Can the polycentric spatial structure achieve pollution and carbon emission reduction in China's urban agglomerations?

Regional spatial optimization is in favor of alleviating the negative environmental externalities generated during urbanization. Investigating its impact on environmental performance is of great significance for improving the livability level of the region and promoting the green and sustainable development of the region. Based on the nighttime light data from 2006 to 2023, this study conducts the rank-size rule and super-efficiency SBM model to assess the urban agglomeration spatial structure (UASS) and carbon emission efficiency (CEE), exploring the direct and spillover effect of UASS on CEE and pollution (POL), examining whether there is a threshold of them. Analysis of empirical results indicates that most cities within the 19 urban agglomerations progressively exhibit a polycentric spatial pattern, with eastern coastal cities demonstrating a more pronounced polycentricity than central and western cities. The polycentric spatial structure significantly promotes enhancing CEE and reducing POL in the majority of cities, but hinders them in neighboring regions. Furthermore, there exists a threshold between UASS and both CEE and POL. When the economic density (ECDE) exceeds 8941.773, a polycentric spatial structure significantly promotes the reduction of both carbon emissions and pollution. When 643.024 < ECDE ≤8941.773, the polycentric spatial structure primarily influences the reduction of pollution, but not necessarily carbon emissions. If ECDE is less than 643.024, a monocentric spatial structure may be more appropriate for reducing carbon emissions and pollution. This study provides valuable insights into urban spatial planning and enhancing environmental efficiency.