© 2015 . This paper investigates the spatial dynamics of per capita carbon dioxide (CO2) emissions in China. The analyses are conducted by employing a continuous dynamic distribution approach and panel data of 286 cities at the prefecture and above-prefecture level. The results show that per capita CO2 emissions tend to converge during the sample period of 2002-2011. However, multimodality is found in the ergodic distribution of the full sample. It is also found that there is more persistence in cities with low per capita CO2 emissions, and more mobility in cities with high per capita CO2 emissions. The analyses also show that the dynamics of per capita CO2 emissions are significantly different among various geographical, income and environmental policy groups. The conditional distribution analyses indicate that multimodality cannot be explained independently by any one of the two factors, namely geographical location or income level. The findings in this study may have important policy implications for CO2 abatement in China.