Exploring The Effect Of Urban Sprawl On Carbon Emissions Based On The Multi-source Nighttime Light Remote Sensing Data In Southwest China

Since the reform and opening up,China has experienced a rapid urbanization process,and the urbanization rate has increased from 17.92% in 1978 to 60.60% in2019.However,the rapid urbanization process has brought about the rapid expansion of urban space,resulting in the inefficient use of urban land and the dispersion of urban form.Too much emphasis on the growth rate of urbanization and the low-quality development that is eager for success have led to the phenomenon of land expansion faster than population growth in China,that is,urban sprawl,which leads to the shortage of urban resources,urban congestion,haze pollution and other ecological and environmental problems,especially carbon emissions,which has become the focus of policy-making departments and scholars.Therefore,understanding and grasping the relationship between urban sprawl and carbon emissions can provide reasonable and scientific reference for China to better practice the concept of "sustainable development",and develop a sustainable low-carbon urban development model and low-carbon urban planning system to provide a scientific reference.Although the academic community has accumulated some research results in the field of urban sprawl and its effect on carbon emissions,there are still three deficiencies.(1)Few studies have been carried out to measure and analyze urban sprawl from the perspective of large scale and long-term.Also,the studies often employed statistical data such as population/employment distribution to measure urban sprawl,and the conclusions of the effect of urban sprawl on carbon emissions are biased or inconsistent due to the lack of spatial information,slow update speed,and statistical caliber differences.In addition,the single dimension of traditional visible light remote sensing data and the complexity of model method also limit its long-term analysis.Although some studies have used nighttime light data to measure and quantify urban sprawl,rough data correction further limited the accurate measurement and analysis of urban sprawl from the perspective of large scale and long-term.To date,it is unable to grasp the evolution characteristics and laws of urban sprawl in Southwest China as a whole.(2)On the basis of considering heterogeneity,few studies could effectively test and analyze the effect of urban sprawl on carbon emissions under the conditions of large scale and long-term.In addition to the lack of the data,there are also some issues,such as focusing on a single city or a typical area in a particular year,lack of large samples,long time series and heterogeneity analysis.Moreover,what are the transmission factors of carbon emissions from urban sprawl? The existing research is also slightly involved.(3)Few studies have explored the effect of urban sprawl on carbon emissions from the perspective of spatial agglomeration.Spatial agglomeration economy and agglomeration diseconomy are important theoretical basis for understanding the effect of urban sprawl on carbon emissions.From a macro perspective,the effect is the results of urban socio-economic spatial agglomeration or non-agglomeration.Although some studies have revealed that urban sprawl might affect carbon emissions by affecting spatial agglomeration,most of them focused on the micro-scale industrial spatial agglomeration,e.g.,the spatial agglomeration analysis of residential areas.It is rarely carried out the effect from the perspective of spatial agglomeration.In view of the above shortcomings,taking Southwest China as the study object,this study has developed long time series nighttime light dataset based on the corrected multi-source nighttime light data for making up for the deficiencies of traditional statistical data and remote sensing data.Ultimately,this study has provided a new objective data source for the measurement of urban sprawl.Then,urban sprawl index has been constructed by using nighttime light data for analyzing spatial-temporal evolution of urban sprawl in Southwest China.Furthermore,we identified the effect of urban sprawl on carbon emissions from the perspective of large-scale and long-term time series,and analyzed the heterogeneity from different regions,different scales and different periods.It also evaluates the transmission factors of the effect from the perspectives of urban transportation,construction industry,and urban heat island effect.In addition,considering the heterogeneity of time,region,scale,and multicollinearity,this study explored the effect of urban sprawl on carbon emissions from the perspective of spatial agglomeration by developing the spatial Gini coefficient.The main conclusions are as follows:(1)The urban sprawl index quantified by nighttime light data could effectively represent the level of urban sprawl