[Spatio-temporal Trajectory and Driving Factors of Carbon Emissions Based on Bayesian Hierarchical Spatio-temporal...

The investigation of the spatiotemporal trajectory and driving factors of urban carbon emissions is crucial for achieving carbon peaking， controlling global carbon emissions， and protecting the ecological environment. Compared with the traditional carbon emission-driven identification， the Bayesian hierarchical spatio-temporal model can deal with more complex nonlinear and multilateral variable relationships， better cope with data missing， and improve the estimation accuracy of the results. Based on this， this study uses the carbon accounting coefficient method to measure the carbon emissions of urban agglomerations in the middle reaches of the Yangtze River. The Theil index， gravity center migration model， and spatial autocorrelation analysis are used to explore the spatial and temporal trajectory of urban carbon emissions. Further， the Bayesian hierarchical spatial-temporal model is used to analyze the driving factors affecting carbon emissions. The results showed that： ① The total carbon emissions of the study area increased from 78 459.68×10⁴ t in 2006 to 123 350.56×10⁴ t in 2021， with the rate of increase in carbon emissions slowing down from 1.95% to 1.61%. The Wuhan urban agglomeration was the core area for carbon emissions， accounting for 47.48% of the total emissions. The difference in carbon emissions in the Poyang Lake urban agglomeration was the largest； the focus of carbon emissions changed from south to north. ② The carbon emissions of the urban agglomeration in the middle reaches of the Yangtze River exhibited significant spatial correlation， primarily manifesting as high-high or low-low clustering types， and displayed a spatial distribution characteristic of higher emissions in the west and lower emissions in the east. ③ The degree of influence of driving factors on carbon emissions was ranked as follows： urbanization rate>industrial structure>level of economic development>actual use of foreign capital>energy efficiency>expenditure on science and technology>total population. The positive impact of urbanization rate， economic development level， actual utilization of foreign capital， and energy efficiency on carbon emissions was gradually increasing， the positive impact of industrial structure on carbon emissions was gradually weakening， and the positive impact of science and technology expenditure and total population on carbon emissions was fluctuating. The local changes in carbon emissions in the study area were obviously different， and the overall performance was 'weak in the upper part and strong in the lower part'. The hot spots were mainly concentrated below the study area， the local trend of carbon emissions in the study area was obviously different， and the rapid growth area was mainly distributed in Wuhan urban agglomeration. The study results are significant for understanding the spatiotemporal characteristics of carbon emissions and their driving variables and hold important theoretical and practical implications for the subsequent application of Bayesian hierarchical spatiotemporal models in the field of carbon emissions.