Enhancing climate resilience is essential for advancing China’s comprehensive green transition and the Beautiful China initiative. The synergy between low-carbon city policies and digital policies represents an important institutional innovation for strengthening urban climate resilience (UCR). Using panel data from 284 Chinese cities, this study treats the Low-Carbon City Pilot (LCCP) and the Broadband China Demonstration City Pilot (BCDCP) as a quasi-natural experiment of dual-policy synergy. We employ double machine learning (DML) to identify the effects of policy synergy on UCR and explore its underlying mechanisms. The empirical results indicate that: (Acemoglu 2002) Dual-policy synergy significantly enhances UCR, and its enabling effect exceeds that of either policy alone. This result remains robust across multiple checks (Ahrens et al. 2021). Dual-policy synergy enhances UCR indirectly through technological innovation, talent agglomeration, and capital allocation (Araya-Muñoz et al. 2016). Heterogeneity analysis indicates that the synergistic effect is more pronounced in less-developed cities, non-resource-based cities, and environmental protection priority cities. Moreover, dual-policy synergy generates positive spatial spillover effects on neighboring cities’ UCR, following an inverted “U-shaped” pattern with distance and fading at around 160 km. Overall, this study provides new empirical evidence on the role of digital-green policy synergy(DGPS) in enhancing UCR and clarifies its underlying mechanisms.
In recent years, ambient PM2.5 and O3 have emerged as major environmental concerns, making it crucial to elucidate their spatiotemporal patterns and driving factors. Accordingly, this study characterizes the spatiotemporal evolution and primary drivers of concurrent PM2.5–O3 pollution across the YRDMC, based on hourly observational data from 266 monitoring stations during 2015–2020. The yearly cumulative number of concurrent PM2.5–O3 threshold breaches indicates that joint pollution across the YRDMC declined by 22.28
House gardening has been proposed as one of the available solutions to the global environmental crises driven by the food system. Nevertheless, large scale estimates of potential production and environmental benefits of gardens are currently missing. This study evaluates the overall potential of gardening to provide food and mitigate environmental impacts on the level of the European Union. We map the potential residential gardens in the EU, estimating their potential area at up to 0.7 million hectares. Coupling the garden area with site-specific agro-climatic potential yield, we quantify the potential vegetable production on those gardens when 20 percent of the area is utilized. In such a situation, house gardens could supply around a third of the current EU vegetable demand. Considering life-cycle environmental impacts, substituting conventional agricultural production could lead to significant benefits across all evaluated impact categories. Specifically, direct substitution could lead to greenhouse gas emission savings of 12 Tg CO2 eq. Furthermore, we apply an agricultural market model to evaluate the potential effect of the changes across the food system. These results suggest that market effects would likely curtail the potential benefits. We discuss further unquantified environmental and social benefits of gardening as well as the socio-economic constraints that currently limit a wide adoption of this practice. House gardening could play a substantial role in food system transformation toward sustainability, yet the full potential can be achieved only in the context of wider systemic changes.
Spatially explicit assessments of cultural ecosystem services (CES) in China’s border regions remain limited, although these services are important for ecological security and regional sustainability. This study focuses on Xishuangbanna Dai Autonomous Prefecture in Yunnan Province. It combines CES points of interest (POIs) derived from social media data with key environmental variables. An optimized Maximum Entropy (MaxEnt) model, calibrated using the kuenm package in R, was used to predict the potential spatial distribution of CES. Spatial clustering patterns were then analyzed based on the model outputs. The results show that: (1) Under optimized parameter settings, all CES models achieved good predictive performance, with AUC values exceeding 0.75. (2) Medium and high suitability areas for aesthetic landscape CES are mainly located in the central and western parts of the prefecture, accounting for 27.84
Adoptive social welfare, providing essential accommodation and services for vulnerable groups, plays a vital role in fostering social equity and harmony. This study delves into the relationship between air pollution and adoptive social welfare, emphasizing the impact on the social welfare of disadvantaged groups such as disabled individuals, bereaved elderly, and orphans. Despite increasing global concerns about air pollution and its direct threat to human health and social welfare, its implications for adoptive social welfare remain under-explored. This paper extends the existing literature by examining how and to what extent air pollution affects adoptive social welfare, using county-level panel data from China from 2000 to 2021. The results show that higher PM2.5 concentrations significantly reduce the number of adoptive social welfare institutions and available beds. This finding remains robust after addressing endogeneity with thermal inversions, replacing the core explanatory variable, excluding outliers, and applying a double machine learning model. Heterogeneity analysis indicates that the negative effect is more pronounced in counties with higher urbanization and welfare levels, while counties with lower educational endowment are more vulnerable due to weaker risk-coping capacity. These findings suggest that air pollution is not only an environmental and health issue, but also a welfare and equity issue, highlighting the need to coordinate environmental governance with social welfare protection.
