Livable Cities - London AMPS | City, University of London Page 114 reaching the minimum of 0.52 and the maximum of 0.98 in 2017 and then decreased to around 0.85. Overall, the model fits best in cities in the southwest and less well in the northeast. Figure 3. Local R2 results for 2015, 2017, 2019, 2021. GDP per capita In Figure 4. (a), the different shades of colors reveal the extent to which GDP per capita (PCGDP) contributes to carbon emissions in different cities. Dark blue shades in northeastern regions such as Dazhou, Bazhong and Guangyuan, indicate that PCGDP had a greater negative impact on carbon emissions compared to cities in the southwest of CCEC. From the time span, the negative impact of PCGDP on carbon emissions gradually intensified during 2012- 2016. After a brief reduction in 2017, the negative impact intensified again and even turned positive in southwestern CCEC cities such as Ya'an, Leshan and Yibin. Imports and exports It can be seen from Figure 4. (b) that between 2012 and 2017, the total imports and exports of the cities in the northeast showed a positive correlation on carbon emissions, reflecting that foreign trade in these cities was relatively more developed, and the import and export industry had a greater impact on carbon emissions. Since 2018, imports and exports turned to have a negative impact, which was more pronounced in the southwestern cities. GDP share of the secondary industry In Figure 4. (c), The area with darker red color shifted from the northeast to the west. In 2012, cities with intensive energy-consuming industrial activities, such as Chongqing, Dazhou, and Guang'an in the eastern region, produced relatively more carbon emissions. By 2015, the contribution of the southwestern cities exceeded that of the east. Nonetheless, in general, there wasn’t a significant geographic difference among the CCEC cities in any given year. And the positive impact of the secondary industry GDP share on carbon emissions gradually decreased over the entire period. Total Electricity Consumption Figure 5. (d) demonstrates how total electricity consumption positively impacted urban carbon emissions. In the first five years, its impact was geographically unstable but then significantly positive in the second five years in the eastern cities, especially Chongqing. Throughout the ten years, the positive contribution rose steadily, which may be a result of rising energy demand. Population density As it is shown in Figure 5. (e), from 2012 to 2016, the positive influence of the northeastern region exceeded that of the southwest, and the gap was small. However, in the latter five years, the positive influence became more significant in the southwest, and the gap between the regions widened.
