Next Article in Journal
Differential Stomatal Responses to Surface Permeability by Sympatric Urban Tree Species Advance Novel Mitigation Strategy for Urban Heat Islands
Previous Article in Journal
Green Distribution Route Optimization of Medical Relief Supplies Based on Improved NSGA-II Algorithm under Dual-Uncertainty
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 15
Issue 15
10.3390/su151511940
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

The Impact of Urban Sprawl on Carbon Emissions from the Perspective of Nighttime Light Remote Sensing: A Case Study in Eastern China

by
Ling Zhang
1,
Jiawei Zhang
2,
Xiaohui Li
2,
Kaidi Zhou
2 and
Jiang Ye
2,*
1
College of Management Science, Chengdu University of Technology, Chengdu 610059, China
2
College of Earth Sciences, Chengdu University of Technology, Chengdu 610059, China
*
Author to whom correspondence should be addressed.
Sustainability 2023, 15(15), 11940; https://0-doi-org.brum.beds.ac.uk/10.3390/su151511940
Submission received: 24 June 2023 / Revised: 23 July 2023 / Accepted: 31 July 2023 / Published: 3 August 2023
Browse Figures
Review Reports Versions Notes

Abstract

:
This paper collects the nighttime light data and six control variables of 77 cities in the eastern China region from 2000 to 2020 and analyzes the spatio-temporal changes of urban sprawl and carbon emissions in the eastern China region. Using the baseline regression model, the correlation and influencing factors between urban sprawl and carbon emissions are explored. The results show that although the impact of urban sprawl on carbon emissions in the eastern China region is positive, it is the result of a trade-off of various factors. Different factors have different impacts on carbon emissions, and urban expansion of different city sizes also has different impacts on carbon emissions in the eastern China region. This paper uses nighttime light data to measure the urban sprawl index more finely, directly explores its impact on carbon emissions from the perspective of urban sprawl and conducts an in-depth analysis of multiple external factors and different city types, providing references for decision-makers to construct sustainable low-carbon city development models and low-carbon city planning systems from different dimensions.

1. Introduction

1.1. Research Background

Since the industrial revolution, with the progress of science and technology and economic recovery, the process of urbanization in the world has achieved unprecedented rapid development. With the suburbanization of the urban population, the reverse development of cities and many other ‘pie-type’ development phenomena, the phenomenon of urban sprawl has become increasingly prominent. For China, it only took ~20 years to complete the urbanization process, while Western countries needed a hundred years to complete it. However, the rapid advancement of urbanization also leads to the drastic expansion of urban space, occupying a large amount of land around the city, which will result in the phenomenon of disorderly urban expansion, and this will inevitably bring some negative effects, such as environmental pollution [1] and soil damage, which will lead to a series of social and ecological problems such as a shortage of urban resources and imperfect public service facilities [2,3]. Therefore, urban sprawl is a disorderly, low-density rapid expansion phenomenon in which the expansion rate of urban areas exceeds the demand of the urban population [4,5,6]. Urban sprawl, which is different from the concentration and agglomeration of population and urban structure in urbanization, will cause a series of social problems. For example, urban land use efficiency is low [7], the urban form is scattered and urban functional zoning is unreasonable [8], resulting in traffic congestion [9], environmental pollution, cultivated land erosion, urban ecological environment destruction and other phenomena, preventing urban development and further economic progress [2,3,4,10].
On the other hand, the phenomenon of global warming has been the main culprit in threatening the earth’s ecology, and numerous studies have pointed out that the increase in greenhouse gases has contributed to global warming [11,12,13]. The World Climate Organization (WCO), in the WCO Greenhouse Gas Bulletin, states that 43% of the warming effect on the global climate is contributed by greenhouse gases, 82% of which are caused by carbon dioxide [14], and that most of the carbon dioxide comes from emissions from human economic production and life [15]. The United Nations World Commission on Environment and Development proposed the concept of “sustainable development” in 1987, and there has been a growing awareness of ecological issues around the world; as the world’s largest emitter of carbon emissions [16], the Chinese government has been working to reconcile urban development with carbon emissions. China has pledged to peak its carbon emissions by 2030 and become carbon neutral by 2060. East China, one of China’s economically developed regions (including Shanghai, Jiangsu, Zhejiang, Anhui, Fujian, Jiangxi, Shandong and Taiwan provinces), has also become a major energy-consuming and carbon-emitting region in China.
Therefore, from the perspective of urban sprawl, studying the impact of regional urban sprawl on carbon emissions is of profound significance for practicing the green and sustainable development of cities and provides a scientific reference for countries to build sustainable urban development models and low-carbon urban planning systems.

