firm-level carbon footprint accountings to explore the distribution of GHG emissions from MWWTPs. This suggests that assessing the potential for mitigation with various operational indicators is quite time-consuming, and relevant scenario analysis can only be conducted based on relatively subjective assumptions, such as the penetration and diffusion of advanced treatment processes [51], [52]. In fact, assessing the potential for mitigation in MWWTPs under different operational conditions is of paramount importance [53]. Such an evaluation will provide more precise benchmarks for guiding the industry’s low-carbon transition, particularly through the implementation of more intelligent management practices.

Overall, understanding the emission patterns of MWWTPs is challenging due to the complexity of GHG formation and the high monitoring costs [54], [55], [56], [57], [58], [59]. In this study, a data-driven approach is adopted to bridge the gap between existing research and expected objectives. The methodology comprises several essential steps. First, we the monthly GHG emissions of MWWTPs throughout China from 2007 to 2020 are estimated. This is accomplished by employing conventional operational indicators and established emission factors, thereby elucidating the spatiotemporal trends of GHG emissions at a high level of detail. Subsequently, structural decomposition analysis (SDA) is employed to identify the driving factors behind the growth in the carbon footprint of MWWTPs over time, while also scrutinizing the influences of treatment processes and scale. Finally, for the key drivers of GHG emissions that are feasible for engineering practice, quantile regression forest based on the random forest algorithm is adopted to obtain benchmarks for GHG emissions tailored to each plant in China, contingent upon their specific operational conditions. These benchmarks are then compared with actual emissions to gauge the potential for mitigation and empirical strategies aimed at enhancing the energy and materials efficiency of MWWTPs are recommended.

2. Materials and methods

Fig. 1 [61], [122] depicts the framework of this study along with the methodologies used in each section. When calculating the carbon footprint, the approach chosen avoids discussions concerning the complex mechanisms underlying GHG emissions. Rather than solely comparing carbon footprints from various sources, SDA is employed to precisely identify the key driving factors influencing carbon footprint growth. Furthermore, to assess the potential for mitigation in MWWTPs, the quantile regression forest algorithm is used to obtain the distribution of critical drivers influenced by different operational conditions within MWWTPs, focusing on the key drivers of GHG emissions that can be practically managed in engineering.

2.1. GHG emissions accounting

The National Urban Wastewater Treatment Management Information System provides accurate and comprehensive details, including fundamental data and monthly operational information of MWWTPs across China. This information encompasses the name, location, volume, influent and effluent concentrations of chemical oxygen demand (COD) and total nitrogen (TN), electricity consumption, types and consumption of carbon sources, phosphorus removing agents, dewatering agents, and more [62].

On this basis, we account for the carbon footprints of 6228 MWWTPs across China, which had operational records from 2007 to 2020 by month, are accounted for with the emission factors provided by the Intergovernmental Panel Climate Change Guidelines (IPCC) for National GHGs Inventories and existing studies [20], [47], [54], [63], [64], [65], [66], [67], [68], [69], [70], [71], [72], [73], [74], [75], [76], [77], [78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], [92], [93], [94], [95], [96], [97], [98], [99], [100], [101], [102], [103], [104], [105], [106], [107], [108], [109], [110], [111], [112], [113] (Eqs. S1–S14 and Tables S1–S9 in Appendix A). Fig. 1 displays the system boundary for calculating GHG emissions, which consists of: ① direct emission of CH₄, N₂O, and CO₂ during the treatment process, ② indirect emissions of CO₂ during the treatment process due to the energy and chemical consumption, and ③ direct emissions of CH₄ and N₂O from the effluent.

It’s noteworthy that electricity is the primary form of energy used in most MWWTPs [53], [114], so the consumption of other forms of energy is not considered in this study. Furthermore, due to the lack of available data and being beyond the scope of the plant, the carbon footprint related to sludge treatment and disposal is not accounted for. In addition, given that most CO₂ emissions during the wastewater treatment process originate from biogenic sources, while the specific proportion of non-biogenic CO₂ is highly uncertain [47], [53], only the portion brought by non-biogenic carbon source is included and calculated as supplementary COD. Since the removal amounts of COD and TN are derived from the actual monitoring values of MWWTPs, the accounting results of associated carbon footprints are more precise than estimates based on indirect indicators such as population [20].

2.2. Identification of key driving factors

SDA is a common method for analyzing the factors influencing GHG emissions, widely used in previous research [115], [116], [117]. It quantifies the impact of individual factors on the carbon footprint, thereby ranking their importance and determining the critical ones. The principle of SDA is illustrated through considering GHG emissions led by energy consumption as an example, which is shown in Eqs. (1), (2). The carbon footprint caused by electricity usage

E_{C O_{2}}^{EC}

(EC: electricity consumption) is determined by the product of the wastewater volume (

Q_{T r e a t e d}

), electricity consumption per unit volume of treated wastewater (

I_{EC}

), and the emission factor of the power grid

E F_{PG}

(PG: power grid). The change in GHG emissions caused by electricity consumption from year t₀ to year t₁ (

Δ E_{C O_{2}}^{EC}

) is ultimately decomposed into the sum of changes caused by the variation in these three operational indicators during the same stage (

Δ Q_{T r e a t e d}

Δ I_{EC}

, and

Δ E F_{PG}

). Similarly, changes in direct and indirect carbon footprints resulting from chemical consumption over time can be decomposed into the sum of changes caused by variations in wastewater volume, wastewater quality, chemical consumption per unit volume of treated wastewater, and emission factors of chemicals during the same period.

(1)

E_{C O_{2}}^{EC} = Q_{Treated} \times I_{EC} \times E F_{PG}

(2) $\begin{aligned} \Delta E_{\mathrm{CO}_{2}}^{\mathrm{EC}}= & \frac{1}{2}\left(\Delta Q_{\text {Treated }}\right)\left(I_{\mathrm{EC}}^{t_{0}} \times \mathrm{EF}_{\mathrm{PG}}^{t_{0}}+I_{\mathrm{EC}}^{t_{1}} \times \mathrm{EF}_{\mathrm{PG}}^{t_{1}}\right) \\ & +\frac{1}{2}\left(\Delta I_{\mathrm{EC}}\right)\left(Q_{\text {Treated }}^{t_{0}} \times \mathrm{EF}_{\mathrm{PG}}^{t_{1}}+Q_{\text {Treated }}^{t_{1}} \times \mathrm{EF}_{\mathrm{PG}}^{t_{0}}\right) \\ & +\frac{1}{2}\left(\Delta \mathrm{EF}_{\mathrm{PG}}\right)\left(Q_{\text {Treated }}^{t_{1}} \times I_{\mathrm{EC}}^{t_{1}}+Q_{\text {Treated }}^{t_{0}} \times I_{\mathrm{EC}}^{t_{0}}\right) \end{aligned}$

2.3. Assessing the potential for mitigation

Limiting the key drivers of carbon footprint, such as electricity and chemical consumption, can achieve a significant reduction in emission. However, the complex relationship between the operational indicators of MWWTPs makes it challenging to directly calculate the impact on GHG emissions when a single indicator changes. An interdisciplinary approach employing data science provides a novel solution to this problem [118], [119], [120]. In particular, statistical inference based on the random forest algorithm, such as, “quantile random forest” [121], can obtain the distribution of critical drivers of carbon footprint growth under various conditions. The monthly operational records of MWWTPs in the latest year, 2020, from the dataset, which includes 5444 plants and 63 551 pieces of data such as water quality, water volume, type of chemical, type of treatment process, season, and the located city, serve as input variables for the random forest model (Tables S10–S13 in Appendix A). The output results are the intensities of electricity and chemical consumption, which have a significant impact on the GHG emissions, as previously identified. Next, quantile random forest [121] is adopted to obtain the median values of these critical indicators under the same treatment conditions as the benchmarks. The logic of the methodology is outlined as follows.

