Analyzing the spatial association of household consumption carbon emission structure based on social network

In recent years, the energy consumption and associated carbon emissions from household consumption are increasing rapidly. It is an essential indicator to evaluate the extent of building a low-carbon society in China under the background of carbon peaking and carbon neutrality. Thus, we firstly calculate the information entropy of direct household consumption-induced carbon emission structure (IDHCES) in China during 2005–2019. Secondly, the spatial association network of the IDHCES is constructed by using the modified gravity model. Finally, we apply the social network analysis (SNA) to investigate spatial association characteristics of the spatial association network and explore influential factors by constructing the quadratic assignment procedure (QAP) model. There are four primary discoveries: (1) The balance of inter-provincial direct carbon emission structure from residential consumption is quite different. And the spatial linkage of the IDHCES is not just geographical proximity, but shows the complex network pattern. The extent of this network linkage is getting higher over time. (2) The spatial association network of the IDHCES presents an evident core-edge distribution. Most of the eastern provinces situated at the core of this network, such as Shanghai, Beijing and Tianjin, play essential roles, while most of the central and western provinces such as Qinghai, Guizhou, Xiangjiang and Ningxia are on the edge and have slight influence to this network. (3) The spatial association network for the IDHCES can be divided into four blocks, which are strongly related to each other and have obvious stepwise spillover effects. (4) The expansion of differences in per capita GDP, energy consumption per unit of GDP, family size and government investment in science and technology promotes the formation of the spatial association network of the IDHCES. While, the expansion of differences in geographical distance, population density and engel coefficient acts as a barrier. Based on the above analysis, we put forward some related suggestions for optimizing the information entropy of the direct carbon emission structure from Chinese residents’ consumption.