1.2. Research Review

The term ‘urban sprawl’ became popular after it was first proposed in the 1960s, mainly describing the rapid expansion and extension of cities and their surrounding areas [17]. Urban sprawl is a product of the process of urbanization. It is mainly manifested in the low density and scattered use of land resources [18,19] and the lack of planning guidance [20,21], which has a negative impact on social development [22,23]. Current research has different definitions of urban sprawl from different perspectives [24,25]. Green environmentalists worry that urban sprawl will damage the environment and human health [1,26]; economists believe that the spread of the city may be related to market failure, the increasing cost of infrastructure construction and other factors [24,26]. Sociologists pay more attention to the impact of urban sprawl on society [16,27]; urban planners argue that urban sprawl leads to the loss of vitality and individuality of traditional communities [15,24,25,27]. Some studies separate the causes and consequences of urban sprawl from the phenomenon of urban sprawl itself to clarify the definition of “urban sprawl”. They argue that urban sprawl has different causes and outcomes in different regions and regulatory contexts [28]. In terms of measuring the driving factors of urban sprawl, scholars’ research also shows different dimensions. For example, they measured and explored the driving factors of urban sprawl in a certain region from the population, economic and social dimensions [29,30,31,32]; from the ecological dimension, Glaeser et al. innovatively constructed as many as nine indicators to measure Romania’s urban sprawl from the perspective of green space through the analytic hierarchy process [33]. Deng et al. combined population and economy with ecological dimension to measure urban sprawl in China [27]. From the policy, political and cultural dimensions, they discussed the dynamics and driving factors of urban sprawl [34,35]. Studies generally agree that pursuing rapid urbanization excessively will lead to the phenomenon of urban sprawl, which will inevitably bring some negative impacts and a series of social and ecological problems [2,36,37,38].
In addition, studies have different measurement methods for urban sprawl. The index method measures the urban sprawl situation through the statistical results of indicators [39]. Specifically, it is divided into single-dimensional [40,41,42] and multi-dimensional indicator systems [33,34,43]. However, when the indicator system is large, it causes difficulties in data acquisition and processing. The model method is often used in the fields of urban planning and geography. The most widely used model is the cellular automaton model (CA) [44,45]. However, it does not consider the driving factors of urban sprawl, such as population growth, distance, etc. Traditional remote sensing data have the characteristics of objectivity and spatiality, which can enable us to use multiple scales and long-term observations from a unified standard to simulate socio-economic development and natural environmental changes [1,2,46], but there are problems with the clarity and acquisition of remote sensing images. Therefore, this paper uses the long-term night light dataset to construct the urban sprawl index more accurately and solves the problems of lack of spatial information, difficulty in obtaining information, and complicated processing of the traditional data for long-term, large-scale research, and measures and analyzes the urban sprawl situation in China, to further enrich the quantitative theory of urban sprawl.
Scholars discussed the impact of various phenomena caused by urban sprawl on urban carbon emissions [47,48,49,50]. Burchfield et al. concluded that there is a positive correlation between urban sprawl and the popularity of private cars [51]. The rapid development of the transportation industry has directly or indirectly led to an increase in carbon emissions [52,53,54,55,56]. At the same time, urban form and land use also have an impact on carbon emissions [57,58,59,60,61,62], and the influence of policy factors on carbon emissions cannot be ignored [63,64]. Some studies also believe that the density of urban sprawl will have an impact on carbon emissions [65,66,67,68]. Although more and more studies focus on the impact of urban expansion on carbon emissions, this impact is more often manifested as indirect and single-factor analysis [68,69,70]. At present, many studies tend to focus on the impact studies of large cities or some ecologically developed regions [71]. For example, Carpio, A et al. analyzed the urban expansion of the Monterrey Metropolitan Area (MMA) Mexico from 1990 to 2019 using satellite imagery and Geographic Information Systems (GIS) to determine its relation to carbon emissions [72]. Lee, C measured urban form as an indicator of metropolitan sprawl and explored its impact on commuting trips and NOx and CO2 emissions from road traffic in all metropolitan statistical areas (MSAs) and four groups’ MSAs separated by population in the continental United States [9]. Li et al. conducted a study on 34 large and medium-sized cities in China and concluded that urban sprawl will further aggravate environmental pollution, but its low-density nature can improve energy efficiency [24]. Bereitschaft et al. explore the relationships between urban form and air pollution among 86 major U.S. metropolitan areas [14]. Later, Bereitschaft also proposed that regional and temporal differences need to be considered to more accurately study the impact of urban sprawl on carbon emissions [14].

