Study On The Carbon Emissions From Energy Consumption In China Using DMSP/OLS Night Light Imageries

According to the United Nations Statistics of 2010, China has become the largest CO2 emitter in 2008. Maintaining the great amount and rapid rate of CO2 emissions, China would probably product even-larger CO2 emissions in future, which attracted push great pressure Chinese government. Due to the significant differences in economic and carbon consumption between different regions, the first step for making a reasonable mitigating target of CO2 emissions and making robust carbon reduction planning was to investigate the spatial and temporal characteristics of China’s CO2 emissions and reveal the potential driving forces. However, most of current researches depended on the statistical data published by China’s national, provincial and municipal Bureau of Statistics. As the statistical approach on data collection, reporting and validation was opaque, there were always significant discrepancies between national, provincial and city-level official statistics of energy consumptions. Additionally, the city-level statistics of energy consumptions were always blank in some special decades, especially in underdeveloped regions. Therefore, it is bring great difficulties for the Chinese government to make accurate, comprehensive and differentiate planning of CO2 emission reductions. So, in order to rationally assessing China’s CO2 emissions, new spatial information estimating methods are urgently needed.DMSP/OLS Nighttime stable light(NSL) images, which are capable of detecting the low lights from urban, small human habitats, and even the vehicles, are proven to be effective in monitoring the intensities of human activities. The usage of DMSP/OLS NSL to estimating CO2 emissions have been proved feasible by international researchers. But such studies are still rare and also at the initial stage. Moreover, most studies mainly focused on the globe, country, and provincial scale carbon emissions. Fewer have pay attention to the city scale.To fill up the gaps of historical city-level CO2 emissions and unify the national, provincial and city-level statistical data of China, this paper aimed at the following objectives:(1) to develop a normalized faithful approach for accurately assessing China’s city-level CO2 emissions of energy consumptions using DMSP/OLS NTL imageries:(2) to analyze the spatial and temporal dynamics of CO2 emissions of cities at different development stage and pattern;(3) to study its internal relationship with the government polices, energy structures and industry structures, then reveal the major driving forces, and then propose corresponding potential mitigation policies.(1) A neighborhood statistics analysis(NSA) method is developed for extracting the built-up urban areas using DMSP/OLS NTL imageries. The proposed method is rooted in the terrain relief analysis method of DEM images, using the Neighborhood Statistics Analysis method in the surface analysis tools of Arc GIS, and multiple raster calculating methods. Multi-temporal built-up urban areas that were extracted from the Landsat TM/ETM+ images in the same period were selected to evaluate those extracted from the DMSP-OLS NSL data. The root mean squared error(RMSE) is 193.56 pixels, with relative accuracy of 87.21% and kappa coefficient of 0.731. There are also significant positive correlations in LSI(R2 = 0.46, p<0.001), CONTIG(R2 = 0.38, p<0.001), PARA(R2 = 0.82, p<0.001), and CONNECT(R2 = 0.96, p<0.001) between DMSP-OLS NSL- and Landsat TM/ETM+-derived built-up urban areas. The proposed NSA method significantly overcomes the disadvantages that are associated with the threshold-based methods; it accurately maps both the large patches of built-up areas in urban regions and the small patches of built-up areas in surrounding towns.(2) A new method of accurately mapping the city-level energy-related carbon emissions is developed based on the DMSP/OLS NTL imageries. The proposed method firstly extracts the built-up urban area and obtains the total night lights(SDN) using the NSA built-up extracting method. By conducting the correlation analysis with statistical energy-related carbon emissions, we developed an accurate linear model for estimating China’s city-level energy-related carbon emissions. Results show that the SDN of all cities correlate well with the statistical energy-related carbon emissions(RMSE=943.79×104t; RE=7.65%; R2=0.818). Thus, the proposed method is feasible in monitoring the spatiotemporal pattern of China’s energy-related carbon IXemissions.(3) The spatiotemporal patterns of China’s energy-related carbon emissions are comprehensively studied. During the period 1992-2010, China’s total CO2 emissions increase continuously from 1.49 ×1010t in 1992 to 1.01×1011t in 2010, at an average annual growth rate of 11.22%. The increase rate differs from different cities. The SLOPE index methods indicate that most cities in Eastern China along the coastal lines belonged to the moderate and rapid growth type, while most cities in Western, Central and Northern China belonged to the slow growth type. The global Moran’s I index analysis suggests that China’s CO2 emissions became more and more clustered during the past 19 years. Local Moran’s I indices of each city are further calculated to analyze the regional spatial clustering patterns of China’s current CO2 emissions between neighborhood cities. Result show that the High–High cluster phenomenon is clearly identified in the coastal regions, while the Low-Low spatial clusters mainly are located at the western regions(4) The potential mechanisms influencing China’s energy-related carbon emissions are studied in this study, by analyzing the total amount of carbon emissions, carbon emission intensities per capital, and carbon emission intensities per GDP. The total amounts of CO2 emissions in Eastern, Central China and Western, Northern China are mainly determined by the regional total GDP and greatly related to national polices. The six city clusters also show the similar trends of carbon emissions. The CO2 emissions of per capita(PCCE) in four major economic regions basically maintained the following orders: Eastern China >Northeast China> Western China> Central China. The CO2 emissions of per GDP(PGCE) of Northeastern and Western China were much higher than that of Eastern and Central China. In order to further study on the carbon emission characteristics for making regionalized mitigation suggestions, the double mass curves of cumulative energy consumptions against cumulative CO2 emissions of 10 representative cities or provinces(Beijing, Tianjin, Guangdong, Shanghai, Henan, Hubei, Heilongjiang, Liaoning, Shanxi and Yunnan) are produced. The CO2 emissions of per energy consumption(PECE) of Beijing and Tianjin city in Beijing – Tianjing- Tangshan economic zone around China’s capital changed little during these years. As for Guangdong province and Shanghai city, which were representatives of the Southern and Eastern coastal economic zone, the PECE decreased obviously during 2000/2005 period. However, the PECE of Central(Henan and Hubei provinces), Northern(Heilongjiang and Liaoning provinces) and Western(Shanxi and Yunnan provinces) regions all increased significantly since the 2009-2010 periods. Results reveal that GDP increment is the major factor determining the carbon-emission growth rate, while industry structures and energy efficiencies are the major factors influencing regional CO2 emission intensities.(5) Considering the different economic development levels, industry and energy structures, this paper propose regionalized polices for different cities to mitigate carbon emissions. The major efforts should be focused on optimizing the industrial structures in Eastern and Central China where industries mainly belonged to technology-intensive, labor-intensive and light industry types, and guiding companies to increase the energy efficiencies in Northeastern and Western China, where industries mainly belonged to heavy and energy-related types.