The blue bay initiative (BBI), an essential marine conservation policy in China, advocates for the restoration of coastal environment and facilitates coastal economic and social development. Utilizing intensity difference-in-differences, mechanical effect model, and synthetic control method, we construct a panel database for Chinese coastal cities from 2010 to 2023, to evaluate the policy effect of BBI on the coastal tourism economy and analyze the mechanical channels and spatial effects. We find that the BBI promotes the coastal tourism economy. Meanwhile, BBI impacts coastal tourism economy by strengthening financial support, optimizing industrial structures and stimulating market vitality. The BBI also generates spatial spillovers and heterogeneity effects for the coastal tourism economy. The results suggest that coastal ecological restoration policies should be further refined, and tourism resources need be allocated in a manner tailored to local conditions to boost the coastal tourism economy. Overall, this study examines the policy impact process to establish an interpretive framework for assessing ecological restoration policies and the sustainable coastal tourism economy.
Regional green and low-carbon transitions depend on the coordinated evolution of ecological capacity, institutions, and innovation, yet evidence on how their dominant roles change over time remains limited. This study integrates the Wuli-Shili-Renli (WSR) methodology with Haken synergetics to model ecological resilience, environmental regulation, and green technological innovation as interacting subsystems. Using a balanced panel of 30 Chinese provinces from 2008 to 2022, we estimate pairwise motion equations, identify order parameters under the adiabatic approximation, derive potential functions and synergy scores, and conduct alternative-variable, winsorisation, and temporal-breakpoint tests. Results show that green technological innovation governed system evolution in 2008–2015, whereas ecological resilience became the order parameter in 2016–2022; environmental regulation remained the principal coupling variable in both stages. The shift coincided with a move from innovation-led coordination to resilience-oriented adjustment. Overall synergy rose, regional gaps narrowed, and central China outperformed the eastern and western regions in the later stage, although coordination remained unstable during the transition. Robustness tests preserved the phase-specific order-parameter pattern. These findings indicate that green governance should strengthen ecological carrying capacity, adapt regulatory instruments to the dominant driver at each stage, and differentiate interventions across regions. The framework offers a dynamic basis for designing evidence-based policies under China’s dual-carbon strategy.
This study develops a resilience-oriented framework to evaluate industrial transformation capability (ITC) in resource-based regions. Using kernel density estimation, Dagum Gini coefficient decomposition, convergence analysis, and spatial autocorrelation methods, the study examines the spatiotemporal evolution of ITC in China’s π-shaped Curve Area during 2010–2023. The results reveal persistent and widening regional disparities, driven primarily by interprovincial gaps and differences in transformation competitiveness, particularly digitalization, innovation capability, and industrial ecology. Spatial analysis further indicates increasing core-periphery polarization and stable clustering patterns between innovation-oriented cities and resource-dependent cities. The findings highlight the diagnostic value of the resilience framework. By identifying the specific capacity deficits constraining industrial transition, the proposed approach extends conventional outcome-based assessments and provides a transferable analytical perspective for resource-dependent regions facing resource depletion, price volatility, and decarbonization pressures.
Prefabrication technology is promising and attracts increasing attention in the construction industry to reduce the emissions and achieve carbon neutrality. However, how to evaluate the emissions when preparing the prefabricated components remains as a technical challenge. This study develops a P-G-I LCA model to evaluate emission reduction potential from three concrete walls. The results indicate that prefabricated hollow-core wall, prefabricated wall, and cast-in-place wall have different levels of environmental impact, and prefabricated hollow-core wall has the least carbon emissions contribution. The carbon emissions from the buildings with prefabricated walls were generally lower than those cast-in-place walls, and construction method, structural design and specifications all play a key role in emissions. In addition, from environmental and economic perspectives, the prefabricated hollow-core wall is the optimal choice. The research provides a method to quantitatively evaluate carbon emissions of prefabricated building, and points out a way to mitigate the emissions in the construction industry.