1.3. Research Ideas

In general, most studies agree that there is a close relationship between carbon emissions and urban expansion. However, few studies directly explore the impact of urban sprawl on carbon emissions, most of them indirectly analyze the impact of urban sprawl on carbon emissions from individual factors, such as urban form, population, economy, ecology, etc., and there is also less analysis of the impact of external factors other than urban sprawl on carbon emissions. There is a lack of in-depth exploration of the relationship between the two, such as what are the effects of considering and not considering external factors on their relationship? What are the differences? What is the magnitude of the effect of each external factor on this relationship? Is there heterogeneity? This part of the in-depth exploration is more realistic for decision-makers. At the same time, in terms of measuring the urban sprawl index, most studies use traditional data, which has problems such as large differences in data statistics from different departments, difficulty in obtaining and processing data, etc. With the demand for long-term and large-scale data, especially for small and medium-sized cities, the authenticity and accuracy of the data will be affected, which will also affect the measurement results of urban sprawl. Therefore, this paper will collect nighttime light data from 2000 to 2020, which has characteristics such as data uniformity, easy access, high authenticity and accuracy. It will take 77 cities with different types in East China as research objects, with urban sprawl as the main explanatory variable, carbon emissions as the dependent variable and six representative control variables as external factors to construct a baseline regression model. It hopes to explore the issue of urban sprawl’s impact on carbon emissions more deeply.

2. Study Area

East China is an important administrative and geographical sub-region in eastern China, including Shanghai, Jiangsu, Zhejiang, Anhui, Fujian, Jiangxi, Shandong and Taiwan provinces, with a total area of 83.43 km2, accounting for 8.7% of China’s total area. Due to the lack of data from Taiwan, this study only covers the other six provinces and one municipality in the research area (see Figure 1). Located on China’s third terrace, the main topography of Eastern China includes plains, basins, mountains and hills, and is one of the economically developed regions in China with relatively complete topographic units.
East China is one of the economically developed regions in China, with a relatively fast economic development speed, and it has a very important position in China in terms of industry, agriculture and tertiary industry. East China can be said to be a microcosm of the whole of China; it not only includes plain cities, river-open cities, mountain cities, resource-based cities and tourism service cities, but also includes cities with different population sizes and economic scales and different development speeds. Therefore, the study of the impact of urban sprawl on carbon emissions in 77 cities in East China has certain representativeness.

3. Materials and Methods

3.1. Data Selection and Source

3.1.1. Nighttime Light Data

Two types of nighttime light data were selected for this study: the 2000–2013 DMSP-OLS nighttime light data and the 2013–2020 NPP-VIIRS nighttime light data. Both DMSP-OLS and NPP-DNB data were processed and distributed by the National Oceanic and Atmospheric Administration (NOAA) and are freely available. Details of the two types of data are shown in Table 1.
Due to the benchmark difference between the two nighttime light datasets, it was necessary to calibrate the data. The specific procedure is as follows: First, the DMSP-OLS data from 2000 to 2013 are pre-processed to ensure that each year’s data have the same coordinate system, and then the multi-sensor calibration was performed using the region in East China where the night light data remained stable for a long period of time as the benchmark to complete the intra-annual fusion and inter-annual continuity correction of the multi-sensor night light data. Second, the monthly NPP-VIIRS data from 2013 to 2020 were merged into annual data, and background denoising, negative value removal and continuity correction were performed. Finally, the DMSP-OLS data and NPP-VIIRS data of the same year (2013) were selected to establish regression equations, and the NPP-VIIRS data from 2014 to 2020 were corrected. This results in a benchmarked and harmonized long time series luminescence data set from 2000 to 2020, part of which is shown in Figure 2.

