China’s provincial carbon emission peaking pathways and probabilistic forecasting based on the ME-QRNN model

Journal of Chemical Ecology · Springer Nature · 2026-09-25 · DOI: 10.1007/s10651-026-00761-2

Аннотация

Regional differences in China in terms of resource endowments, industrial structure, and development stages are significant, leading to pronounced spatial heterogeneity in the driving mechanisms, evolutionary pathways, and emission-reduction potential of carbon emissions. Accurately characterizing this heterogeneity is of great importance for formulating differentiated carbon mitigation policies. Based on this, this paper develops a Mixed-Effects Semiparametric Quantile Regression Neural Network model (ME-QRNN) that integrates provincial fixed effects and regional heterogeneity effects. It further combines scenario analysis and Monte Carlo simulation, while introducing a Copula approach to characterize the multivariate dependence structure, thereby enabling probabilistic forecasting and uncertainty quantification of future carbon emission pathways. The results show that under the SDP scenario, all Chinese provinces are expected to achieve carbon peaking before 2030. Provinces with low uncertainty are mainly concentrated in the developed eastern coastal regions, whereas provinces with high uncertainty are mostly distributed in old industrial bases and energy-rich regions. At the national level, under the MP scenario, carbon emissions are projected to peak around 2034, with a peak value of approximately 14.227 billion tons. Under the SDP scenario, emissions are expected to peak before 2030, with a peak value of approximately 12.500 billion tons, while under the SEGP scenario, no obvious turning point is observed up to 2040. Overall, this study provides a unified framework that combines interpretability and predictive capability for carbon emission forecasting in China, and offers quantitative evidence to support the formulation of regionally differentiated emission-reduction policies.

Читать у издателя →

Похожие статьи