Investigating the spatial correlation characteristics of CO2 emissions is crucial for promoting regional collaborative emission reduction. This study examines 334 Chinese cities, employing an adjusted gravity model to construct a carbon emission spatial correlation network (CESCN). Utilizing social network analysis (SNA) methods, it systematically examines the network structure characteristics, dynamic evolution patterns across 2005, 2010, 2015, and 2020, and the driving factors for 2020. Key findings include: (1) China’s CESCN exhibits overall characteristics of “high resilience, low density, and weak hierarchy.” Inter-city carbon emission linkages are predominantly passively driven by macroeconomic activities, while networks based on proactive collaborative emission reduction remain underdeveloped. (2) The network demonstrates distinct node role differentiation and spatial clustering, with eastern coastal developed cities like Shanghai, Shenzhen, Hangzhou, and Suzhou serving as core hubs. (3) The network’s core-periphery structure underwent a clear evolution from being primarily geographically driven to being economically driven, with the core zone stabilizing in the more economically developed eastern region. (4) geographical distance, disparities in economic development levels, and differences in population size inhibit the formation of CESCN; conversely, variations in industrial structure and carbon intensity exert a positive promoting effect. By constructing a more comprehensive urban sample network, this research deepens the understanding of the overall spatial correlation patterns of China’s CO2 emissions, providing empirical evidence and decision-making references for optimizing regional low-carbon collaborative governance policies.
Yang et al. (2026) studied this question.