3.1.2. Carbon Emission Data

The study uses the carbon emission data provided by the United Nations Intergovernmental Panel on Climate Change (IPCC) from 2000 to 2020 in East China to analyze the spatial and temporal changes in carbon emissions in East China.

3.1.3. Economic and Social Data

The study used the per capita GDP, population size, total fixed asset investment, the proportion of the first, second and third industries and the green coverage rate of built-up areas in the provinces and cities in the ‘China Urban Statistical Yearbook’ to explore the impact of different influencing factors on urban sprawl and carbon emissions.

3.1.4. Other Auxiliary Data

The vector data of China’s national administrative regions were downloaded from the China Basic Geographic Information Center. At the same time, the mask extraction function in ArcGIS was used to obtain vector layers of different scales such as provincial, prefecture-level and county-level in Shanghai, Jiangsu, Zhejiang, Anhui, Fujian, Jiangxi and Shandong.

3.2. Variable Selection and Data Description

3.2.1. Core Explanatory Variable: Urban Sprawl Index (US)

The study combined the existing urban sprawl measurement methods to measure and analyze the cities in East China. Some studies used the ratio of urban built-up area growth rate to urban population growth rate to quantify urban sprawl [73], but this index is difficult to apply to cities with negative urban area or urban population growth, and the core of the index is still urban population density, which cannot fundamentally overcome the inherent defects of the average density index. The significant characteristics of urban sprawl are low density and decentralization. The average population density is often difficult to distinguish whether a certain number of people are evenly distributed or concentrated in a certain area, which can only be used as a rough measure of urban sprawl. Because the lights in rural areas are very weak compared to the lights inside the city, the greater the light radiation value in the nighttime light data, the higher the local human activities and the degree of urban development tend to be; therefore, the light area of nighttime light data can effectively reflect the spatial area of urban built-up areas, the scope of social and economic activities and the size of light radiation value can reflect the intensity of social and economic activities [6].
To precisely explore the phenomenon and characteristics of urban sprawl in 77 prefecture-level cities in six provinces and one municipality directly under the central government in East China, the study combines the existing urban sprawl measurement methods [6,74] and uses the thresholds of DN values obtained from the long time-series of nighttime light data in East China for the calendar years 2000–2020 to define the target areas for measuring urbanization area and urban sprawl index (see Table 2) and analyzes them.
The calculation formula for the analysis is as follows:
U S i = S A i × S P i
where U S i is the sprawl index of city i and the value ranges from 0 to 1. The larger the value, the higher the degree of urban sprawl; S A i   is the horizontal sprawl index (see Equation (2)), mainly reflecting the area dimension of lighting areas in urban spaces; L A i and H A i are the proportion of the area of light in the built-up area of a city that is below or above the average light intensity of the province in which the city is located, respectively.
S A i = 0.5 × ( L A i − H A i ) + 0.5
S P i is the vertical sprawl index (see Equation (3)), mainly reflecting the value dimension of lighting in an urban space. In contrast to previous methods of measuring urban sprawl using population density, the urban sprawl index constructed in the study does not consider the population of the city but rather the area and intensity of light within the city. This index provides a more comprehensive picture of urban sprawl in both horizontal and vertical terms.   L i and H i are the proportion of the total light values in the built-up area of a city that are below or above the average light intensity of the province in which the city is located, respectively, to the total light values of the city.
S P i = 0.5 × ( L i − H i ) + 0.5
Both S A i and S P i range from 0 to 1, with higher values indicating a higher level of horizontal or vertical urban sprawl.