In the quantile random forest (Eq. (3),

\hat{F} (y | X = x_{k})

is the probability that the target variable Y (the intensity of electricity or chemical consumption) is less than or equal to the quantile y for the k^th observation x_k (including water quality, water volume, types of used chemicals, treatment process, season, and located city). When the j^th target variable Y_j is less than or equal to the quantile y, the indicator function

I {Y_{j} \leq y}

equals 1; otherwise, it equals 0. Here, n is the number of training observations, and T is the number of trees in the ensemble. In calculating

\hat{F} (y | X = x_{k})

, the quantile random forest allocates a weight

ω_{tj} (x_{k})

to each sample in each decision tree [121], then estimates the conditional quantiles of emission intensity by summing over every sample and averaging the results of all the decision trees in the ensemble. As the sample size increases, the estimated quantiles converge to the true ones, indicating that this statistical inference method is consistent and can reasonably reflect the data distributions and potentials for mitigation. The process is implemented through the quantile predict function in MATLAB.

(3)

\hat{F} (y | X = x_{k}) = \frac{1}{T} \sum_{t = 1}^{T} \sum_{j = 1}^{n} ω_{tj} (x_{k}) I \{Y_{j} \leq y\}

When the actual carbon footprint is higher than the median value, there exists potential for mitigation indicated by the difference between them. This is because the potential MWWTP operation under the same conditions (such as the treatment process and the natural and socioeconomic conditions) as the “benchmark” will have more efficient material and energy uses as inferred by the quantile regression forest method.

3. Results and discussion

3.1. Overview of MWWTPs and carbon footprints in China

The primary step in comprehensively studying the emission patterns of MWWTPs is to illustrate the overall state of the industry’s development and carbon footprint. According to our accounting results, the number of China’s MWWTPs in operation reached 5444 with a total treatment capacity close to 1.9 million cubic meters per day (million m³·d⁻¹) by 2020 benefitting from decades of sustained and active construction. This remarkable progress is validated by rapidly growing GHG emissions at the industry level every month (Fig. 2). Since 2007, the overall carbon footprint has risen from 8.3 to nearly 42 million tonnes (Mt) of CO₂ equivalent (CO₂-eq). The average annual growth rate surpasses that of the national economy [122]. Previous studies focusing on the GHG emissions of the wastewater sector present discrepancies due to the differences in settings such as system boundaries. For instance, considering upstream infrastructure construction (129 Mt CO₂-eq in 2015 [52]) or downstream sludge treatment and disposal (53.0 Mt CO₂-eq in 2019 [51]; 58.3 Mt CO₂-eq in 2019 [53]) as sources of carbon footprint will increase total GHG emissions to different extents. Meanwhile, excluding chemical consumption tends to reduce the overall carbon footprint (approximately 30 Mt CO₂-eq in 2019 [50]). In general, GHG emissions from the wastewater industry expand sharply and have been confirmed to reach the scale of 10 Mt. An interesting observation from Fig. 2 is the periodic shrinkage of GHG emissions at the start of each year, which then returns to normal levels. For instance, the total carbon footprint decreased by 12.5% from January to February and increased by 10.0% in March 2019. This phenomenon is attributed to the Chinese Lunar New Year, typically occurring in late January or early February, when a large crowd returns from cities to rural areas and the national urban population temporarily drops to the lowest point of the year. Thus, the volume of municipal wastewater undergoing treatment, along with its associated carbon footprint, will also decrease. It indicates that the carbon footprint of MWWTPs is closely related to the urban population.

In terms of emission sources depicted by Fig. 2 [123], the electricity consumption plays the most crucial role. This is attributed to the increased demand for electricity resulting from improved pollutant removal rates, such as COD and TN, in recent years (Table S14 and Fig. S1 in Appendix A). The next significant contributor is N₂O emission during the treatment process, with its greenhouse effect far exceeding that of CH₄ in the same phase due to its much higher global warming potential. The carbon footprints led by the consumption of carbon source and phosphorus removing agent began to increase suddenly and eventually occupied a considerable proportion. The reason behind this is the issuance of the action plan for preventing and controlling wastewater by the State Council of the People's Republic of China in April 2015 [123]. This plan, the most influential regulation ever enacted in the wastewater management sector, prompted MWWTPs nationwide to add more chemicals to meet the increasingly stringent discharge standards, consequently expanding the corresponding carbon footprints. GHG emissions in the effluent remain relatively stable as the decreasing concentration of COD and TN after treatment partially offsets the effect of the growing volume of municipal wastewater.

Despite the sharp increase in the carbon footprint of China’s MWWTPs in just over a decade, the industry also provides clean water resources for the world’s largest urban population. When considering annual per capita values, the electricity consumption (26.1 kW·h) and GHG emissions (47.6 kg of CO₂-eq) of China’s MWWTPs do not appear prominent compared to some European countries (23–47 kW·h, 7–161 kg of CO₂-eq) [4], [124], [125], [126], [127], [128], [129], [130], [131]. However, the accounting results also reveal that when treating an equivalent volume of wastewater, the carbon footprint in China is likely to be 60.5% higher than that of the United States. Evidently, there remains substantial potential for mitigation, even though the industry construction is almost completed with the coverage of MWWTPs in county-level administrative regions sharply climbing from 29.1% to 90.3% (Fig. S1). This fully mirrors the characteristics of China as a developing country with a large population base and a relatively extensive economy. The future development of the municipal wastewater treatment industry will certainly place a greater emphasis on quality aspects such as low-carbon transition.

Although sludge treatment is excluded from the carbon footprint accounting scope, it remains an integral part of wastewater treatment, and the process of anaerobic digestion will generate onsite CH₄ and N₂O emissions [2]. According to existing research [53], the contribution of sludge treatment and disposal to the total carbon footprint is even as high as nearly 50%. On the other hand, the processes of composting and pyrolysis also provide opportunities to recover energy and materials from sludge [2]. Generally, the GHG emissions from the wastewater sector are minor in scale compared to traditional major emitters such as the energy, transportation, and chemical industries from a global perspective [132], but the mitigation costs and comprehensive effects of the wastewater sector are lower and more significant [20]. Meanwhile, some wastewater treatment plants have achieved 150% and even 180% energy self-sufficiency in Denmark by 2019 [24], [133], [134]. With the operation of the first wastewater resource recovery factory serving as a paragon of energy conservation and emission reduction, China’s MWWTPs are expected to play a more active role in achieving the national target of carbon neutrality before 2060 [41], [135].

