Journal of Environmental Management最新文献

The Water-Energy-Food (WEF) Nexus, which provides a critical framework for coordinating the management of multiple interrelated fundamental resources, such as water, energy and food, is key to achieving the Sustainable Development Goals (SDGs). Urgently conduct the in-depth research on the spatiotemporal evolution driving mechanism of WEF efficiency, and scientific predictions of its dynamic response in different future development scenarios. Therefore, this study constructed a comprehensive framework that integrates the super-efficiency network SBM model, Geographically and Temporally Weighted Regression (GTWR) model, Explainable Machine Learning model and XGBoost-LSTM model, to conduct the quantitative evaluation, driving factor analysis, and multi-scenario simulation of the WEF system efficiency in the Yellow River Basin (YRB) of China. The results show that: (1) From 2000 to 2022, the efficiency of WEF system and its subsystems followed a declining-then-rising trend; (2) Driven by economic development and ecological resilience, the eastern and southern regions demonstrated higher system efficiency and better subsystem balance compared to the central and western regions; (3) The spatiotemporal heterogeneity significantly shapes the critical impact of government fiscal capacity in improving the WEF system and its energy and food subsystems, while water use control is crucial for enhancing the water subsystem efficiency; (4) Compared to single XGBoost and LSTM models, the XGBoost-LSTM integrated model improves prediction accuracy by approximately 12.11 % and 8.9 %, respectively; (5) Multi-scenario projections based on the XGBoost-LSTM model identify the WEF synergistic enhancement scenario as the optimal pathway for enhancing WEF system efficiency in the YRB. The WEF comprehensive framework proposed in this study can provide case demonstration and technical method for other regions in the world facing similarly resource management and sustainable development issues.

Regulators, companies, and financial players are increasingly focusing on adverse climate events and environmental risks. Using a novel integration of high-resolution geospatial and firm-level financial data, this study provides the first empirical evidence on how rainfall-induced soil loss affects the financial performance and capital structure of Italian agricultural firms. We find that unsustainable soil erosion is associated with significantly lower profitability, manifesting as a decrease of 1.20% in Return on Assets (ROA) and 2.10% in Return on Equity (ROE). Unsustainable levels of soil loss also impair the ability to access external financing, as firms located in these areas exhibit lower levels of external bank financing (-2.00%) and (-3.30%) supplier short-term debt and rely more on equity financing (+4.80%). We also find partial support for the view that unsustainable soil loss impairs a firm's credit risk profile, evidenced by a negative relationship with the interest coverage ratio (-4.69). This research is highly relevant to international studies because it offers a concrete financial framework for understanding the economic consequences of environmental degradation. By providing quantifiable data linking soil loss to a firm's financial health, this study can inform policymakers and regulators globally of the hidden risks in agricultural supply chains. The methodology and insights can be applied to other countries facing similar challenges, providing a basis for considering how sustainable land management practices can contribute to mitigating systemic risks and fostering greater resilience in the agricultural sector.

Nitrous oxide (N₂O) is a greenhouse gas produced during wastewater treatment. Recent advancements in computer technology have facilitated the generation and collection of large-scale multi-source datasets, rendering machine learning (ML) a powerful tool to predict N₂O emissions from these processes. In the narrative review, the emission characteristics and pathways of N₂O from wastewater treatment processes have been reviewed, while summarizing the mechanistic models, ML methods, and hybrid models used for N₂O emission prediction. In particular, the performance of different models in predicting the N₂O emission flux (accuracy), corresponding pathways and key influencing factors has been analyzed. Support vector machine (SVM), random forest (RF), and artificial neural network (ANN) algorithms were the most commonly used N₂O emission prediction models with high performance (R² > 0.90), computational speed, and interpretability. Key influencing factors identified by these models were nitrogen compounds, DO, and C/N, which was consisted with the domain knowledge. Hybrid models of mechanistic and ML algorithms (e.g., Long Short-Term Memory) were superior to the respective individual components in predicting N₂O emission flux and pathways because of the fewer data requirements and higher interpretability. However, the issues of data availability, interpretability, and transferability challenge the applicability of ML models. Hence, further studies on performance improvement strategies (e.g., generative models, interpretable ML, and transfer learning) should be conducted. Nevertheless, the studied prediction methods are important for controlling global warming.