3.2.2. Selection of Control Variables

To explore the impact of different influencing factors on the relationship between urban sprawl and carbon emissions, Fang et al. used government intervention, industrial structure, energy structure and economic development level control variables to study the impact of urban sprawl on carbon emissions in the Yangtze River Economic Belt [74]. Liu analyzed the impact of urban sprawl on urban carbon emissions based on the spatial Durbin model by integrating six control variables: population density, economic development, industrial structure, foreign direct investment, traffic characteristics and energy consumption [75]. This paper also selects the factors that studies generally involve when studying urban sprawl and carbon emissions and combines the characteristics of huge differences in population density, economic development level and urban greening level among cities in East China. At the same time, considering the availability and validity of the data, six control variables covering population, economic development, industrial structure and ecological environment are finally selected (see Table 3).
In statistical analysis, to avoid the collinearity problems caused by too strong of a relationship between dependent variables, a variance inflation factor test (VIF) is needed. When there is no serious multicollinearity between variables, the value of VIF should usually be less than 10. The mean value of the variance inflation factor (VIF) was 2.839, and the VIF of each variable index is not greater than 10, so there is no multicollinearity between the variables (see Table 4).

3.2.3. Benchmark Regression Model Construction

(1)
Theoretical analysis of the impact of urban sprawl on carbon emissions
Firstly, urban sprawl is characterized by the outward expansion of urban boundaries, the relatively scattered spatial distribution of population and the increase in the frequency of use of transportation, resulting in increased carbon emissions. However, urban sprawl may also cause some positive factors. For example, urban sprawl is conducive to alleviating population concentration in central urban areas and facilitating traffic diversion, thereby reducing carbon emissions.
Secondly, urban sprawl will lead to an increase in urban land use to maintain the scope of resident living and economic activities, thereby increasing the laying and use of infrastructure such as domestic water and electricity, and thus increasing carbon emissions. On the contrary, urban sprawl may lead to lower population density and scattered distribution of residents’ activities and industries, which may reduce the impact of people on the environment and thus reduce carbon emissions.
Thirdly, the destruction of the ecosystem and natural environment by urban sprawl may have a negative impact, weakening the ecological regulation function of urban areas, exacerbating the heat island effect, reducing the quality of the natural environment of the city and further increasing carbon emissions. On the contrary, urban decentralization is conducive to air circulation and reducing urban temperature, heat island effect and energy consumption, thereby also reducing carbon emissions.
Thus, the impact of urban sprawl on carbon emissions may arise through a combination of trade-offs among the above factors.
(2)
Benchmark regression model construction
Based on the above discussion on the impact of urban expansion on carbon emissions and the models used by other researchers, this paper develops the following baseline regression model (see Equation (4)) to examine the causal link between urban sprawl and carbon emissions:
C E i t = C + β 1 U S i t + β 2 X i t + γ + ε i t
In Equation (4), C E i t   is the carbon emission of city i in year t; the core explanatory variable U S i t represents the spread of city i in year t, and its coefficient β 1 represents the impact of urban sprawl on carbon emissions. If positive, it indicates that urban sprawl will increase urban carbon emissions, and vice versa [76]. X i t is a set of control variables related to the city, and the specific variable parameters are shown in Table 2. γ represents fixed effects, including urban fixed effects and year fixed effects, which are used to reduce the bias of missing variables; ε i t is the error term.

4. Results

4.1. Urban Sprawl Index Measurement

Using Equations (1)–(3), the urban sprawl index of 77 cities from 2000 to 2020 was calculated. Using the natural breakpoint grading method, the urban sprawl in East China was divided into four different levels. By comparing the spatial distribution of urban sprawl in East China in 2000, 2008, 2012 and 2020 (see Figure 3), most cities in East China were found to have medium or high levels of sprawl from 2000 to 2020, and the level of sprawl increased over time. This is reflected in the main cities in the Yangtze River Delta region gradually extending to the surrounding areas, forming a north–south urban belt. Along this urban belt, the connection and urbanization level between cities was increasing, forming some new urban hubs, which affect the scale of urban sprawl to varying degrees.

4.2. Empirical Analysis of Carbon Emissions

Based on Equation (4), the carbon emissions of the provinces in East China in the past 20 years were calculated. Figure 4 shows the time change in the total annual carbon emissions of the provinces in East China.
The total carbon emissions of Shandong Province were ranked first over the years and far ahead of other provinces and cities in East China. It increased from about 195 Mt in 2000 to about 940 Mt in 2020, an increase of about 4.8 times. It was at a high level in the carbon emissions of the whole of East China and is the focus of environmental protection in East China in the future. The second is Jiangsu, a major industrial province; the total carbon emissions of Zhejiang, Anhui and Shanghai are in the third, fourth and fifth positions, respectively. Although the carbon emission intensity per unit area of Shanghai is large, the carbon emission base is small due to the small area, so the total amount is not large. At the same time, the last two are the Fujian Province, which has the highest forest coverage rate in China, and Jiangxi Province, which ranks second, reflecting that the ecological environments of the Fujian and Jiangxi Provinces are in good condition.