3.2. Spatiotemporal trends of carbon footprints

China can be divided into six parts based on geographical orientations (Fig. S2 in Appendix A). Development across different regions in China is unbalanced, and the GHG emissions of MWWTPs also show a distinct tendency to cluster around major urban areas (Figs. S3 and S4 in Appendix A). As the economy grows, an increasing number of regions begin to access municipal wastewater treatment services, generating carbon footprints. Meanwhile, GHG emissions in areas where MWWTPs are already been located become more significant. The Beijing–Tianjin–Hebei region in north China, the Yangtze River Delta region in east China, and the Pearl River Delta region in south central China are the most economically active areas across the country, with pronounced carbon footprints from MWWTPs. In terms of emission intensity (the GHG emissions for every cubic meter of municipal wastewater treated), values and upward tendencies in northern China are generally higher and more noticeable than those in the south, especially in north and northwest China (Figs. S3 and S5 in Appendix A). Changes in carbon footprint and emission intensity over time reveal that regions with higher total values also experience greater increases (Figs. S3, S6, and S7 in Appendix A). Consequently, it is necessary to consider implementing stricter binding measures in these important cities when formulating national-level policies to effectively prevent the rapid growth of the sector’s GHG emissions.

The high-resolution spatiotemporal trends also reveal strong correlations between carbon footprint and socioeconomic factors. Metropolitan areas with dense populations and intensive socioeconomic activities have a stronger demand and capacity for the construction of municipal wastewater treatment facilities, leading to more substantial GHG emissions (Figs. S2 and S8–S12 in Appendix A). By contrast, the load rates of MWWTPs in remote regions are relatively lower (Fig. S13 in Appendix A). Additionally, effluent discharge standards are more stringent in more developed cities (Fig. S14 in Appendix A). Consequently, regions with prominent carbon footprints on the maps highly overlap with metropolitan areas.

On the other hand, the distribution of emission intensity tends to be more closely related to environmental conditions rather than socioeconomic factors. Traditional energy resources are abundant in north and northwestern China. Fossil fuels, such as coal and petroleum, constitute a larger share of the local power generation structure compared to renewables, thereby increasing the emission intensity associated with electricity consumption (Fig. S15 in Appendix A). Besides, the annual average temperature in the northern regions is lower than that in the south (Fig. S16 in Appendix A). As a result, microorganisms there are less active, requiring more energy and chemical inputs for compensation during the treatment process.

Another factor contributing to the heightened emission intensity could be the uneven distribution of rainfall (Fig. S17 in Appendix A). Following the dilution of influent, achieving the required effluent standards becomes more feasible. Notably, the concentration of COD and TN in the influent is indeed higher in northern regions (Figs. S18 and S19 in Appendix A). In addition, higher amounts of chemicals and electricity are required to remove pollutants in these areas, leading to elevated emission intensities (Figs. S20 and S21 in Appendix A). In contrast, despite the Pearl River Delta in south-central China playing a pivotal role in the country's economic development and local MWWTPs releasing a large amount of GHGs, the emission intensities remain relatively low. In practice, prioritizing emission intensity over the total carbon footprint proves to be more pragmatic. Leveraging high-resolution spatiotemporal trends allows for the rapid identification of areas with abnormal GHG emission patterns, facilitating the formulation of targeted strategies.

3.3. Impacts of operational factors on carbon footprint growth

Identifying the critical influencing factors for carbon footprint growth can unlock the full potential of big data, as depicted in Fig. 3. According to our accounting results, China’s MWWTPs collectively emitted approximately 320 Mt of CO₂-eq from 2007 to 2020. Notably, electricity consumption constitutes over one-third of this total, with N₂O emissions directly generated during the treatment process (Fig. S22 in Appendix A) closely following. These factors combined contribute to more than 65% of the total GHG emissions, further underscoring the importance of controlling energy usage and effectively managing TN removal for emission reduction within MWWTPs.

Fig. 3(a) illustrates the driving factors behind temporal changes in GHG emissions within MWWTPs, as determined through SDA. Reference points for analysis are specified for the years 2010, 2015, and 2020. Bar lengths are employed to depict the influence of relevant variables on carbon footprint changes over five-year intervals. These operational indicators are categorized into four primary domains: water volume, water quality, chemicals, and electricity.

During the initial period (2010–2015), the primary contributors to the carbon footprint of MWWTPs were the volume of treated wastewater and the influent’s TN concentration. However, during the subsequent stage (2015–2020), their impact diminished significantly. Simultaneously, the importance of carbon source, phosphorus removal agent, and electricity consumption escalated. In the context of MWWTPs located in north China, the impacts of operational factors on carbon footprint growth exhibited analogous trends, as depicted in Fig. 3(b). Additional findings from SDA concerning GHG emissions in MWWTPs across various regions in China can be found in Figs. S23–S27 in Appendix A. The significant role of efficient electricity use and decarbonization in determining carbon footprint has also been emphasized in an evaluation of GHG emissions from the European urban wastewater sector [136]. This universal phenomenon underscores a notable shift in carbon footprint drivers, transitioning from uncontrolled variables like influent water quality towards factors encompassing electricity and chemical utilization—a transformation that MWWTP operators can actively improve. For the municipal wastewater treatment sector, this shift signals a positive development, indicating that the plants themselves can play a more significant role in promoting the industry’s low-carbon development. Moreover, reducing the intensity of energy and materials consumption aligns with the pursuit of economic benefits, further enhancing the enthusiasm of MWWTPs to actively reduce their GHG emissions. In the future, policymakers should prioritize the management of these critical operational indicators to achieve efficient and precise reduction of the industry’s carbon footprint.

3.4. Impacts of treatment process and scale on emission intensity

The treatment process is not directly involved in the carbon footprint accounting of MWWTPs. However, it can exert important influences on the GHG emissions through energy and material consumption [20]. There are 30 types of treatment processes with various cumulative shares that have been successfully applied in practice across China (Table 1). The spatiotemporal distribution of proportions of different treatment processes is displayed in Fig. S28 in Appendix A. To explore the impacts of treatment processes on the carbon footprint of MWWTPs, the emission intensity of each process is calculated. Fig. 4(a) shows the distribution of emission intensity for different treatment processes. It reveals that most values fluctuate below 1.5 kilograms of CO₂-eq per cubic meter (kg CO₂-eq⋅m⁻³), with noticeable differences among different treatment processes. Only a few are mainstream processes with cumulative market shares exceeding 5%. Taking the median value as a reference, the emission intensity of the activated sludge (AS) process is the lowest among all the mainstream processes, while the process related to the sequencing batch reactor (SBR) is also low-carbon. Meanwhile, processes associated with anaerobic/anoxic/oxic (A²/O) and oxidation ditch (OD) tend to release more GHGs when treating the same volume of wastewater. In terms of direct emission intensity, the majority of values among wastewater treatment technologies fall within 0.4, with median values around 0.2 kg CO₂-eq⋅m⁻³ (Fig. S29 in Appendix A). Similar to the distribution of total emission intensity, AS and SBR are more climate-friendly than other mainstream processes. The primary reason could be that the AS process is broadly applicable for treating moderately polluted wastewater, while the SBR process is capable of achieving more efficient removal of pollutants by flexibly adjusting operational parameters. Besides, a study on Spain’s wastewater treatment system demonstrates that the carbon footprint of nature-based technologies such as high-rate algal ponds and constructed wetlands is significantly lower than that of conventional AS technology [137], [138].

Typically, the emission intensities of mainstream processes, particularly the direct components, are relatively high in China’s market. When planning the construction or imminent reconstruction of MWWTPs in the short term, it is advisable to consider processes characterized by lower emission intensities. Such a strategy holds promise for greater potential in mitigating climate change.