Accurate prediction of riverine water quality is often hindered by sparse sampling and limited streamflow data, a common outcome of resource-constrained watershed monitoring. To address this, we propose a three-module machine-learning framework-prediction (graph neural networks or recurrent networks), interpretation (explainable AI), and management (counterfactual analysis)-and apply it to chromaticity prediction in the Hantan River Basin, Republic of Korea. The dataset includes 1667 monthly observations from 59 monitoring sites (December 2021-October 2024) covering 37 hydro-environmental variables. Performance was assessed using independent training, validation, and test sets. Graph-based models outperformed the recurrent baseline, with the enhanced Graph Sample-and-Aggregate model achieving a test R² of 0.82, demonstrating that representing pollution-source characteristics and transport pathways improves prediction. Interpretability analyses revealed management-relevant insights: PGExplainer highlighted strong upstream influences from the SC sub-watershed, identifying it as the primary intervention region. Feature attribution distinguished long-term influences (e.g., TOC near major WWTPs) from short-term episodic drivers associated with facility-specific effluent spikes. Counterfactual analyses quantified the reductions in effluent chromaticity and proximal indicators required to achieve downstream targets at site HT Y4. Counterfactual success rates-defined as the proportion of model-generated cases meeting the target-were 26 % and 40 % for chromaticity targets of 14 and 15 color units (CU), respectively. Given these outcomes and considering that 14-15 CU is generally acceptable for basin-scale management, a downstream target of 14-15 CU was proposed as feasible and practical. Overall, the framework serves as a cost-effective and interpretable decision-support tool for watershed management under data-limited monitoring conditions.

To solve the problem of intelligent waste classification, a new waste classification detection model named WCD-YOLO (Waste Classification Detection -You Only Look Once) is proposed. We first optimize the backbone network of YOLOv10 by developing an MCA module to heighten the overall model's feature extraction capability and refine the model's precision. At the same time, a more powerful feature extraction module, FNC2f, with high efficiency and multi-scale characteristics, is created to enrich the network's feature extraction and meet the needs of high-precision target recognition. A brand-new FNC2f-BiFPN feature pyramid network structure is also designed, strengthening the detection ability of the waste with insufficient feature capture. Finally, we use Inner-CIoU as the loss function of the WCD-YOLO model and control the scale of the auxiliary boundary. Experimental results show that the WCD-YOLO model developed in this research has better precision on the self-built dataset at two IoU thresholds than other models, with mAP₅₀ reaching 95.8%, an increase of 1.6% over the original model, and mAP_50:95 reaching 74.0%, an increase of 2.6%. The model parameters are only 7.2MB, and the GFLOPs are 8.5G. The proposed model is characterized by low consumption and high precision in waste recognition and classification, providing a reference for future academic research and engineering practice.

Digitalisation and Human Capital (HC) are key drivers of economic transformation, yet their joint influence on environmental outcomes remains underexplored in developing economies. Existing evidence largely focuses on linear or isolated effects, overlooking potential nonlinearities and complementarities crucial for sustainable development. This study investigates the nonlinear and interactive impacts of digitalisation and HC on carbon emissions in 23 developing countries from 2000 to 2023. A multidimensional digitalisation index is constructed using Principal Component Analysis (PCA), integrating digital infrastructure, industrialisation, and innovation. Employing a dynamic system Generalised Method of Moments (GMM) estimator, the study captures endogeneity, persistence, and feedback mechanisms. Results confirm that both digitalisation and HC exhibit U-shaped relationships with carbon emissions. While early stages enhance efficiency and environmental awareness, advanced levels intensify emissions through energy-intensive infrastructure and industrial upgrading. The interaction between digitalisation and HC is insignificant, revealing unrealised synergies in developing contexts. The study contributes to the environmental economics literature by integrating digital and HC dynamics into a unified nonlinear framework and offers policy insights for aligning digital transformation with renewable energy deployment, green skill formation, and institutional coordination to advance low-carbon, human-centric growth.

