ZHANG Jin-ping, QIN Yao-chen, ZHANG Yan, ZHANG Li-jun
Under the background of the urban low-carbon as a basic strategy for sustainable development, urban carbon dioxide emission structure and low-carbon standard is not only the basic point of view shifting to low-carbon economy and society, but also the methodological basis for micro-scale study on the low-carbon city. This research emphasizes mainly on driving factors of urban CO2 emission, the carbon cycle and metabolism, planning for low-carbon city, and environmental benefit-governance of low-carbon city. The urban carbon dioxide emission models have been developed such as logarithmic mean Divisia index method model, urban carbon flux balance model, Hybrid-EIO-LCA model, scenario analysis model and computable general equilibrium model. The research in China involves development strategies for low-carbon city, the assessment on low-carbon standard of city, CO2 emissions from urban residents, spatial planning of low-carbon city, the structure of urban carbon circulation, urban carbon footprint, and so on. The theory in China is lagging behind, especially in urban CO2 emission accounting with Shanghai City, which is the main region for practice. There are no inter-city comparison and short-term prediction, besides, urban CO2 emission structure is not clear enough. Taking the calculation methods adopted by '2006 IPCC Guidelines for National Greenhouse Gas Inventories' as references and comprehensive consideration of various uncertain factors, urban CO2 emissions and low-carbon standard in Beijing, Tianjin, Shanghai and Chongqing cities from 1995 to 2013 were measured by mathematical model combining with BP neural network in this paper. The study shows that, primarily, CO2 emissions in four cities increase year by year with great differences. Secondly, urban CO2 emissions and the trends depend on CO2 emission structure and its dynamic evolution in four cities. Thirdly, the measurement for low-carbon standard shows that the role of industrial structure optimization and upgrading is significant to improve the low-carbon standard, though economic growth in four cities still depends on the carbon-based energy consumption. Finally, BP neural network is more reasonable and accurate than the traditional method when used in short-term prediction.