4.3. Analysis of the Influencing Factors of Urban Sprawl on Carbon Emissions

4.3.1. Regression Model Selection

The previous section used the degree of urban sprawl as the core explanatory variable for calculating the total carbon emissions and only analyzed the temporal changes of the total carbon emissions in East China. However, to accurately explore the relationship between urban sprawl and carbon emissions, further research is needed. Based on the specific requirements of control variables, the study establishes two types of panel data models: ordinary least squares (OLS) and fixed effects (FE). The main basis for selecting the above two types of panel data models are: (1) Controlling individual heterogeneity: there may be fixed characteristics and heterogeneity among individuals in the panel data. By introducing a mixed-panel regression model and a panel fixed-effect regression model, it can control individual fixed effects, eliminate the impact of individual heterogeneity and pay more attention to the impact of urban sprawl on carbon emissions. (2) Considering the characteristics of time series: the panel data contain observations at multiple time points, which can capture the long-term trends and changes of urban sprawl on carbon emissions. The mixed panel regression model and the panel fixed effect regression model consider the variability of the time dimension, which can analyze the impact of urban sprawl on carbon emissions more comprehensively and analyze the time series. (3) Provide more accurate estimation and inference: The mixed panel regression model and the panel fixed effect regression model consider the random effects part while controlling the individual fixed effects, providing more accurate estimation and inference results. Through these models, it can better understand the impact of urban sprawl on carbon emissions and draw more reliable conclusions.

4.3.2. Full Sample Regression Analysis

Table 5 shows the full sample regression results to explore the impact of urban sprawl on carbon emissions in East China. Column (1) in Table 5 is the result of a mixed panel OLS regression model without introducing control variables and fixed effect analysis, and Column (3) is the result of a panel fixed-effect FE regression model without introducing control variables. The core explanatory variable, urban sprawl, coefficient was positive but not significant in both models. Columns (2) and (4) are the results of introducing control variables and fixed-effect experiments under the mixed-panel OLS regression model and the panel fixed-effect FE regression model, respectively. The core explanatory variable, urban sprawl, has a significant positive coefficient at the 5% level, which means that urban sprawl has a positive impact on carbon emissions in East China, and this impact is more significant under the effect of control variables. The impact of urban sprawl on carbon emissions is the trade-off result of multiple factors.
In addition, according to the full sample regression estimation of the East China region in Table 5, for each control variable, it is not difficult to see: (1) The coefficient of population size was negative but not significant, that is, population size was negatively correlated with urban carbon emissions, and the relative decentralization of the population is one of the important external manifestations of urban sprawl. (2) The coefficient of urban economic development level was positive, and it was significant at the 1% level, indicating that the level of urban economic development and carbon emissions show a strong positive correlation. (3) The coefficients of industrial structure and tertiary industry were both positive, but industrial structure was significant at the 1% level, indicating that the impact of industrial structure on carbon emissions is far greater than that of tertiary industry on carbon emissions. (4) The positive effect of fixed asset investment was slightly stronger than the negative effect, indicating that the increase in public fixed asset investment has a certain promoting impact on carbon emissions. (5) The urban greening coefficient was negative, and when all control variables and fixed effects are combined, the regression results in column (4) indicate that if the urban greening index increases by 1%, carbon emissions will decrease by 0.998%. This shows that improving urban greening can play a role in energy conservation and emission reduction.