Human activities, dominated by fossil fuel-powered transportation, have unequivocally driven global warming and triggered the ongoing climate emergency (IPCC, 2023). In contrast, electric vehicles, featuring low carbon emissions and high energy efficiency, have achieved rapid development and are currently recognized as a core direction for the global transportation sector transformation. As the stock of retired power batteries, the key component of electric vehicles, is projected to surge substantially in the coming years, and improper disposal of spent batteries poses severe pollution risks to soil, water and air, the recycling of retired electric vehicle power batteries has attracted growing global attention. Meanwhile, as an integral part of global initiatives to achieve carbon neutrality, a variety of carbon emission regulatory policies have been enacted by numerous countries and regions, generating synergistic incentives for promoting resource recycling and reducing greenhouse gas emissions. This study incorporates major carbon emission policies, including mandatory carbon emission policy, carbon tax policy, carbon offset policy and carbon trading policy to construct a low carbon multi-level reverse logistics network model. The proposed model is formulated as a cost minimization problem under the constraints of carbon emission policy. Through the constraints and penalty based carbon costs, emission considerations are effectively internalized. The effectiveness of the proposed model is validated through a real-world case study involving the establishment of a reverse logistics network for recycling waste electric vehicle batteries by a Beijing based new energy automobile manufacturer. Additionally, a sensitivity analysis is conducted on key model parameters. The results demonstrate that carbon emission policies significantly influence decision-making processes related to battery recycling. Moreover, under different regulatory frameworks, both the economic costs and environmental benefits associated with waste electric vehicle battery recycling exhibit notable variations.
Open burning of rice straw remains widespread across many rice-producing regions, generating avoidable greenhouse gas emissions, air pollution, and soil degradation. Yet the transition to sustainable straw management remains limited, especially in climate-stressed smallholder systems. This study explores farmers’ behavioral readiness and economic preferences to adopt rice-straw incorporation, willingness to pay (WTP) for its adoption, and identifies how socioeconomic, farm and climatic factors shaping these decisions. Using a combination of contingent valuation method, ordered logit, Heckman selection, and double-hurdle models, we analyze survey data of 435 rice-farmers collected through stratified-random sampling from three climate-hazardous (i.e., drought, salinity, and flood) districts and one less climate-risk district of Bangladesh. Adoption of straw-incorporation is very low (14
Megacities are experiencing acute governance challenges and uneven development, while the traditional governance frameworks are often insufficient to reflect the modernization of urban governancecapacity. To address this gap, a novel governance framework of Development-Autonomy-Inclusiveness (DAI) was constructed in this study, to conceptualize the urban governance capacity modernization (UGCM). Using an integrated measurement approach, UGCM is assessed for 19 Chinese megacities for the period 2013–2022. The coordination capacity and its spatiotemporal evolution are depicted and the major obstacles are identified. Empirical evidence shows that: (1) The UGCM level remains relatively low, with development capacity ranking highest, followed by inclusiveness capacity and autonomy capacity ranking lowest. Megacities with high, moderate, and low UGCM exhibit development-led, autonomy-supported, and inclusiveness-guaranteed patterns respectively; (2) Inclusiveness capacity is the major coordinating factor. The coordination capacity displays a spatial declining trend from east to west; (3) Development capacity is both a key contributor and major obstacle. The UGCM is constrained by 13 obstacle factors, including the number of artificial intelligence patents, disaster governance and emergency response capability, etc. Three types of cities should adopt differentiated strategies: tri-capacity synergy, autonomy promotion, and inclusive development. The findings of this study provide a new perspective, conceptual framework, and empirical evidence for understanding the modernization of megacity governance capability in China, and also offer a meaningful reference for governance modernization efforts in the metropolitan of other countries.