in Southwest China.For calculating urban sprawl index,the nighttime light data were used to replace statistical data,urban light saturation threshold areas were used to replace urban built-up area,and light gray value could represent population activity density.In addition,a panel regression analysis was developed between the single urban sprawl index based on statistical data and the urban sprawl index calculated by nighttime light data.The results show that all models are positive,and four models are significant above 1% level,and one model is significant at10% level,which indicates that the regression results are reasonable,and proves that there is a stable correlation between the urban sprawl index of nighttime light data and the urban sprawl index of statistical data.(2)Urban sprawl and carbon emissions in Southwest China present significant temporal and spatial heterogeneity.From the perspective of temporal and spatial changes,the urban sprawl in Southwest China is relatively serious and in a dynamic process.The degree of urban sprawl in Sichuan,Chongqing,Guizhou and Yunnan gradually increased.The urban sprawl index of Sichuan province was 0.452-0.529,that of Chongqing was 0.443-0.501,that of Guizhou was 0.440-0.513,and that of Yunnan was 0.446-0.524.The severity of urban sprawl within different city size was in order of medium-sized city,big city and small city.Thus,the cities in Southwest China should need to prevent the adverse effects of urban sprawl.It can be seen from the time trend that carbon emissions in Southwest China increased with the increase of time.The annual carbon emissions of cities in Southwest China increased from about 100 tons in 2000 to about 240 tons in 2018,an increase of about 2.4 times.Carbon emissions of Chongqing and Sichuan province are at a high level in Southwest China.From the perspective of time heterogeneity.carbon emissions in Southwest China was in a linear growth trend before 2009,and then it tended to be stable.From the perspective of spatial distribution,the areas with high carbon emissions mainly distributed in Chongqing metropolitan area,Chengdu,Guiyang,and other core cities in 2000.By 2018,carbon emissions of the cities within Sichuan Chongqing,Guiyang,Kunming,and some surrounding cities increased significantly compared with that of 2000.On the contrary,due to the constraints of regional society,history,nature,and other factors,carbon emissions in Southwest Yunnan,Southeast Guizhou,and Sichuan Chongqing autonomous prefecture presented a low trend.(3)Urban sprawl in Southwest China could aggravate carbon emissions and present heterogeneities of time,region,and city size.Urban transportation,construction and heat island effect are important transmission factors affecting carbon emissions.From the full sample analysis,it is concluded that urban sprawl would promote the increase of carbon emissions.By replacing the core explanatory variables,eliminating outliers,and two-stage regression of instrumental variables,the stability and reliability of the results were also verified.From the analysis of time heterogeneity,it can be concluded that urban sprawl would have a positive effect on carbon emissions in the period before and after.The analysis of regional heterogeneity shows that urban sprawl in different regions could aggravate carbon emissions.From the analysis of city size heterogeneity,it can be seen that urban sprawl within small-sized cities in Southwest China had a significant positive correlation with carbon emissions.Although the regression results of medium and large-sized cities were not significant,the coefficient was still positive,which proved the possibility of urban sprawl increasing carbon emissions to some extent.Furthermore,it is found that urban sprawl could affect carbon emissions mainly through the urban transportation,construction industry,and heat island effect.From the path of affecting urban traffic,we can find that urban sprawl extended the travel distance between house and workplace,leading to the increase of car ownership and dependence on private car travel,and urban public transport facilities lag behind the speed of urban construction,thus affecting carbon emissions.From the perspective of the path affecting the construction industry,urban sprawl increased the demand for buildings and public infrastructure,and then affected the consumption of construction operations and energy consumption,and then affecting carbon emissions.From the path of affecting the heat island effect,urban sprawl would cause the increase of urban temperature,aggravate the heat island effect,and then increase the energy consumption caused by residents' cooling,thus increasing carbon emissions.