This paper extends the Stochastic Impacts by Regression on Population, Affluence and Technology (STIRPAT) model to be a panel error-correction model, which allows for a rich dynamic analysis where changes of urbanization account for carbon dioxide emissions. Applying the mean group (MG), pooled mean group (PMG), and dynamic fixed effect (DFE) estimation, we examine a China province-level data over the period from 1995 to 2011. Our estimates indicate that rapid urbanization increases carbon dioxide emissions both in the short-run and in the long-run. However, the lower magnitude of the environmental impact of urbanization in the short-run than that in the long-run indicates that the process of urbanization would have a long-lasting impact on the growth rate of carbon dioxide emissions. In addition, our results suggest that the impact of urbanization on carbon dioxide emissions would usually take some time to materialize fully, and would be useful to policymakers who seek to coordinate economic development and reduction of carbon dioxide emissions.