The Loess Plateau region exhibits extreme ecological vulnerability, where vegetation serves as the primary entry point for energy flow and the central hub of material cycling within its ecosystems. Consequently, deciphering the drivers of vegetation dynamics is critical for advancing regional ecological restoration research. While existing studies predominantly focus on the plateau-wide scale, they often overlook spatial heterogeneity in driver importance, threshold effects, and interaction mechanisms across its internal sub-regions. Based on multi-source remote sensing and ground observations, this study employed machine learning combined with the SHapley Additive exPlanations (SHAP) method to reveal the key drivers of vegetation change, their thresholds, and their interactions across different spatial regions of the Loess Plateau. The results showed that from 2000 to 2020, approximately 59.2
Exploring the spatiotemporal evolution and key influencing factors of environmental governance efficiency (EGE) in industrial and mining cities is conducive to the improvement of relevant theories and the formulation of environmental protection policies. This paper comprehensively employs the global super-EBM (Epsilon Based Measure) model with undesirable outputs, the GML (Global Malmquist-Luenberger) index, the Gini coefficient and KDE (Kernel Density Estimation), as well as dynamic QCA (Qualitative Comparative Analysis) and NCA (Necessary Condition Analysis) methods to measure the EGE of 45 industrial and mining cities from 2013 to 2022, and conduct an analysis of spatiotemporal changes, regional differences, distribution characteristics, and driving factors. The outcomes showcase that: (1) The overall EGE of these cities demonstrated a generally increasing tendency. The high EGE demonstrated a trend of spatial diffusion from local concentration to the whole country. Meanwhile, the total factor productivity of environmental governance in these cities has generally shown an upward trend, and it is mainly driven by technological progress. (2) The differences in EGE among these cities in China have fluctuated, but the overall trend has been a reduction. (3) No single variable condition has been found to achieve high EGE in these cities. (4) In the configuration analysis, configurations H1, H2, H3.1, H3.2, H3.3, H4.1, and H4.2 were obtained, which can be classified into four types of high EGE driving factor combinations: social-driven, government-society-driven, market-society-driven, and government-market-society-driven. Finally, several policy suggestions are put forward.
Clean energy transitions increasingly rely on collaborative innovation across organizational and regional boundaries, and such collaborative processes are inherently embedded in relational networks rather than confined to isolated regional systems. While existing studies highlight the role of green innovation in environmental performance, less attention has been paid to how the structure of clean energy innovation networks shapes regional green development. Drawing on knowledge-network and regional innovation system perspectives, we examine whether broader and deeper positions in clean energy innovation networks are associated with regional green development through interregional flows of technology, R D personnel, and R D capital. Using inter-provincial collaborative patent data for 30 Chinese provinces from 2008 to 2022, we construct China’s clean energy innovation network and employ social network analysis and two-way fixed-effects models to examine its environmental implications. The results show that the network has evolved from a monocentric to a more polycentric structure. Both the breadth and depth of innovation cooperation are significantly associated with lower PM₂.₅ emissions and higher green development efficiency, with technology, personnel, and capital flows acting as key transmission channels. These findings highlight innovation networks as relational infrastructures shaping green transition outcomes beyond local innovation capacity.
Mobile source pollution is a major contributor to air pollution, so it is significant to evaluate the environmental effect of transportation policies for facilitating their optimization. The concentration of air pollutants is influenced not only by emissions but also by meteorological factors, so raw air pollution monitoring data cannot be directly used to evaluate the environmental effects of policies. To further explore the true effect of transportation policies on pollutant emission reduction, this study develops a WN-Ensemble-SCM-DID model, which integrates machine learning algorithms and intelligent optimization algorithms with a synthetic control method. This integrated approach is used to analyze the meteorologically normalized pollutant concentration data and evaluate the actual impact of traffic policies on the emission levels in the target city. In this study, the Traffic Power Construction Outline is used as a case to test the feasibility and effectiveness of the proposed model. The results show a significant downward trend in PM2.5 and PM10 levels in both Shenzhen and Chongqing, as well as in NO2 levels in Chongqing. WN-Ensemble-SCM-DID serves as a reliable machine learning-based evaluation tool for assessing the environmental effects of policies. Moreover, the results support the conclusion that the Traffic Power Construction Outline has a positive impact on reducing fine particulate matter emissions within two years of implementation, providing a valuable reference for policy promotion and future related decisions. This study provides a machine learning-based causal model for policy effect evaluation. WN-Ensemble-SCM-DID can be effectively evaluate the environmental effects of transportation policy. ‘Traffic Power Construction Outline’ reduced the emission of PM2.5 and PM10.
