On the spatio-temporal distribution regularity of the carbon emissions based on the GIS in China

The paper intends to offer the research results of the spatio-temporal distribution statistics of carbon emissions,such as the carbon emissions of the different provinces and regions in the country,the regional difference over the carbon emissions,with the global and local spatial autocorrelation data of the carbon emission intensity per land area in China from 1980 to 2011. We have also worked out the amount of carbon emissions each year over the provinces of China from 1980 to 2011 by using the data of the national carbon emissions and Chinese provincial fossil fuel consumption and cement output published by CDIAC. The results of our calculation data indicate that the carbon emissions of the country both at the national and provincial levels represent obvious phase distribution characteristic features,for example,the slow growth stage from 1980 to 1997,the fluctuant growth stage from 1997 to 2002 and the fast growing stage from 2002 to 2011. The regional difference in the carbon emissions shows a trend fromhigh to low and then another high to low. In other words,the difference is biggest in 1980 s,gradually increasing when returning to the lowest level in 1999,and then reaching the maximum difference again on the local conditions. What is more,the global spatial autocorrelation has also been found in the carbon intensity per land unit area in the three distinct stages,namely,the fluctuating growth stage from 1980 to 1994,the fluctuant downward stage from 1994 to 2001,and the fast growth phase from 2001 to2011; the carbon intensity per land area shows the positive spatial autocorrelation mainly due to the lowering or downward-falling concentration whose trend is changing and gradually getting stabilized after 2006. Under the significant level of 0. 05,the characteristics of the carbon intensity per land-area eventually shows the significanthigh-highcluster area in Beijing,Jiangsu and Zhejiang,whereas the significantlow-lowcluster areas can be found in Gansu,Qinghai,Sichuan,Xinjiang,Ningxia and Yunnan,with lots of non-significant areas under the stable distribution.