| The global environmental problems caused by carbon emissions have attracted close attention from the international community.As the largest developing country,China ’s carbon emission problem has become the main reason that restricts the green and coordinated development of China’s economy.Accurately obtaining the spatio-temporal dynamic information of carbon emission intensity and analyzing the influencing factors of carbon emission intensity are the premise of exploring energy conservation and emission reduction strategies suitable for China ’s national conditions.In order to achieve this goal,an estimation model was constructed based on long-term nighttime light data sets and carbon emission statistics.Taking 282 prefecture-level units as sample cities,the spatial and temporal evolution of carbon emission intensity was explored.The spatial spillover effect model of urban carbon emission intensity was constructed by using Markov model,and the influencing factor model was constructed by combining Geographically and Temporally Weighted Regression(GTWR)model and STIRPAT(Stochastic Impacts by Regression on Population,Affluence,and Technology)model.Analyze the spatial transfer trend of urban carbon emission intensity and the common characteristics and regional differences of the impact of key factors on carbon emission intensity.The specific research contents and results are as follows :(1)Construction of carbon emission estimation model based on nighttime light data.Firstly,the two types of light data of DMSP-OLS and NPP-VIIRS are corrected respectively,and the overlapping years are selected to perform regression fitting on the light data.The goodness of fit of the equation is 0.9077,and the fitting effect is good.The two types of light data are further fused and corrected,thus constructing the corrected DMSP-OLS scale night light data set from 2000 to 2019.Combined with the provincial-level energy consumption statistics carbon emissions,a carbon emission estimation model based on the corrected nighttime light data set was constructed.The fitting accuracy R2 reached 0.874,and the model estimation effect was good.(2)Spatial autocorrelation and hotspot analysis were used to explore the spatial and temporal dynamic changes of carbon emission intensity.From 2000 to 2019,the carbon emission intensity in the study area generally showed a stable downward trend,showing the characteristics of high in the northwest and low in the southeast,and there was an obvious Hu Huanyong phenomenon;the results of global Moran index and hot spot analysis show that cities with similar carbon emission intensity are in a state of spatial aggregation,showing more significant aggregation characteristics,and the spatial form tends to be stable.The high value area of carbon emission intensity shows the characteristics of double center agglomeration to single center agglomeration.The low value area of carbon emission intensity is characterized by single center aggregation,and there is a convergence phenomenon in time dimension between regions with high carbon emission intensity and regions with low carbon emission intensity.(3)The spatial spillover effect of urban carbon emission intensity is analyzed by using spatial Markov model.Based on the estimated urban carbon emission data from 2000 to 2019,combined with the spatial Markov transfer matrix,the spatial spillover effect of urban carbon emission intensity is analyzed.The results show that the probability matrix of interval transfer of urban carbon emission intensity is in a relatively stable state as a whole,showing a good trend of upward transfer.The transfer spillover phenomenon of carbon emission mostly occurs between adjacent areas,which is a progressive transfer,and there is no sudden increase and decrease.(4)Combined with GTWR model and STIRPAT model,regression analysis was carried out on the influencing factors of carbon emission intensity.Taking urban carbon emission intensity data as the explained variable,population density,per capita GDP,proportion of secondary industry output value,proportion of tertiary industry output value and labor efficiency as the explanatory variables,combined with GTWR model and STIRPAT model,the regression analysis of influencing factors is carried out.The results show that the goodness of fit R2 is 0.8175,and the model fitting accuracy is good.The regression results of 2000,2004,2009,2014 and 2019 are selected to analyze the spatial and temporal heterogeneity of the influence of various influencing factors on urban carbon emission intensity.Among them,per capita GDP has a negative effect on urban carbon emissions,while population density,the proportion of secondary industry output value,the proportion of tertiary industry output value and labor efficiency have both positive and negative effects on urban carbon emissions.The innovations of this paper are summarized as follows :(1)Firstly,the DMSP-OLS and NPP-VIIRS light data are fused and corrected,and the corrected DMSP-OLS long-term nighttime light data set from 2000 to 2019 is constructed.Based on the corrected nighttime light data and provincial carbon emission statistics,a grid-scale carbon emission estimation model is constructed by using panel data model to study the spatial and temporal evolution characteristics of carbon emission intensity.(2)Considering that the carbon emission intensity of municipal units has a certain degree of difference in time and space,this paper selects the GTWR model as the main model of influencing factor analysis,and combines the STIRPAT model to explore and analyze the spatial and temporal heterogeneity of the influencing factors of carbon emission intensity in 282 cities at the national level,so as to provide new reference for energy conservation and emission reduction. |
PDF Full Text Request