Establishing a low-carbon sustainable manufacturing system necessitates a profound understanding of how emerging digital technologies reshape pathways to green production. Despite artificial intelligence (AI) demonstrating formidable transformative potential, systematic evidence remains scarce regarding its interactive relationship with the green transition in manufacturing—particularly concerning nonlinear coupling development patterns and their spatially heterogeneous driving mechanisms.Based on China’s provincial-level balanced panel data from 2011 to 2023, this study constructs a Coupling Coordination Degree (CCD) analytical framework. Integrating Dagum Gini decomposition, kernel density estimation, Markov chains, and geographic detectors, it systematically characterises CCD’s spatiotemporal evolution and its driving mechanisms.Findings reveal: CCD has continuously improved, with spatial agglomeration effects intensifying; National disparities show an overall convergence trend, yet inter-regional differentiation remains the primary source of imbalance.Spatial Markov chains reveal pronounced neighbourhood effects: high-level regions catalyse surrounding areas’ advancement, while low-level neighbours tend to induce development lock-in. Economic growth, population agglomeration, and urbanisation constitute core drivers of CCD enhancement; these factors exhibit stronger synergistic amplification through mutual interactions. More importantly, the spatial coupling mechanisms uncovered in this study not only expand systematic understanding of AI-enabled green industrial transformation but also provide universally applicable and transferable insights for developing nations worldwide. These insights support the construction of intelligent, green manufacturing systems at urban and regional scales, thereby advancing the achievement of the Sustainable Development Goals (SDGs).
This study focuses on the coupled impact of wildfires and climate change on BGW resources in the Weihe River Basin(WRB) in China. Based on the Soil and Water Assessment Tool (SWAT), two climate change scenarios (SSP126 and SSP585) and eight wildfire-climate coupled scenarios (Fr1–Fr8) were constructed to systematically analyse the spatiotemporal evolution of blue and green water (BGW) resources. The key findings reveal that, relative to the baseline period (2000–2020), multi-year average blue water (BW) increased by > 55.3
Under China’s National Unified Carbon Emissions Trading System (CN-ETS), regional disparities have raised questions about institutional fit and environmental justice. This study constructs a Regional Policy Friendliness Index (RPFI) that integrates ecological, economic, and institutional dimensions. Using PCA-based weighting validated by EWM, Coupling Coordination Degree (CCDM), and Obstacle Degree models, we evaluate the carbon-market adaptability of 30 provinces from 2011 to 2023. The results show four main patterns. First, RPFI rises unevenly and retains a persistent “High-East, Low-West” gradient, indicating potential regressive effects on underdeveloped resource-based provinces. Second, ecological endowment and economic capacity remain spatially mismatched, while institutional design is the dominant driver of policy friendliness. Third, Benchmark-Leading provinces (high RPFI, high coordination) show benign coupling, whereas Structure-Locked provinces (low RPFI, low coordination), mainly in central and western China, face a “low friendliness + low coordination” dilemma. Fourth, historical emissions and carbon productivity remain rigid barriers, with quota gaps becoming increasingly binding. The study provides a quantitative perspective on regional inequality and suggests that a just transition requires differentiated quota allocation, partial use of auction revenues for a Just Transition Fund, and ecological compensation for carbon-sink regions.
Against the backdrop of China’s “dual carbon” goals and the rapid development of the digital economy, this study examines the interaction between corporate digital transformation and green transformation and their implications for ESG performance. Using panel data on 2,653 Chinese A-share listed manufacturing firms from 2011 to 2024, we examine the bidirectional relationship between digital and green transformation, test their synergistic effects, and develop a coupling coordination index to capture the degree of coordination between the two transformations. A three-dimensional kernel density estimation is further employed to characterize the dynamic evolution and regional heterogeneity of digital-green synergy. The results indicate that: (1) digital and green transformation exhibit a significant bidirectional positive relationship and both significantly enhance corporate ESG performance; (2) their interaction generates a significant synergistic effect that further amplifies ESG performance; and (3) these effects are heterogeneous across industry characteristics, ownership structure, financing constraints, and market competition, while the synergistic effect remains robust across subsamples. The kernel density results further show a steady improvement in firms’ digital-green synergy, though notable regional differences persist in its evolution and distribution. This study contributes to the literature by providing evidence on the bidirectional relationship and synergistic effects of digital and green transformation and offers policy insights for fostering coordinated transformation and sustainable corporate development.
