全球变化知识资源中心

Carbon dioxide (CO2) emissions have a significant impact on global climate change. Previous studies have mostly focused on the impact of economic performance of sectors on carbon emissions, however, there is a lack of information about the influence of the structural roles on the full industrial chain. This article first built global embodied carbon emission transfer networks from 2000 to 2015, combining multi-regional input-output analysis and complex network theory, to explore the structural characteristics and recognize the structural roles of sectors. Then, panel regression analysis was applied to quantify the contributions of the structural roles of sectors to their carbon emissions. The results first showed that there were significant scale-free network characteristics in the global embodied carbon emissions transfer networks. The majority of global carbon emission transfers was concentrated on a limited number of sectors over time. Moreover, the network analysis revealed that sectors played different roles with different levels of importance in the global embodied carbon emission transfer networks. The results also suggested that the carbon emissions of sectors related to economic activity were significantly determined by some structural roles in the global embodied carbon emission transfer networks. Degree in, strength-out and closeness-out and betweenness had significant positive influences, while degree out, strength-in and closeness-in had significant negative impacts. These results indicated that carbon emissions mainly were increased by the domestic markets rather than international trades. Thus, adjusting the domestic consumption structures can be helpful for global carbon emission reduction. It also indicated that carbon emissions were not only driven by direct production demand from downstream sectors, but also were significantly influenced by indirect production demand in the industrial chain. Therefore, reducing carbon emissions should be established in both the direct and the indirect relationships in the production process. (C) 2018 Elsevier Ltd. All rights reserved.