In the existing literature, efficiency-related concepts commonly applied in the wastewater treatment sector include energy efficiency, environmental performance, eco-efficiency, and carbon efficiency. However, these concepts differ substantially in terms of research focus, input–output boundaries, and policy implications (Kuosmanen and Kortelainen, 2005; Zaim and Taskin, 2000; Zhang et al., 2014). The concept of CEEWT proposed in this study is not a simple rebranding of these existing measures, but rather a context-specific and integrated efficiency indicator.

From the perspective of environmental economics, CEEWT emphasizes the minimization of carbon emissions directly or indirectly associated with wastewater treatment processes while delivering pollution abatement services under given economic and institutional constraints. Its essence lies in resource allocation efficiency in the presence of internalized environmental externalities (Pigou, 1920). From the standpoint of efficiency analysis theory, CEEWT represents a typical multi-input–multi-output efficiency problem, incorporating conventional inputs such as capital, energy, and labor, desired outputs such as treatment capacity and pollutant removal, and carbon emissions explicitly treated as undesirable outputs (Färe et al., 1989; Chung et al., 1997).

Within the framework of urban metabolism and sustainable development theory, wastewater treatment systems constitute a critical node in urban material and energy flows. Their operational performance affects not only water environmental quality but also urban low-carbon transition through energy consumption and greenhouse gas emissions (Kennedy et al., 2011; Zhang et al., 2015). Therefore, CEEWT highlights a comprehensive evaluation of wastewater treatment system performance under the dual objectives of pollution reduction and carbon mitigation, distinguishing it from indicators that focus solely on unit energy consumption or single pollution control outcomes.

Empirical research on efficiency in the water and wastewater sector generally falls into three categories: studies on operational or technical efficiency, environmental or eco-efficiency, and carbon emission or energy–environment efficiency.

Regarding operational and technical efficiency, a large body of studies has employed DEA and SBM-DEA models to assess the efficiency of municipal wastewater treatment plants under different scales and treatment technologies. These studies consistently find that scale effects, process configurations, and management practices are key determinants of efficiency (Molinos-Senante et al., 2016; Athanassopoulos, 2000). However, most of these analyses focus on treatment capacity, operating costs, or energy consumption, with carbon emissions rarely incorporated explicitly into the evaluation framework.

In studies on environmental and eco-efficiency, researchers have progressively incorporated pollutant removal as desired outputs and pollution loads or excessive discharges as undesirable outputs to assess the environmental performance of wastewater treatment systems under regulatory constraints (Yang et al., 2018; Hu and Wang, 2006). While these approaches provide useful tools for characterizing pollution reduction performance, carbon emissions are often treated as ancillary variables rather than central evaluation targets.

More recently, driven by growing concerns over climate change, some studies have examined the carbon emission characteristics and efficiency of wastewater treatment systems. These studies typically adopt LCA or MFA to quantify carbon footprints of different treatment technologies, or integrate carbon emissions as undesirable outputs within DEA-based frameworks (Gu et al., 2017; Longo et al., 2016). Although these contributions offer important insights into emission sources and mitigation potential at the process or plant level, systematic assessments of wastewater treatment carbon emission efficiency at the regional or river-basin scale remain limited.

At the theoretical level, wastewater treatment efficiency and carbon performance are jointly influenced by institutional, technological, and economic factors. According to the Porter Hypothesis, well-designed environmental regulations may stimulate technological innovation and managerial improvements, thereby enhancing resource-use efficiency and environmental performance (Porter and van der Linde, 1995). In the wastewater sector, stricter discharge standards and energy or carbon constraints may incentivize process upgrading and operational optimization, ultimately improving CEEWT (Ambec et al., 2013).

The Environmental Kuznets Curve (EKC) hypothesis provides another important lens for understanding the relationship between economic development and environmental performance (Grossman and Krueger, 1995). Some studies suggest that in the context of water environmental governance, increased economic development is associated with higher investment in wastewater treatment infrastructure and technological advancement, potentially leading to phased improvements in environmental performance. However, this relationship exhibits substantial heterogeneity across regions and environmental indicators, and its applicability to wastewater treatment carbon emission efficiency remains to be empirically verified (Dinda, 2004).

In addition, technological conditions, fiscal investment, pricing and cost-recovery mechanisms, energy prices, and regional governance arrangements have all been identified as important drivers of wastewater treatment efficiency and carbon performance (OECD, 2010; IPCC, 2019). Nevertheless, existing studies often examine these factors in isolation, and comprehensive analyses integrating multiple drivers within a unified analytical framework remain relatively scarce.

In summary, although the literature on wastewater treatment efficiency and carbon emissions has expanded rapidly, several gaps remain. First, at the conceptual level, the distinction between carbon emission efficiency and related concepts such as energy efficiency and environmental efficiency has not been sufficiently clarified. Second, in terms of spatial scale, systematic comparisons of wastewater treatment carbon emission efficiency at the regional or river-basin level are still limited. Third, variations in indicator systems and methodological choices across studies hinder the comparability of existing findings (Wang and Su, 2019; Wang et al., 2021).

