Decomposition of carbon emission drivers and urban typology in the Guangdong-Hong Kong-Macao Greater Bay Area: implications for regional environmental management

The Guangdong–Hong Kong–Macao Greater Bay Area is marked by pronounced urban heterogeneity, so that a uniform account of its carbon emission drivers is unlikely to hold. Using panel data for 11 cities from 2015 to 2023, this study applies the Generalized Divisia Index Method to decompose eight drivers of emission change and to classify the cities by their factor contribution structures. Economic growth (G) and output carbon intensity (CI) are the largest positive and negative contributors in ten of the eleven cities (mean G = +19.18%, mean CI = −9.94%), the exception being Macao. An exploratory Ward clustering of the four most variable factors, cross-validated against k-means, distinguishes four types of driving structure—carbon-intensity suppression, economic-scale driven, energy-scale expansion, and a specialized service-structure type (Macao); an innovation index combining research and development (R&D) intensity and patent density varies independently of these types. Decoupling is likewise uneven, ranging from strong decoupling in three cities (Foshan, Zhongshan and Hong Kong; Tapio elasticity t = −0.152 to −0.437) to weak decoupling in six (0.133 ≤ t ≤ 0.462) and expansive coupling in Huizhou and Macao. Per-capita emissions and energy-structure carbon intensity show marked dispersion. These findings inform a three-tiered policy discussion—differentiated measures, collaborative governance, and data support—for coordinated low-carbon governance in urban agglomerations.