Abstract: This paper examines the impacts of population size, population structure, economic level, and technology level on carbon emissions, from 1997 to 2012, in the BTH economic circle by using the STIRPAT model. Results indicate that carbon emissions is positively influenced by economic level, population size, urbanization level, population age structure, population structure of employment, and GDP per capita, but negatively influenced by average family size and energy intensity. Moreover, economic level is the main driving factor of carbon emissions. Average family size is the main factor to control carbon emissions. In terms of population structure, the rapid urbanization contributes to energy consumption, and the effect of age structure acted through increasing supply of the labor and consuming demands. Shrinking average family size increases per capita consumption expenditure and total number of households, thus leading to more energy consumption. Employment structure increases the industry sector employment, resulting in larger carbon emissions. Combining these empirical findings and specific circumstances, some policy recommendations are made.