To address these gaps, this study focuses on the YREB and, based on a clear conceptualization of CEEWT, develops a multi-indicator comprehensive evaluation framework to analyze the spatiotemporal dynamics and driving factors of regional wastewater treatment carbon emission efficiency. By doing so, this study aims to provide new empirical evidence and policy-relevant insights that complement and extend the existing literature.

The data used in this study were obtained from authoritative public sources, including the China Urban and Rural Construction Statistical Yearbook and the China Statistical Yearbook. These datasets ensure data consistency, inter regional comparability, and long-term temporal continuity, which are essential for a spatiotemporal efficiency analysis covering multiple provinces over an extended period.

Carbon emissions from wastewater treatment were estimated following the Intergovernmental Panel on Climate Change (IPCC) Guidelines for National Greenhouse Gas Inventories. Given the regional and temporal scope of this study, which covers 11 provinces and municipalities in the YEB over a 10-year period, the IPCC Tier 1 methodology was adopted to ensure data consistency, transparency, and cross-regional comparability.

The IPCC approach allows carbon emissions associated with wastewater treatment to be decomposed into emissions from methane (CH₄) and nitrous oxide (N₂O) generated during wastewater and sludge treatment processes. The total carbon dioxide equivalent emissions from wastewater treatment are calculated as Equation 1: E F W W T P = E F C H 4 + E F N 2 O + E F s l u d g e × G W P C H 4 (1)

Where E F W W T P denotes total carbon emissions from wastewater treatment (t CO₂e), E F C H 4 represents methane emissions from wastewater treatment, E F N 2 O represents nitrous oxide emissions, and E F s l u d g e denotes methane emissions from sludge landfill disposal.

Methane emissions from wastewater treatment are calculated as Equation 2: E F C H 4 = T O W × E F C H 4 (2) where T O W is the total amount of degradable organic matter in wastewater, expressed as biochemical oxygen demand (BOD), and E F C H 4 is the methane emission factor. Due to the lack of consistent provincial-level BOD monitoring data, the BOD/COD ratio recommended by the IPCC (0.43) was adopted to derive BOD values from COD statistics.

Nitrous oxide emissions are calculated as Equation 3: E F N 2 O = N E × E F E × 44 28 (3) where N E is the nitrogen content in wastewater and E F E is the emission factor for N₂O.

Methane emissions from sludge landfill treatment are calculated as Equation 4: E F s l u d g e = S T × L 0 − R × 1 − O X (4) where S T is the amount of sludge generated, L 0 is the methane generation potential, R represents methane recovery, and O X denotes the oxidation factor.

All emission factors and coefficients were taken from the IPCC Tier 1 default values to maintain methodological consistency across provinces. While these parameters do not explicitly capture spatial heterogeneity in wastewater composition, treatment technologies, or climatic conditions, they provide a conservative and standardized estimation framework suitable for comparative efficiency analysis at the regional scale. The limitations associated with this assumption are further discussed in the Conclusion section.

Figure 1 reports the indicator weights derived from the entropy-weighting method. It should be emphasized at the outset that a higher entropy weight reflects greater information variability across provinces rather than a stronger causal impact on CEEWT. Accordingly, the weights are interpreted as indicators’ relative contributions to differentiating regional efficiency performance, rather than as evidence of economic dominance or policy effectiveness.

Among all indicators, Import and Export Value (IEV) exhibits the highest average weight (approximately 22%), followed by Government R&D Expenditure (around 16%). This pattern indicates substantial cross-provincial heterogeneity in trade openness and public innovation investment within the YEB. Coastal provinces such as Jiangsu and Shanghai exhibit significantly higher external trade exposure than inland provinces, while fiscal disparities across regions translate into markedly different levels of government-supported technological innovation.

Several indicators—including GDP, urbanization rate, value added of the tertiary industry, and urban environmental protection expenditure—display relatively similar weights (approximately 10%). This reflects the fact that most YEB provinces maintain relatively high levels of economic development and urbanization compared with the national average, resulting in moderate interregional dispersion rather than dominance by a single dimension.

To address potential concerns regarding indicator redundancy and multicollinearity, variance inflation factor (VIF) diagnostics and pairwise correlation analyses were conducted and are reported in Supplementary Appendix A. The results indicate that multicollinearity remains within acceptable thresholds, supporting the robustness of the entropy-based weighting scheme. In addition, principal component analysis (PCA)–based alternative weighting results are provided in Supplementary Appendix B, which yield highly consistent provincial rankings, further confirming the stability of the CEEWT measurement.