There is an overarching trend of decreasing emission intensity as the treatment scale of MWWTPs increases. According to our accounting results, the actual monthly treatment scales range from 100 m³ to 100 million m³. Fig. 4(b) shows that as the scale increases, the distribution of emission intensity presents an overall descending trend, especially between the quantiles of 10% and 60%, although the median value seems to be stable. Thus, the influences of treatment scales on the emission intensity of MWWTPs are complex and deserve further exploration. Furthermore, it is noteworthy that the top 5% of records contribute to over 35% of the total volume of treated wastewater. This underscores the importance for operators to maintain lower emission intensities within large-scale plants to prevent a significant increase in carbon footprint.

3.5. Mitigation potentials through reducing electricity and chemical use

Operations during the treatment process, often based on experience, can lead to excessive input of electricity and chemicals, thus increasing the indirect GHG emissions of MWWTPs. Measures such as refined control and management of the treatment status can effectively improve the efficiency of resource usage. In practice, these adjustments to enhance the operational performance of MWWTPs are much more feasible for achieving decarbonization than completely changing the treatment process, treatment scale, or wastewater quality. The extent of potential reduction hinges on the difference between the actual values and a benchmark, which is determined as the median value derived from the distribution of energy and material consumption under equivalent treatment conditions such as water quality, water volume, type of chemical, type of treatment process, season, and located city (Fig. S30 in Appendix A). The benchmarks are established through the random forest algorithm and quantile regression forest.

The findings reveal that in 2020, China’s MWWTPs have the potential to mitigate emissions by 3.3 Mt CO₂-eq, constituting 12.6% of indirect GHG emissions. Reductions in the intensity of phosphorus removing agent and electricity consumption contribute significantly more to emission reduction than that achieved through adjustments in carbon source and dewatering agent (Fig. S31 in Appendix A).

In Fig. S30(a), the benchmark intensity of indirect GHG emissions generally exhibits an upward trend with increasing influent concentration of COD or decreasing effluent concentration. As more COD is removed during wastewater treatment, the chemical and electricity consumption tends to increase. A similar trend is observed in Fig. S30(e), which shows the trends of the benchmark intensity of emissions caused by electricity consumption changing with the concentration of COD in wastewater. In Fig. S30(c), the trends are even more consistent with the pattern above, as the intensity of phosphorus removing agent consumption is directly related to the influent and effluent concentrations of total phosphorus (TP). The benchmark intensity of emissions within certain intervals exhibits a downward trend, influenced by other operational indicators with greater negative impacts.

The spatial distribution of the total potential for mitigation and the components related to phosphorus removing agent and electricity across the country in 2020 are shown in Figs. S30(b), (d), and (f), respectively. Metropolitan areas demonstrate a greater potential for mitigation, indicating that the decarbonization capacity is closely connected with the baseline of the total carbon footprint. Moreover, the large areas of blank space in Fig. S30(d) reveal that the intensity of phosphorus removing agent consumption in a considerable number of regions is already below the median level.

Various insights into specific decarbonization strategies, focusing on reducing resource consumption during operation, have been provided [11]. To decrease energy consumption during wastewater treatment, MWWTP operators can concentrate on the low-energy operation of devices that demand a significant amount of electricity, such as the pumps and aeration equipment. Automatic/intelligent control of the energy-intensive devices based on real-time monitoring of wastewater quantity and quality, is a promising pathway for achieving energy consumption reduction. Regular maintenance of equipment can also improve energy utilization efficiency by keeping equipment in optimal performance state. Besides, optimizing the spatial distribution of aeration equipment based on oxygen mass transfer characteristics can improve aeration and bio-chemical reaction efficiency, thereby reducing energy consumption. While energy recovery is a promising but investment-dependent approach to realizing energy self-sufficiency in WWTPs, it still relies on further scientific and technological progress to break through the bottlenecks. Currently, adopting anaerobic processes as the mainstream method for CH₄ recovery instead of the aeration unit is an efficient approach to reduce electricity usage. Chemicals are also conserved due to sludge reduction.

The strategy for reducing chemical consumption aligns with the procedures mentioned above. Using precise chemical dosing facilitated by online monitoring equipment and feedback systems, driven by extensive data models, can effectively accommodate variations in wastewater volume and quality. This approach not only mitigates chemical waste but also contributes to a reduction in indirect carbon emissions.

4. Conclusions

In pursuit of advancing the climate-friendly transformation of China’s municipal wastewater sector, a comprehensive study was undertaken, systematically examining spatiotemporal trends in the carbon footprints of MWWTPs, the driving forces behind these trends, and the potential for mitigation achieved through reductions in electricity and chemical usage.

The annual carbon footprint of MWWTPs across China increased to 41.9 Mt CO₂-eq in 2020, surpassing the pace of national economic growth. The rapid development of MWWTPs is strongly supported by resources such as the electricity which is the primary source of GHG emissions. As regulations on wastewater discharge become increasingly stringent, the chemical consumption also plays a more significant role in the carbon footprint growth of MWWTPs.

At the macroscopic level, the spatiotemporal trends of GHG emissions indicate their influence by various environmental and socioeconomic factors. Metropolitan areas exhibit greater carbon footprints due to higher production, collection, and treatment of municipal wastewater, coupled with stricter local effluent discharge standards. North and northwest China demonstrate higher emission intensities attributed to carbon-intensive electricity structures and increased energy and material consumption influenced by climatic conditions of lower temperature and less rainfall.

Over time, the significance of factors such as the volume of treated wastewater and the concentration of TN in influent as drivers of emission growth notably decreases. In contrast, the contributions from factors like the intensity of carbon source, phosphorus removing agent, and electricity consumption exhibit a significant increase. This shift in influential factors of the carbon footprint growth signifies a transition from external to internal drivers within MWWTPs, empowering them to assume a more proactive role decarbonization processes.

The analysis underscores that the reduction in electricity and chemical consumption intensities represents a promising and viable strategy for reducing the carbon footprint of MWWTPs. The potential for emissions reduction through more efficient utilization of materials and energy is 3.3 Mt CO₂-eq, approximately 12.6%. Primary mitigation strategies encompass the implementation of real-time monitoring equipment and feedback systems, rooted in big data models. These systems facilitate timely management of operational conditions and precise control of chemical quantities, thereby enhancing overall carbon efficiency.

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.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (52200228 and 72022004); the National Key Research and Development Program of China (2021YFC3200205 and 2022YFC3203704).

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.eng.2024.01.021.

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Liu GG, Qu JH, van Loosdrecht M. ‘Blue Route’ for combating climate change. Natl Sci Rev 2021; 8(8):nwab099.

[2]	Li L, Wang X, Miao J, Abulimiti A, Jing X, Ren N. Carbon neutrality of wastewater treatment—a systematic concept beyond the plant boundary. Environ Sci Ecotechnol 2022; 11:100180.

[3]	Zhang X, Liu Y. Circular economy is game-changing municipal wastewater treatment technology towards energy and carbon neutrality. Chem Eng J 2022; 429:132114.

[4]	Goliopoulos N, Mamais D, Noutsopoulos C, Dimopoulou A, Kounadis C. Energy consumption and carbon footprint of Greek wastewater treatment plants. Water 2022; 14(3):320.

[5]	Daw J, Hallett K, DeWolfe J, Venner I. Energy efficiency strategies for municipal wastewater treatment facilities.Report. Golden: National Renewable Energy Lab (NRE L); 2012 Jan. Report No.: NRE L/T P-7A20-53341.

[6]	Goldstein R, Smith WUS. Electricity consumption for water supply & treatment—the next half century. Water Supply 2002; 4(4):93.

