Search

Addressing the urgent need for accurate carbon emissions forecasting to support global emissions reduction goals, this paper introduces a novel approach for carbon emissions prediction that dynamically considers the spatial–temporal correlation of carbon emissions across forecasting targets. Traditional statistical and machine learning models have limitations, such as oversimplification and inefficiency in capturing evolving dynamics among multiple forecasting targets, alongside a focus on longer forecasting horizons like monthly or yearly. To tackle these, we propose an innovative Dynamic Spatial–Temporal Graph Convolutional Recurrent Network (DSTGCRN), a blend of graph convolutional and recurrent neural network structures customized for up-to-date multi-regional spatial–temporal predictions at a daily level. Comprehensive empirical analyses on datasets from China, the US, and the EU confirm DSTGCRN's superior performance and robustness across diverse geographical contexts. It outperforms the second-best among 10 baseline models, achieving 40.6%, 24.6%, and 38.5% improvements in MAE, RMSE, and MAPE, respectively, across various horizons. The incorporation of environmental data, including temperature and Air Quality Index, as supplementary predictors in the DSTGCRN model, has demonstrated its efficacy in the context of the China dataset. The DSTGCRN model's improved accuracy and daily multi-regional forecasting deepen emission dynamics understanding, supporting the development of informed and region-specific environmental policies that can respond promptly to changes. [ABSTRACT FROM AUTHOR]

Exploring the key influencing factors (KIFs) of Urban Low-Carbon Innovation (ULCI) from urban characteristics is essential for devising customized promotion strategies. However, existing studies are hampered by methodological limitations that lead to an inability to effectively discern KIFs among urban characteristics or unravel complex, non-linear relationships, and interaction effects. To address these gaps, this paper adopts a synthetic approach based on interpretable Machine Learning (ML). Firstly, the influencing factors are identified through the Delphi method and a systematic literature review. Subsequently, three single AI models (KNN, SVR, and CART) and three ensemble models (RF, XGBoost, and LBGM) are employed to fit the data. Finally, the SHapley Additive exPlanations (SHAP) algorithm is integrated to identify the KIFs and disentangle their impact effects. The findings indicate that (1) 41 influencing factors are identified, from which 10 KIFs, such as Expenditure on Research, Carbon Emissions, Local General Budgetary Revenue, and Education Expenditure, are determined, (2) the developed interpretable ML model tailored for ULCI's KIFs analysis demonstrates high precision and effectively capturing non-linear relationships (R2 = 0.841, RMSE = 0.591, MAE = 0.463), and (3) the global impact, interactive effects, and individual sample impact of the KIFs are explained, and two categories of KIFs dominated by positive and negative influences are revealed respectively. Results of the KIFs identification can provide policy-makers with insight for designing ULCI enhancement paths and consequently promote emission mitigation in China. • 41 influencing factors of ULCI is identified from urban characteristics. • An interpretable ML model is developed to quantitatively explore KIFs. • 10 KIFs are determined and the comprehensive impact, interactive effects, and individual sample impact of the them are explained. [ABSTRACT FROM AUTHOR]