An important aspect in the management of forests is how the resources should be used. Governance structures put in place for forest management need to reflect resource users' divergent preferences, rules governing resource use to mirror local conditions, and the need for collective choice arrangements to ensure users' participation in appropriation and provision rules. These mechanisms could promote coordination as well as cooperation, and influence preference for resource management and extraction behavior. In this study, we consider forest governance structures that mimic these core issues and examine whether resource users choose to extract more tree resources less often or fewer tree resources more often. Based on a choice experiment on forest resource users involving mainly farmers and youth, a mixed logit model was used to explore preferences and estimate willingness to accept. Findings show that users are willing to switch to forest management regimes that delay resource extraction by allowing more resting periods for trees to rehabilitate. Moreover, results related to our time preference measure imply that users with a lower discount rate are likely to prefer governance structures that increase the resting period. On the contrary, users have negative preferences for resting higher proportion of trees. Instead, they prefer to tap more trees and are willing to pay for it. Overall, results highlight resource users' strong preferences for forest governance structures that allow resource extraction from more trees but less often.

As demand for non-sewered sanitation continues to increase globally, faecal sludge management (FSM) continues to have critical implications for public health and sustainable development. However, understanding of the life-cycle performance of alternative FSM technologies is lacking. Here, we address this gap and present a framework for combined life cycle assessment and cost analysis to systematically quantify the carbon footprint and economic viability of faecal sludge treatment plants (FSTP). We apply the framework to two case study FSTPs in Beijing (FSTP1 and FSTP2), which serve as representative examples of FSM in mega-cities in China and the world. These plants use different FSM technology and process different waste mixtures, with FSTP1 using traditional physicochemical processes to treat single-source faecal sludge and FSTP2 co-processing faecal sludge, food waste, and municipal sludge through anaerobic digestion, combined with biogas and crude oil recovery. Carbon emissions analysis shows that FSTP1 emits 54.7 ± 2.08 kg CO₂-eq/t organic waste (OW), which is considerably lower than FSTP2's net emission of 70.5 ± 8.45 kg CO₂-eq/t OW. This accounts for FSTP2's 43.6 % emission reduction from resource recovery, which partially offsets the plant's otherwise much higher carbon emissions. Transportation distance and grid GHG emission intensity are key factors that affect each plant's carbon footprint. Thus, emissions could be reduced through development of low-carbon faecal sludge treatment technologies and optimization of regional logistics and energy structures. Economic analysis shows that the life cycle costs for FSTP1 and FSTP2 are 79.7 million CNY (≈11.2 million USD)¹ and 282 million CNY (≈39.7 million USD), respectively. Although FSTP2 requires a higher initial investment, its diversified revenue structure and stable treatment fees result in a higher operating income, shortening the investment payback period to just over 10 years, compared to the 14 years required by FSTP1. This indicates the greater economic feasibility of the synergistic treatment and resource utilization approach in the medium to long term. The research findings can inform optimized urban faecal sludge management strategies and selection or promotion of lower-carbon treatment technologies and circular economy models.

Lakes cover only 1.8 % of Earth's land yet emit 0.35-0.55 Pg C yr^-1 as CO₂, 60-120 Tg CH₄, and 60-150 Gg N₂O, about 5 % of global fossil-fuel emissions. This review synthesizes recent advances in lake GHG mechanisms, measurement, and mitigation, and outlines key research directions. Net respiration generally exceeds primary production, maintaining CO₂ supersaturation, while anaerobic methanogenesis dominates CH₄ release; oxic pathways linked to methyl-phosphonate cleavage and cyanobacterial leakage add up to 28 % of CH₄ in warm, P-limited, dissolved organic carbon (DOC)-rich waters. Roughly 60 % of N₂O originates from nitrifier-denitrification. Multi-scale monitoring now combines satellite retrievals, eddy covariance, and buoy sensors, yet global budgets still vary ten-fold because small ponds, littoral "hot spots", and winter ice-outs are under-sampled and gas-transfer coefficients are inconsistent. A range of mitigation measures offers substantial climate-related benefits: (i) enhanced wastewater treatment, urine-diverting sanitation, and vegetated buffers cut CH₄ + N₂O by 15-25 %, while generating climate and water quality co-benefits can value at billions of USD globally; (ii) biomass harvesting and floating macrophytes lower CH₄ + CO₂ by up to 57 %; (iii) micro-bubble destratification or moderate sulfate dosing trim residual CH₄ by 25-60 %. Together, these measures can abate 0.01-0.08 Mt CO₂-eq yr^-1 per 100 ha of intensively managed urban lake, enough to achieve 6-10 % of the Global Methane Pledge. Future priorities are resolving the oxic-methane paradox, standardising monitoring, and applying AI-driven up-scaling to embed lake mitigation within SDGs, turning lakes from overlooked emitters into actionable components of climate and water-quality policy.