[7]	Shin C, Szczuka A, Liu MJ, Mendoza L, Jiang R, Tilmans SH, et al. Recovery of clean water and ammonia from domestic wastewater: impacts on embodied energy and greenhouse gas emissions. Environ Sci Technol 2022; 56(12):8712-8721.

[8]	Arzbaecher C, Parmenter K, Ehrhard R, Murphy J. Electricity use and management in the municipal water supply and wastewater industries.Report. California: Water Research Foundation, Electric Power Research Institute; 2013 Nov. Report No.: 3002001433.

[9]	Simon-Várhelyi M, Cristea VM, Luca AV. Reducing energy costs of the wastewater treatment plant by improved scheduling of the periodic influent load. J Environ Manage 2020; 262:110294.

[10]	Hu G, Ou X, Zhang Q, Karplus VJ. Analysis on energy–water nexus by Sankey diagram: the case of Beijing. Desalin Water Treat 2013; 51(19–21):4183-4193.

[11]	Ren ZJ, Pagilla K. Pathways to water sector decarbonization, carbon capture and utilization. London: IWA Publishing; 2022.

[12]	Zeng S, Chen X, Dong X, Liu Y. Efficiency assessment of urban wastewater treatment plants in China: considering greenhouse gas emissions. Resour Conserv Recycling 2017; 120:157-165.

[13]	Pasqualino JC, Meneses M, Abella M, Castells F. LCA as a decision support tool for the environmental improvement of the operation of a municipal wastewater treatment plant. Environ Sci Technol 2009; 43(9):3300-3307.

[14]	Lundin M, Bengtsson M, Molander S. Life cycle assessment of wastewater systems: influence of system boundaries and scale on calculated environmental loads. Environ Sci Technol 2000; 34(1):180-186.

[15]	Zhang Y, Huo J, Zheng X. Wastewater: China’s next water source. Science 2021; 374(6573):1332.

[16]	Rothausen SGSA, Conway D. Greenhouse-gas emissions from energy use in the water sector. Nat Clim Chang 2011; 1(4):210-219.

[17]	Liao X, Tian Y, Gan Y, Ji J. Quantifying urban wastewater treatment sector’s greenhouse gas emissions using a hybrid life cycle analysis method — an application on Shenzhen city in China. Sci Total Environ 2020; 745:141176.

[18]	Figueroa FL, Korbee N, Abdala-Díaz R, Gómez-Pinchetti JL, Aci FGén. Growing algal biomass using wastes. D.P. Häder, G.S. Erzinger (Eds.), Bioassays, Elsevier, Oxon 2018; 99-117.

[19]	Li WW, Yu HQ, Rittmann BE. Chemistry: reuse water pollutants. Nature 2015; 528(7580):29-31.

[20]	Xi J, Gong H, Zhang Y, Dai X, Chen L. The evaluation of GHG emissions from Shanghai municipal wastewater treatment plants based on IPCC and operational data integrated methods (ODIM). Sci Total Environ 2021; 797:148967.

[21]	Hockstad L, Hanel L. Inventory of U.S. greenhouse gas emissions and sinks: 1990–2017. Report. Hawaii: Environmental System Science Data Infrastructure for a Virtual Ecosystem; 2018 Jan. Report No.: cdiac:EPA-EMISSION S.

[22]	Hua H, Jiang S, Yuan Z, Liu X, Zhang Y, Cai Z. Advancing greenhouse gas emission factors for municipal wastewater treatment plants in China. Environ Pollut 2022; 295:118648.

[23]	European Commission. EDGAR-emissions database for global atmospheric research [Internet].Brussels: European Commission; 2023 Nov 20 [cited 2024 Feb 29]. Available from: https://edgar.jrc.ec.europa.eu/.

[24]	Farag Mò, Damgaard A, Rebsdorf M, Nielsen PH, Rygaard M. Challenges in carbon footprint evaluations of state-of-the-art municipal wastewater resource recovery facilities. J Environ Manage 2022; 320:115715.

[25]	Nielsen OK, Plejdrup MS, Winther M, Nielsen M, Gyldenkærne S, Mikkelsen MH, et al. Denmark’s national inventory report 2020: emission inventories 1990–2018—submitted under the United Nations Framework Convention on Climate Change and the Kyoto Protocol.Report. Aarhus: Scientific Report; 2020.

[26]	Nicole M, Marion P. Annual European Union greenhouse gas inventory 1990–2018 and inventory report 2020.Report. Brussels: European Commission, Directorate-General Climate Action European Environment Agency; 2020 May.

[27]	Qu J, Dai X, Hu HY, Huang X, Chen Z, Li T, et al. Emerging trends and prospects for municipal wastewater management in China. ACS EST Engg 2022; 2(3):323-336.

[28]	Zhang Q, Smith K, Zhao X, Jin X, Wang S, Shen J, et al. Greenhouse gas emissions associated with urban water infrastructure: what we have learnt from China’s practice. WIREs Water 2021; 8(4):e1529.

[29]	Yan X, Qiu D, Zheng S, Cheng K, Han Y, Sun J, et al. Spatial and temporal distribution of greenhouse gas emissions from municipal wastewater treatment plants in China from 2005 to 2014. Earths Futur 2019; 7(4):340-350.

[30]	Zaborowska E, Czerwionka K, M Jąkinia. Integrated plant-wide modelling for evaluation of the energy balance and greenhouse gas footprint in large wastewater treatment plants. Appl Energy 2021; 282:116126.

[31]	Kammen DM, Sunter DA. City-integrated renewable energy for urban sustainability. Science 2016; 352(6288):922-928.

[32]	Wang X, McCarty PL, Liu J, Ren NQ, Lee DJ, Yu HQ, et al. Probabilistic evaluation of integrating resource recovery into wastewater treatment to improve environmental sustainability. Proc Natl Acad Sci USA 2015; 112(5):1630-1635.

[33]	Wang X, Daigger G, Lee DJ, Liu J, Ren NQ, Qu J, et al. Evolving wastewater infrastructure paradigm to enhance harmony with nature. Sci Adv 2018; 4(8):eaaq0210.

[34]	Flores L, García J, Pena R, Garfí M. Carbon footprint of constructed wetlands for winery wastewater treatment. Ecol Eng 2020; 156:105959.

[35]	Larsen TA, Hoffmann S, Lüthi C, Truffer B, Maurer M. Emerging solutions to the water challenges of an urbanizing world. Science 2016; 352(6288):928-933.

[36]	Lu L, Guest JS, Peters CA, Zhu X, Rau GH, Ren ZJ. Wastewater treatment for carbon capture and utilization. Nat Sustain 2018; 1(12):750-758.

[37]	Qiao S, Hou C, Wang X, Zhou J. Minimizing greenhouse gas emission from wastewater treatment process by integrating activated sludge and microalgae processes. Sci Total Environ 2020; 732:139032.

[38]	Huang F, Shen W, Zhang X, Seferlis P. Impacts of dissolved oxygen control on different greenhouse gas emission sources in wastewater treatment process. J Clean Prod 2020; 274:123233.

[39]	Caniani D, Caivano M, Pascale R, Bianco G, Mancini IM, Masi S, et al. CO₂ and N₂O from water resource recovery facilities: evaluation of emissions from biological treatment, settling, disinfection, and receiving water body. Sci Total Environ 2019; 648:1130-1140.

[40]	Marinelli E, Radini S, Foglia A, Lancioni N, Piasentin A, Eusebi AL, et al. Validation of an evidence-based methodology to support regional carbon footprint assessment and decarbonisation of wastewater treatment service in Italy. Water Res 2021; 207:117831.

