Cities are one of many carbon sources. According to the Intergovernmental Panel on Climate Change (IPCC) AR5,CO2emissions from fossil fuel combustion and industrial processes contributed about 78% to the total Green House Gas (GHG) emission increase between 19702010. Total annual anthropogenic GHG emissions have increased by about 10GtCO2-eq between 20002010. The increase directly came from energy (47%) ,industry (30%) ,transport (11%) , and building (3%) sectors,which mainly exist in cities. Urban expansion and urbanization can affect urban carbon emission. Studies show that there is a long-term and stable relationship between urbanization and carbon dioxide emissions. The relationships between urban carbon emissions and indicators,including urban development intensity,urban land use, and the industrial sector,are studied extensively. During urban expansion,the quantitative and spatial features of urban lands can both affect carbon emissions. Therefore,urban form was added to the possible factors influencing carbon emissions in this study,which may be different from previous research that has focused on the relationship between urban growth and carbon emissions. However,in some related research,when urban form has been added to the indicators,the objects were residents or the transport sector,and they lacked quantitative indicators to verify the conclusions. The definition ofurban formin this study was landscape pattern which was characterized by landscape metrics,and the study area consisted of 13 cities in the Yangtze River Delta. In this study,we analysed the shift of the gravity center from 19952015 for carbon emissions of the study area,and defined the decoupling index as well as analysing the temporal-spatial changes of the decoupling relationships between carbon emissions and urban growth in the study area. We also built panel data models to estimate the impact of urban forms on carbon emissions. Based on that,the conclusions are as follows: (1) The shift of the gravity center from 19952015 for carbon emissions of the study area was southwest-northwest-southeast-northwest. The shift may be related to the development of industry and change of industrial structure in some cities during this period. (2) There was a significant temporal-spatial heterogeneity in the decoupling relationships between carbon emissions and urban growth from 19952015. The leading decoupling relationship between carbon emissions and urban growth of the study area changed from expansive negative decoupling to weak decoupling from 19952015. The difference of decoupling relationships between cities narrowed after 2005 and the overall decoupling relationship of the study area became homogeneous. In 2015,almost 70% of cities reached the decoupling state and the decoupling states of Shanghai,Shaoxing, and Taizhou were strong. (3) Urban carbon emissions can be negatively influenced by the dominance of complete patches, and positively influenced by the degree of fragmentation and aggregation of urban patches. Carbon emissions can be more sensitive to the more aggregative distribution of the urban patches. This study analysed the relationship between carbon emissions and urban growth,as well as exploring how urban form can affect carbon emissions. The conclusions could provide scientific references for the policy making of low-carbon development strategies and land use and urban planning of urban agglomeration in the Yangtze River Delta.