Research On Carbon Emission Simulation From Residential Buildings Based On Night Light Images

Since one hundred years, the global climate is undergoing significant changes, of which the main characteristic is warming. China is one of the most intense regions, and synchronous with global warming. Global change is one of the most serious challenges the human society facing in the 21 st century. Coping with the climatic variation is gradually becoming the important issue of international negotiations, intergovernmental cooperation and leaders’ talks. Intergovernmental Panel on Climate Change(IPCC) ascribes global change to the rising of greenhouse gases concentrations, especially CO2 concentration increased. Carbon emissions from fossil fuels burning and land use change are considered to be the main cause of global warming. The fourth assessment report from IPCC(IPCC-AR4) indicated that the average global temperature rise since the middle of 20 th century most likely results from the increasing greenhouse gases concentrations caused by human activities. The fifth assessment report from IPCC(IPCC-AR5) further pointed out that the global warming of climate system is accurate, and confirmed the causal relationship between human activity and global warming.In 2009, China has become the largest country in CO2 emissions, surpassing the United States. Resident sector is second big departments considering the proportion of national energy consumption, right behind the industrial sector. If taking the indirect residential energy consumption into account, the ratio of residential energy consumption has been more than 50%. With the continuous growth of resident income and the sustaining upgrade of consumption structure, both the residential energy consumption and its carbon emissions are gradually increasing. The diversity of life energy consumption structure, the differences of household energy consumption and the complexity of the statistical data increase the difficulty in calculating residential energy consumption and carbon emissions. The research scale mostly stays on the provincial and municipal administrative unit, and only a little research has been done at county level scales. It’s quite a big work to calculate the residential carbon emissions using statistical data. Additionally, statistical data is inconsistent and rough at different scales, the data at county and township level scales is unavailable, and the accuracy of data is hard to estimate, which all increase the difficulty of calculating residential carbon emissions in high precision level and across administrative boundaries.With high photoelectric magnification effect, DMSP/OLS sensor is sensitive to the lights in the night, and it can detect faint night light, providing the effective data source in monitoring human activities process. DMSP/OLS data is obtained freely and easily, which has been widely used in population estimation, fire monitoring, urbanization, carbon emissions, etc.In this paper, DMSP/OLS data was used along with DNVI and DEM to simulate the residential area density. And then statistical data was combined to calculate residential carbon emissions in Zhengzhou city. Finally, the spatial distribution pattern of residential carbon emission was obtained based on the residential area density in 1 km grid. The main conclusions are as following:(1) The spatial distribution pattern of residential area density was simulated based on DMSP/OLS night light images as well as NDVI and DEM images by building model, which laid a data foundation for the further research. And it’s confirmed that the spatial data can be well used in the study of social and economic problems, expanding the train of thought in process of the research in social and economic fields.(2) The carbon emissions in the center of city and county town is higher than that of surrounding areas, presenting polycentric spatial distribution pattern in a whole, and some centers are about to connect a line. For example, Shangjie, Xingyang and Zhengzhou city center are closely linked. The carbon emissions of Zhengzhou city center are significantly higher than the others. The highest carbon emission in the area of 1 km2 is up to 14466.55 t CO2. The order of regional carbon emissions was as following: Jinshui > Zhongyuan > Guancheng Hui > Huiji > Xinzheng > Zhongmou > Erqi > Xingyang > Gongyi > Shangjie > Dengfeng > Xinmi.(3) The carbon emissions in Zhengzhou city was divided into four levels using standard deviation classification method. The four levels were as following: low carbon emissions areas, medium carbon emissions areas, relative high carbon emissions areas, high carbon emissions areas. High carbon emissions areas are mainly located in Jinshui, Guancheng Hui and Zhongyuan. Relative high carbon emissions areas are mainly located in Zhengzhou city center. Medium and low carbon emissions areas are located in each region and distributed homogeneously relatively.(4) Carbon emissions per capita of eastern of Zhengzhou city were relatively higher than those of western, and those of northern were higher than those of southern overall. The total carbon emissions of Jinshui were dominant, but its per capita carbon emissions were much smaller than the surrounding areas, presenting a pit in spatial pattern.(5) The spatial patterns both of total carbon emissions and per capita carbon emissions were analyzed using the standard deviation ellipse distribution method. The results showed that the distribution range of per capita carbon emissions is greater than that of total carbon emissions, and the ellipse center of per capita carbon emissions was by west compared with that of total carbon emissions. The factor of low per capita carbon emissions influences the shape of standard deviation ellipse, which indicates that population density restricts per capita carbon emissions in a large extent.(6) Geo Da software was used to analyze spatial agglomeration of total carbon emissions and per capita carbon emissions, respectively. It turned out that both the two showed positive spatial autocorrelation to a certain extent. The spatial correlation between per capita carbon emissions and per capita income, as well as spatial correlation between per capita carbon emissions and per capita GDP, was analyzed. The results showed that the per capita carbon emissions and income were positive spatial correlation, and the per capita carbon emissions and GDP were diametrically opposite, which were negative spatial correlation. The main reason was that regions where per capita GDP was higher had a better economic development, and the population density was much higher. Thus, Not only the technological level of energy conservation and emission reduction was higher, but also the efficiency of energy utilization was improved in these regions. In consequence the regions with lower per capita carbon emissions were surrounded by these areas with high per capita GDP.