[41]	Qu JH, Ren HQ, Wang HC, Wang KJ, Yu G, Ke B, et al. China launched the first wastewater resource recovery factory in Yixing. Front Environ Sci Eng 2022; 16(1):13.

[42]	Qu S, Yang Y, Wang Z, Zou JP, Xu M. Great divergence exists in Chinese provincial trade-related CO₂ emission accounts. Environ Sci Technol 2020; 54(14):8527-8538.

[43]	Larsen TA. CO₂-neutral wastewater treatment plants or robust, climate-friendly wastewater management? A systems perspective. Water Res 2015; 87:513-521.

[44]	Yoshida H, M Jønster, Scheutz C. Plant-integrated measurement of greenhouse gas emissions from a municipal wastewater treatment plant. Water Res 2014; 61:108-118.

[45]	Mannina G, Ekama G, Caniani D, Cosenza A, Esposito G, Gori R, et al. Greenhouse gases from wastewater treatment—a review of modelling tools. Sci Total Environ, 551–552 2016; 254-270.

[46]	Mannina G, Rebou TFças, Cosenza A, Chandran K. A plant-wide wastewater treatment plant model for carbon and energy footprint: model application and scenario analysis. J Clean Prod 2019; 217:244-256.

[47]

Deborah B, Michael DS, Yoshitaka E, Juraj F, Céline G, Gregory MP. Wastewater treatment and discharge.In: Eggleston HS, Buendia L, Miwa K, Ngara T, Tanabe K, editors. 2019. 2006 IPC C guidelines for national greenhouse gas inventories. Kitakyushu: Institute for Global Environmental Strategies; 2006.

[48]	Nayeb H, Mirabi M, Motiee H, Alighardashi A, Khoshgard A. Estimating greenhouse gas emissions from Iran’s domestic wastewater sector and modeling the emission scenarios by 2030. J Clean Prod 2019; 236:117673.

[49]	Nguyen TKL, Ngo HH, Guo WS, Chang SW, Nguyen DD, Nghiem LD, et al. A critical review on life cycle assessment and plant-wide models towards emission control strategies for greenhouse gas from wastewater treatment plants. J Environ Manage 2020; 264:110440.

[50]	Wang D, Ye W, Wu G, Li R, Guan Y, Zhang W, et al. Greenhouse gas emissions from municipal wastewater treatment facilities in China from 2006 to 2019. Sci Data 2022; 9(1):317.

[51]	Du WJ, Lu JY, Hu YR, Xiao J, Yang C, Wu J, et al. Spatiotemporal pattern of greenhouse gas emissions in China’s wastewater sector and pathways towards carbon neutrality. Nat Water 2023; 1(2):166-175.

[52]	Chen S, Zhang L, Liu B, Yi H, Su H, Kharrazi A, et al. Decoupling wastewater-related greenhouse gas emissions and water stress alleviation across 300 cities in China is challenging yet plausible by 2030. Nat Water 2023; 1(6):534-546.

[53]	Huang Y, Meng F, Liu S, Sun S, Smith K. China’s enhanced urban wastewater treatment increases greenhouse gas emissions and regional inequality. Water Res 2023; 230:119536.

[54]	Bao Z, Sun S, Sun D. Assessment of greenhouse gas emission from A/O and SBR wastewater treatment plants in Beijing, China. Int Biodeter Biodegr 2016; 108:108-114.

[55]	Ross BN, Lancellotti BV, Brannon EQ, Loomis GW, Amador JA. Greenhouse gas emissions from advanced nitrogen-removal onsite wastewater treatment systems. Sci Total Environ 2020; 737:140399.

[56]	Asadi M, McPhedran K. Estimation of greenhouse gas and odour emissions from a cold region municipal biological nutrient removal wastewater treatment plant. J Environ Manage 2021; 281:111864.

[57]	Delre A, ten M Hoeve, Scheutz C. Site-specific carbon footprints of Scandinavian wastewater treatment plants, using the life cycle assessment approach. J Clean Prod 2019; 211:1001-1014.

[58]	Gruber W, von L Känel, Vogt L, Luck M, Biolley L, Feller K, et al. Estimation of countrywide N₂O emissions from wastewater treatment in Switzerland using long-term monitoring data. Water Res X 2021; 13:100122.

[59]	Dias DFC, Marques R, Martins C, Martins A, Oehmen A. The impact of a seasonal change in loading rate on the nitrous oxide emissions at the WWTP of a tourist region. Sci Total Environ 2022; 804:149987.

[60]	Pahunang RR, Buonerba A, Senatore V, Oliva G, Ouda M, Zarra T, et al. Advances in technological control of greenhouse gas emissions from wastewater in the context of circular economy. Sci Total Environ 2021; 792:148479.

[61]	Qu J, Wang H, Wang K, Yu G, Ke B, Yu HQ, et al. Municipal wastewater treatment in China: development history and future perspectives. Front Environ Sci Eng 2019; 13(6):88.

[62]

The Ministry of Housing and Urban–Rural Development of the People's Republic of China. The information system of national municipal wastewater treatment and management [Internet].Beijing: The Ministry of Housing and Urban-Rural Development; undated [cited 2024 Feb 29]. Available from: https://wushui.mohurd.gov.cn:8088/login.jsp.

[63]	Doorn MR, Strait RP, Barnard WR, Eklund B. Estimates of global greenhouse gas emissions from industrial and domestic wastewater treatment.Report. Washington, D C: US Environmental Protection Agency; 1997.

[64]	Daelman MR, van EM Voorthuizen, van LG Dongen, Volcke EI, van Loosdrecht MC. Methane and nitrous oxide emissions from municipal wastewater treatment—results from a long-term study. Water Sci Technol 2013; 67(10):2350-2355.

[65]	Daelman MR, van EM Voorthuizen, van UG Dongen, Volcke EI, van Loosdrecht MC. Seasonal and diurnal variability of N₂O emissions from a full-scale municipal wastewater treatment plant. Sci Total Environ 2015; 536:1-11.

[66]	Foley J, de D Haas, Yuan Z, Lant P. Nitrous oxide generation in full-scale biological nutrient removal wastewater treatment plants. Water Res 2010; 44(3):831-844.

[67]	Kimochi Y, Inamori Y, Mizuochi M, Xu KQ, Matsumura M. Nitrogen removal and N₂O emission in a full-scale domestic wastewater treatment plant with intermittent aeration. J Ferment Bioeng 1998; 86(2):202-206.

[68]	Wang Y, Lin X, Zhou D, Ye L, Han H, Song C. Nitric oxide and nitrous oxide emissions from a full-scale activated sludge anaerobic/anoxic/oxic process. Chem Eng J 2016; 289:330-340.

[69]	Aboobakar A, Cartmell E, Stephenson T, Jones M, Vale P, Dotro G. Nitrous oxide emissions and dissolved oxygen profiling in a full-scale nitrifying activated sludge treatment plant. Water Res 2013; 47(2):524-534.

[70]	Rodríguez-Caballero A, Aymerich I, Poch M, Pijuan M. Evaluation of process conditions triggering emissions of green-house gases from a biological wastewater treatment system. Sci Total Environ 2014; 493:384-391.

[71]	Masuda S, Sano I, Hojo T, Li YY, Nishimura O. The comparison of greenhouse gas emissions in sewage treatment plants with different treatment processes. Chemosphere 2018; 193:581-590.