Based on the entropy-weighted TOPSIS results, Figure 2 presents the temporal evolution of CEEWT across the YEB from 2011 to 2020. Overall, CEEWT exhibits a gradually improving trend with noticeable short-term fluctuations, particularly during the 2015–2017 period. Substantial heterogeneity persists across provinces, indicating uneven progress in achieving low-carbon wastewater treatment.

From a spatial perspective, provinces can be classified into three distinct echelons according to their long-term efficiency performance (Table 2). This stratification remains relatively stable over time, suggesting the presence of persistent structural factors shaping regional CEEWT.

Figure 3 illustrates the temporal evolution of average CEEWT across the YEB. The most prominent fluctuation occurs during 2015–2017, when most provinces experienced a temporary decline in efficiency. This period coincides with the nationwide implementation of the Action Plan for Water Pollution Prevention and Control, which mandated rapid expansion and upgrading of wastewater treatment capacity.

In the short run, these policy-driven infrastructure investments increased treatment volumes and associated emissions, leading to a transitional efficiency loss. However, as facilities became operationally optimized, CEEWT rebounded, indicating a delayed efficiency payoff from regulatory intervention.

To formally assess convergence dynamics, σ-convergence and β-convergence tests were conducted using panel data methods (Supplementary Appendix A). The results suggest that while overall efficiency dispersion does not decline monotonically—indicating the absence of strong σ-convergence—there is evidence of conditional β-convergence, implying that provinces with lower initial efficiency tend to improve faster when controlling for structural characteristics. This finding supports the presence of conditional efficiency catch-up rather than unconditional convergence.

Unlike most provinces, Jiangsu maintained continuous CEEWT growth during 2016. Rather than attributing this outcome solely to policy narratives, supplementary fixed-effects panel regressions reported in Supplementary Appendix D provide empirical support for the role of regulatory intensity and technological investment.

The regression results indicate that government R&D expenditure and environmental protection spending are significantly associated with higher CEEWT after controlling for province-specific and time-specific effects. These findings suggest that Jiangsu’s resilience reflects a combination of pre-existing technological capacity, fiscal strength, and policy implementation efficiency, rather than a singular policy shock.

While causal inference remains beyond the scope of this study, the consistency between descriptive patterns and econometric associations enhances confidence in the proposed mechanisms.

Taken together, the results highlight that CEEWT in the YEB is shaped by multi-dimensional interactions among economic development, technological capability, industrial structure, and institutional design. Economic scale alone does not guarantee high efficiency; instead, efficiency gains materialize when development is accompanied by innovation-oriented governance and structural upgrading.

The findings contribute to the broader literature by demonstrating that low-carbon wastewater treatment efficiency exhibits conditional convergence, moderated by regional characteristics, and that policy-induced short-term efficiency losses may precede long-term gains. These insights underscore the importance of coordinated regional strategies to facilitate technology diffusion and institutional learning across heterogeneous regions.

This study examines the spatiotemporal evolution of CEEWT in the Yangtze River Economic Belt using an entropy weight–TOPSIS framework combined with IPCC-based carbon accounting. The results show that CEEWT has generally improved over time, but with notable fluctuations and persistent regional disparities.

Significant spatial heterogeneity is observed across provinces, with downstream regions such as Jiangsu, Shanghai, and Zhejiang consistently outperforming upstream and midstream areas. These differences reflect cumulative advantages in economic capacity, technological accumulation, and institutional governance rather than short-term policy effects. No clear evidence of regional convergence is found, indicating that efficiency improvement remains path-dependent.

Methodologically, this study extends carbon efficiency analysis to the wastewater treatment sector at a large regional scale and clarifies the conceptual distinction between carbon emission efficiency and traditional environmental or energy efficiency. The entropy weight–TOPSIS approach provides a comprehensive and comparable assessment of multidimensional efficiency performance without implying direct causal relationships.

Given the differentiated efficiency patterns across the YEB, policy strategies should be region-specific. For leading regions (e.g., Jiangsu and Shanghai), policy priorities should focus on strengthening low-carbon technology leadership through stricter efficiency standards, advanced treatment technologies, and international cooperation. For intermediate regions (e.g., Zhejiang), improving coordination between urban development and wastewater systems, promoting circular economy practices, and enhancing operational optimization are key to sustaining efficiency gains. For lagging regions, targeted fiscal support, technical assistance, and cross-regional cooperation mechanisms are essential to address structural constraints and improve basic treatment capacity and operational performance. At the basin level, enhanced interregional coordination and differentiated regulation can support a more balanced low-carbon transition of wastewater treatment systems.

Several limitations should be noted. First, carbon emissions were estimated using IPCC default parameters, which may not fully capture regional differences in wastewater characteristics and treatment technologies. Second, the entropy weight–TOPSIS method evaluates relative efficiency but does not identify causal relationships. Third, data constraints required the use of macro-level proxy indicators rather than plant-level technological variables. These limitations point to directions for future research as more detailed data become available.