Exploring the impact of urban vitality on carbon emission mechanisms using multi-source data

Urbanization and low-carbon development are critical issues of global concern. As urbanization has reached its middle to late stages, cities face the dual pressures of development and environmental challenges. This study constructed a theoretical framework for urban vitality in six dimensions: social, economic, cultural, environmental, spatial, and perceptual. Using methods such as spatial syntax, entropy-weighted TOPSIS, deep learning models, and geographic detectors, we analysed the distribution characteristics of urban vitality in Yantai's central area, explored how vitality-contributing factors influenced carbon emissions, and elucidated the association of urban vitality with carbon emissions. The results indicated that (1) urban vitality exhibited a multicentred distribution pattern of "low in the hinterland-high along the coast"; (2) significant differences existed in the impacts of various vitality dimensions on urban carbon emissions; (3) different urban vitality factors have varying levels of explanatory power regarding the spatial distribution of carbon emissions, with maximum building height exhibiting the strongest explanatory power, while the selection degree shows the weakest; and (4) the interactions between these factors typically demonstrate a two-factor enhancement, with the interaction between maximum building height and integration having the most significant effect on urban carbon emissions. This study innovatively integrates three-dimensional spatial and cultural perception perspectives, addressing the biases found in previous research that represented urban vitality from a singular viewpoint. It provides a more comprehensive framework and methodology for evaluating urban vitality, and the findings can offer recommendations for building low-carbon, high-vitality, and sustainable urban environments.