[72]	Ahn JH, Kim S, Park H, Rahm B, Pagilla K, Chandran K. N₂O emissions from activated sludge processes, 2008–2009: results of a national monitoring survey in the United States. Environ Sci Technol 2010; 44(12):4505-4511.

[73]	Ni BJ, Pan Y, van B den Akker, Ye L, Yuan Z. Full-scale modeling explaining large spatial variations of nitrous oxide fluxes in a step-feed plug-flow wastewater treatment reactor. Environ Sci Technol 2015; 49(15):9176-9184.

[74]	Pan Y, van B den Akker, Ye L, Ni BJ, Watts S, Reid K, et al. Unravelling the spatial variation of nitrous oxide emissions from a step-feed plug-flow full scale wastewater treatment plant. Sci Rep 2016; 6(1):20792.

[75]	Rodríguez-Caballero A, Aymerich I, Marques R, Poch M, Pijuan M. Minimizing N₂O emissions and carbon footprint on a full-scale activated sludge sequencing batch reactor. Water Res 2015; 71:1-10.

[76]	Masuda S, Suzuki S, Sano I, Li YY, Nishimura O. The seasonal variation of emission of greenhouse gases from a full-scale sewage treatment plant. Chemosphere 2015; 140:167-173.

[77]	Mello WZ, Ribeiro RP, Brotto AC, Kligerman DC, Piccoli AS, Oliveira JL. Nitrous oxide emissions from an intermittent aeration activated sludge system of an urban wastewater treatment plant. Quim Nova 2013; 36(1):16-20.

[78]	Tranvik LJ, Downing JA, Cotner JB, Loiselle SA, Striegl RG, Ballatore TJ, et al. Lakes and reservoirs as regulators of carbon cycling and climate. Limnol Oceanogr 2009; 54(6 part 2):2298-2314.

[79]	Deemer BR, Harrison JA, Li S, Beaulieu JJ, DelSontro T, Barros N, et al. Greenhouse gas emissions from reservoir water surfaces: a new global synthesis. Bioscience 2016; 66(11):949-964.

[80]	Bastviken D, Tranvik LJ, Downing JA, Crill PM, Enrich-Prast A. Freshwater methane emissions offset the continental carbon sink. Science 2011; 331(6013):50.

[81]

St VL Louis, Kelly CA, ÉDuchemin , Rudd JWM, Rosenberg DM. Reservoir surfaces as sources of greenhouse gases to the atmosphere: a global estimate: reservoirs are sources of greenhouse gases to the atmosphere, and their surface areas have increased to the point where they should be included in global inventories of anthropogenic emissions of greenhouse gases. Bioscience 2000; 50(9):766-775.

[82]	Barros N, Cole JJ, Tranvik LJ, Prairie YT, Bastviken D, Huszar VL, et al. Carbon emission from hydroelectric reservoirs linked to reservoir age and latitude. Nat Geosci 2011; 4(9):593-596.

[83]	Li S, Zhang Q. Carbon emission from global hydroelectric reservoirs revisited. Environ Sci Pollut Res Int 2014; 21(23):13636-13641.

[84]	Stanley EH, Casson NJ, Christel ST, Crawford JT, Loken LC, Oliver SK. The ecology of methane in streams and rivers: patterns, controls, and global significance. Ecol Monogr 2016; 86(2):146-171.

[85]	Kaplan W, Elkins J, Kolb C, McElroy M, Wofsy S, Duran A. Nitrous oxide in fresh water systems: an estimate for the yield of atmospheric N₂O associated with disposal of human waste. Pure appl geophys 1978; 116:423-438.

[86]	McElroy M, Elkins J, Wofsy S, Kolb C, Duran A, Kaplan W. Production and release of N₂O from the Potomac Estuary. Limnol Oceanogr 1978; 23(6):1168-1182.

[87]	Goreau TJ, Kaplan WA, Wofsy SC, McElroy MB, Valois FW, Watson SW. Production of NO₂⁻ and N₂O by nitrifying bacteria at reduced concentrations of oxygen. Appl Environ Microbiol 1980; 40(3):526-532.

[88]	McCarthy JJ, Kaplan W, Nevins JL. Chesapeake Bay nutrient and plankton dynamics. Sources and sinks of nitrite. Limnol Oceanogr 1984; 29(1):84-98.

[89]	Seitzinger SP, Nixon SW, Pilson ME. Denitrification and nitrous oxide production in a coastal marine ecosystem. Limnol Oceanogr 1984; 29(1):73-83.

[90]	Kaplan WA, Wofsy SC. The biogeochemistry of nitrous oxide: a review. Adv aquat microbiol 1985; 3:181-206.

[91]	Seitzinger SP. Denitrification in freshwater and coastal marine ecosystems: ecological and geochemical significance. Limnol Oceanogr 1988; 33(4 Pt 2):702-724.

[92]	Yoshida N. ¹⁵N-depleted N₂O as a product of nitrification. Nature 1988; 335(6190):528-529.

[93]	Fauzi R, Mantoura C, Law CS, Owens NJ, Burkill PH, Woodward EMS, et al. Nitrogen biogeochemical cycling in the northwestern Indian Ocean. Deep-Sea Res Pt II 1993; 40(3):651-671.

[94]	Kester RA, De W Boer, Laanbroek HJ. Production of NO and N₂O by pure cultures of nitrifying and denitrifying bacteria during changes in aeration. Appl Environ Microbiol 1997; 63(10):3872-3877.

[95]	Seitzinger SP, Kroeze C. Global distribution of nitrous oxide production and N inputs in freshwater and coastal marine ecosystems. Global Biogeochem Cy 1998; 12(1):93-113.

[96]	Punshon S, Moore RM. Nitrous oxide production and consumption in a eutrophic coastal embayment. Mar Chem 2004; 91(1–4):37-51.

[97]	Frame CH, Casciotti K. Biogeochemical controls and isotopic signatures of nitrous oxide production by a marine ammonia-oxidizing bacterium. Biogeosciences 2010; 7(9):2695-2709.

[98]	Beaulieu JJ, Tank JL, Hamilton SK, Wollheim WM, Hall ROJ, Mulholland PJ, et al. Nitrous oxide emission from denitrification in stream and river networks. Proc Natl Acad Sci USA 2011; 108(1):214-219.

[99]	Yan W, Yang L, Wang F, Wang J, Ma P. Riverine N₂O concentrations, exports to estuary and emissions to atmosphere from the Changjiang River in response to increasing nitrogen loads. Global Biogeochem Cy 2012; 26(4):2010GB003984.

[100]

Zhu X, Burger M, Doane TA, Horwath WR. Ammonia oxidation pathways and nitrifier denitrification are significant sources of N₂O and NO under low oxygen availability. Proc Natl Acad Sci USA 2013; 110(16):6328-6333.

[101]

Ji Q, Babbin AR, Jayakumar A, Oleynik S, Ward BB. Nitrous oxide production by nitrification and denitrification in the Eastern Tropical South Pacific oxygen minimum zone. Geophys Res Lett 2015; 42(24):10755-10764.

[102]

Wang J, Chen N, Yan W, Wang B, Yang L. Effect of dissolved oxygen and nitrogen on emission of N₂O from rivers in China. Atmos Environ 2015; 103:347-356.

[103]

He Y, Wang X, Chen H, Yuan X, Wu N, Zhang Y, et al. Effect of watershed urbanization on N₂O emissions from the Chongqing metropolitan river network, China. Atmos Environ 2017; 171:70-81.

