Structure and driving factors of spatial correlation network of agricultural carbon emission efficiency in China

The study of agricultural carbon emission efficiency is important for the realization of agricultural carbon peak and carbon neutrality goals. There is a lack of studies on agricultural carbon emission efficiency based on relational data and network perspectives. These limitations restrict the development of regional agricultural collaborative emissions reduction activities. Therefore, based on relational data and network perspective, taking the development of the agricultural carbon emission efficiency of 31 provinces (cities and autonomous regions) from 2010 to 2019 as the research subject, the study used the SBM-Undesirable model to measure the efficiency of agricultural carbon emissions, constructed a modified gravity matrix of spatial correlation network of agricultural carbon emission efficiency, analyzed the structural characteristics of the spatial correlation network by applying the social network analysis method, and finally explored the driving factors through a quadratic assignment procedure (QAP) model. There are several main findings. First, despite the wide disparity across the 31 provinces (cities, autonomous regions) in China, agricultural carbon emission efficiency increased rapidly, from 0.400 to 0.756, increasing 88.8% with a creation room for improvement. Second, the network relevance of agricultural carbon emission efficiency in the provinces (cities, autonomous regions) was enhanced. For the spatial correlation networks of agricultural carbon emission efficiency in the 31 provinces (cities, autonomous regions) of China, the number of network relations increased from 121 to 211, and the network density increased from 0.130 to 0.227, while network ranking declined from 0.458 to 0.293, followed by network efficiency, which declined from 0.837 to 0.692. In addition, the spatial correlation network of agricultural carbon emission efficiency among the 31 provinces (cities, autonomous regions) had formed multiple network centers that played an important role in controlling agricultural carbon emission efficiency. Overall, the eastern coastal areas were the main destinations for cyberspace space-related spillover of agricultural carbon emission efficiency in 31 provinces (cities and autonomous regions) in China. Third, the transport-level difference, resident income difference, difference in the output value of the first industry and information-level difference had an important impact on the formation of a spatial correlation network of agricultural carbon emission efficiency in China. Finally, the study findings demonstrated that the differences in transportation level and the output value of the primary industry significantly promoted spatial correlation network development. Similarly, it was found that per capita income, information level, and spatial distance also emphasized spatial correlation network formation. Based on the research conclusions, we proposed some suggestions for enhancing the spatial correlation of agricultural carbon emission efficiency, such as emphasizing the development of inter-regional coordinated emission reduction activities and differences of various provinces (cities and autonomous regions) in spatially related networks, making full use of driving factors strengthening the connection between the agricultural product market and organizations, and enhancing the information and transportation network support.