(4)Urban spatial agglomeration in Southwest China reduces carbon emissions,while spatial agglomeration weakens the effect of urban sprawl in increasing carbon emissions.Spatial agglomeration is conducive to the improvement of population density and thus to the reduction of the overall living cost of residents,resulting in the reduction of traffic and household carbon emissions.At the same time,increasing population concentration will improve the utilization rate of public transportation and other infrastructure,thus reducing energy consumption.Moreover,improving the density of economic activity,spatial agglomeration economy agglomeration of industrial technology,resources,information,and so on,the spillover of spatial agglomeration can attract labor capital,and advanced industrial technology innovation elements and public foundation can be shared among subunits of the city to improve the efficiency of energy utilization.Also,economic agglomeration technology spillover effect is beneficial to the innovation and application of green cleaning technology such as energy conservation and emissions reduction and reduces the energy consumption of the unit of output and reduce emissions and to achieve the goal of low carbon emissions.Through empirical analysis,it is found that there is a negative linear relationship between urban sprawl and spatial agglomeration.Growing urban sprawl reduces the degree of spatial agglomeration,and the reduction of spatial agglomeration level also will make the agglomeration of the negative effect to reduce carbon emissions,thus forming urban ssprawl to reduce spatial agglomeration level and intensify the transmission path of carbon emissions is based on the empirical spatial agglomeration can weaken the urban spread of carbon emissions,but caused by urban spread tightening elements concentration levels drop,the spatial agglomeration effect also gradually decline.According to the overall sample analysis,although urban spatial agglomeration can effectively inhibit the increase of carbon emissions in Southwest China,the inhibiting effect of spatial agglomeration on carbon emissions is gradually weakened with the intensification of urban sprawl.By eliminating the central cities and deleting or changing variables,the reliability of the baseline regression model results is further verified,and the overall intensification effect of urban sprawl on carbon emissions and the mitigation effect of spatial agglomeration on carbon emissions in Southwest China is more robust proved.The impacts of urban sprawl and spatial agglomeration on carbon emissions are heterogeneous in time,region,and city size.From the analysis of temporal heterogeneity,it can be concluded that the negative and positive impacts of spatial agglomeration and urban sprawl on carbon emissions are still established in the early and late stages.Before and after carbon emission,spatial agglomeration weakens the effect of urban sprawl on carbon emission.In different regions,these mechanisms exist in the samples of Guizhou and Sichuan-Chongqing regions,but not in the samples of Yunnan regions.The above-mentioned spatial agglomeration effect exists in cities of different sizes,small and medium sizes,but does not exist in the sample of large cities.This is because the scale effect brought about by the large-scale agglomeration of urban population and economic activities continuously attracts the inflow of population and capital factors,which may cause a crowding effect and form scale diseconomies,and promote carbon emissions.The innovations of this study are mainly reflected in the following three aspects.(1)The urban sprawl index was constructed to accurately measure the urban sprawl in Southwest China based on the correction of multi-source night light data,and the spatiotemporal evolution analysis of large sample and long period is carried out,which makes up for the limitation of the lack of spatial information in statistical data and the complexity of conventional remote sensing data to a certain extent.(2)Breaking through the limitations of the previous single dimension,this study has constructed an analysis framework of "the effect of urban sprawl on carbon emissions".On the one hand,it not only considers the overall impact of urban sprawl on carbon emissions,but also involves the heterogeneity analysis of the impact.On the other hand,it focuses on the prefecture level city scale,which makes up for the limitations of the previous analysis focusing on a single city or regional scale.(3)From the perspective of spatial agglomeration,this study quantitatively analyzed the effect of urban sprawl on carbon emissions.To some extent,urban sprawl might reduce spatial agglomeration and then affect carbon emissions.However,most of the existing studies have focused on the environmental impact of industrial spatial agglomeration at the micro scale.On the one hand,they have ignored the analysis of the effect of urban sprawl and spatial agglomeration on carbon emissions at the macro scale,and on the other hand,they have neglected the heterogeneity analysis.