[104]

The Intergovernmental Panel on Climate Change (IPC C) [Internet]. The Intergovernmental Panel on Climate Change Geneva: IPCC; undated [cited 2024 Feb 29]. Available from: https://www.ipcc.ch/.

[105]

National Center for Climate Change Strategy and International Cooperation. The average CO2 emission factor of regional power grid of China in 2011 and 2012.Beijing: National Center for Climate Change Strategy and International Cooperation; 2014. Chinese.

[106]

Li M. The emission factor of national power grid declined by 4.3%. What’s the influence on carbon footprint accounting for enterprises [Internet]? Beijing: TGCW; 2021 Dec 6 [cited 2024 Feb 29]. Available from: https://m.in-en.com/article/html/energy-2310240.shtml. Chinese.

[107]

Xin W, Yang X, Wang Z. Example of carbon accounting for water treatment and application study of different N₂O emission factors. Adv Analy Chem 2022; 12(2):60-67.

[108]

Zhang Z. Engineering design and carbon emission analysis of municipal WWTP under stricter discharge criteria—case of a WWTP in Ningbo. Water Purif Technol 2023; 42(5):6.

[109]

China Environmental Protection Association. T/CAEPI 49—2022: Technical specification for low-carbon operation evaluation of sewage treatment plant.Chinese standard. Beijing: China Environmental Protection Association; 2022. Chinese.

[110]

European Inorganic Coagulants Producers Association (INCOP A). Life cycle analysis of leading coagulants: executive summary.Report. Brussels: European Inorganic Coagulants Producers Association; 2014 May.

[111]

Parravicini V, Svardal K, Krampe J. Greenhouse gas emissions from wastewater treatment plants. Energy Procedia 2016; 97:246-253.

[112]

Wang HC. Carbon abatement in municipal wastewater treatment area.Jishui Paishui 2010;36(12):1–3. Chinese.

[113]

Carr M. Reducing greenhouse gas emissions industrial biotechnology and biorefining.Washington, D C: Biotechnology Industry Organization; 2009.

[114]

Carlson S, Walburger A. Energy index development for benchmarking water and wastewater utilities.Gander: Awwa Research Foundation; 2007.

[115]

Zhao Y, Wang S, Zhang Z, Liu Y, Ahmad A. Driving factors of carbon emissions embodied in China–US trade: a structural decomposition analysis. J Clean Prod 2016; 131:678-689.

[116]

Brizga J, Feng K, Hubacek K. Drivers of greenhouse gas emissions in the Baltic States: a structural decomposition analysis. Ecol Econ 2014; 98:22-28.

[117]

Hu Y, Yin ZF, Ma J, Du WC, Liu DH, Sun LX. Determinants of GHG emissions for a municipal economy: structural decomposition analysis of Chongqing. Appl Energy 2017; 196:162-169.

[118]

Zhao B, Shuai C, Qu S, Xu M. Using deep learning to fill data gaps in environmental footprint accounting. Environ Sci Technol 2022; 56(16):11897-11906.

[119]

Zhao B, Shuai C, Hou P, Qu S, Xu M. Estimation of unit process data for life cycle assessment using a decision tree-based approach. Environ Sci Technol 2021; 55(12):8439-8446.

[120]

Zhao B, Yu L, Wang C, Shuai C, Zhu J, Qu S, et al. Urban air pollution mapping using fleet vehicles as mobile monitors and machine learning. Environ Sci Technol 2021; 55(8):5579-5588.

[121]

Meinshausen N, Ridgeway G. Quantile regression forests. J Mach Learn Res 2006; 7(6):983-999.

[122]

National Bureau of Statistics. Annual data-gross domestic product [Internet].Beijing: National Bureau of Statistics; undated [cited 2024 Feb 29]. Available from: https://data.stats.gov.cn/easyquery.htm?cn=C01. Chinese.

[123]

The State Council. Action plan for prevention and control of water pollution printed and distributed [Internet].Beijing: The State Council; 2015 Apr 14 [cited 2024 Feb 29]. Available from: https://english.mee.gov.cn/News_service/news_release/201504/t20150427_299595.shtml.

[124]

National Bureau of Statistics. National population and urban population in 2020.Beijing: China Statistics Press; 2021. Chinese.

[125]

The Ministry of Housing and Urban–Rural Development of the People’s Republic of China. The yearbook of urban construction in 2020 [Internet].Beijing: The Ministry of Housing and Urban-Rural Development of the People’s Republic of China; [cited 2024 Feb 29]. Available from: https://www.mohurd.gov.cn/gongkai/fdzdgknr/sjfb/tjxx/jstjnj/index.html.

[126]

Jonasson M. Energy benchmark for wastewater treatment processes—a comparison between Sweden and Austria [dissertation].Lund: Lund University; 2007.

[127]

Balm Pér. Operation costs and consumption of resources at Nordic nutrient removal plants. Water Sci Technol 2000; 41(9):273-279.

[128]

Mamais D, Noutsopoulos C, Dimopoulou A, Stasinakis A, Lekkas TD. Wastewater treatment process impact on energy savings and greenhouse gas emissions. Water Sci Technol 2015; 71(2):303-308.

[129]

Zib L III, Byrne DM, Marston LT, Chini CM. Operational carbon footprint of the U.S. water and wastewater sector’s energy consumption. J Clean Prod 2021; 321:128815.

[130]

Gustavsson DJI, Tumlin S. Carbon footprints of Scandinavian wastewater treatment plants. Water Sci Technol 2013; 68(4):887-893.

[131]

Maktabifard M, Zaborowska E, Makinia J. Energy neutrality versus carbon footprint minimization in municipal wastewater treatment plants. Bioresour Technol 2020; 300:122647.

[132]

Intelligence Research Group; Petroleum B. Analysis about global CO2 emissions in 2020 [Internet].Beijing: Intelligence Research Group; 2021 Aug 3 [cited 2024 Feb 29]. Available from: https://www.chyxx.com/industry/202108/966523.html.

[133]

Aarhusvand.Achieving 150% energy self-sufficiency at Marselisborg wastewater treatment plant. Hasselager Allé: Aarhus Vand; 2020 Jun 18 [cited 2024 Feb 29]. Available from: https://www.aarhusvand.com/news/achieving-150-energy-self-sufficiency-at-marselisborg-wastewater-treatment-plant/.

[134]

VandCenter Syd. VandCenter Syd er energineutral [Internet].Odense C: VandCenter Syd; 2020 May 4 [cited 2024 Feb 29]. Available from: https://www.vandcenter.dk/nyheder/2020/04-energineutral2019.

[135]

Liu Z, Deng Z, He G, Wang H, Zhang X, Lin J, et al. Challenges and opportunities for carbon neutrality in China. Nat Rev Earth Env 2022; 3(2):141-155.

[136]

Parravicini V, Nielsen PH, Thornberg D, Pistocchi A. Evaluation of greenhouse gas emissions from the European urban wastewater sector, and options for their reduction. Sci Total Environ 2022; 838(Pt4): 156322.

[137]

Nguyen TKL, Ngo HH, Guo W, Nghiem LD, Qian G, Liu Q, et al. Assessing the environmental impacts and greenhouse gas emissions from the common municipal wastewater treatment systems. Sci Total Environ 2021; 801:149676.

[138]

Garfí M, Flores L, Ferrer I. Life cycle assessment of wastewater treatment systems for small communities: activated sludge, constructed wetlands and high rate algal ponds. J Clean Prod 2017; 161:211-219.