4.4. Heterogeneity Analysis

Some studies have found that urban scale has heterogeneity in the urban sprawl index [77], so will different urban scales cause urban sprawl to have heterogeneous effects on carbon emissions? Therefore, based on the household registration population in the municipal districts of East China in 2020, the study divides the cities with a household registration population of less than 4.5 million into small-scale cities, the cities with a household registration population of between 4.5 million and 7 million into medium-sized cities, and the cities with a household registration population of more than 7 million into large-scale cities, to analyze the heterogeneity of urban scale (see Table 6).
The results show that city size has a significant heterogeneous effect on carbon emissions.
(1)
The results of large-scale urban regression show that the core explanatory variable coefficients in both models are positive, indicating that urban sprawl in the large-scale urban sample of East China has a positive impact on carbon emissions. When all control variables and fixed effects are combined, the regression results in Column (2) show that if the urban sprawl index increases by 1%, carbon emissions increase by 0.284% accordingly. For large-scale cities, population size can reduce carbon emissions, and the urban economic level significantly promotes carbon emissions. In addition, the coefficients of industrial structure, tertiary industry and fixed investment did not pass the significance test, and most of the experimental coefficients were negative, indicating that the three have a greater inhibitory effect than a promoting effect on carbon emissions. Furthermore, the improvement of urban greening has a significant impact on inhibiting carbon emissions.
(2)
The regression results of medium-sized cities in East China show that although none of the core explanatory variables passed the significance test, the coefficients of the core explanatory variable, urban sprawl, are positive, indicating that urban sprawl has a positive impact on the carbon emissions of medium-sized cities in East China. When all control variables and fixed effects were combined, the regression results in Column (4) showed that if the urban sprawl index increased by 1%, carbon emissions increased by 0.681% accordingly. For medium-sized cities, the impact of population size and urban economic level on carbon emissions was consistent with the results of large-scale cities. However, the impact of industrial structure on carbon emissions in medium-sized cities in East China was more significant than that in large-scale cities, with a significance level of 1%. Although the coefficients of tertiary industry and fixed investment in medium-sized cities did not pass the significance test, they were positive, indicating that both have a promoting effect on carbon emissions. Meanwhile, the impact of urban greening on carbon emissions is not significant, which is completely different from the impact results of large-scale cities in East China.
(3)
The regression results of small-scale cities in East China show that the coefficients of the core explanatory variables in both models are positive, indicating that urban sprawl in the small-scale city sample of East China has a positive impact on carbon emissions and passed the 5% significance test. When all control variables and fixed effects were combined, the regression results in Column (6) showed that if the urban sprawl index increased by 1%, carbon emissions increased by 1.351% accordingly. For small-scale cities, population size did not pass the significance test, but the coefficient was positive, indicating that it has a promoting effect on carbon emissions, which is opposite to large and medium-sized cities. Urban economic level and industrial structure passed the 1% significance test, indicating that they have a significant promoting effect on carbon emissions. The impact of tertiary industry and fixed assets on carbon emissions is the same as that of medium-sized cities; that is, both promote carbon emissions. At the same time, improving urban greening levels for small-scale cities had a significant promoting effect on energy saving and emission reduction.
Overall, urban sprawl had a positive impact on carbon emissions for cities of various sizes, but the degree of impact varied. The impact of urban sprawl on carbon emissions in small-scale cities in East China (1.351) was greater than that of urban sprawl on carbon emissions in medium-sized cities (0.681); the impact of urban sprawl on carbon emissions in medium-sized cities was greater than that of urban sprawl on carbon emissions in large-scale cities (0.284). At the same time, the heterogeneity test results were basically consistent with the full sample regression test results, which further confirms that urban sprawl has a positive impact on carbon emissions to a certain extent.

5. Discussion

The impact of carbon emissions and urban sprawl has been affirmed by many studies, but they have not been able to provide systematic and clear evidence to directly explain how urban sprawl affects carbon emissions. Therefore, this study extracts long-term night light data, which has the advantages of broad coverage, fast timeliness and convenient access compared with traditional data and can be widely applied to multiscale long-term urban problem research. At the same time, based on previous studies on urban sprawl and carbon emission-related issues, six representative control variables in terms of population, industry, economy and ecology were selected to construct a baseline regression model. Taking 77 cities in East China from 2000 to 2002 as the analysis sample, the impact of urban sprawl on carbon emissions was explored. The results show that in the absence of control variables, urban sprawl is positively correlated with carbon emissions, which is consistent with the result of Lu Jing that the increase of urban sprawl degree will increase carbon dioxide emissions [42], but it was not significant. However, after introducing control variables, we found that the coefficient of the core explanatory variable was significant at the 5% level, which proves that the significant impact of urban sprawl on carbon emissions is the result of trade-offs among the various factors.
This study also explored the heterogeneity of the impact of urban sprawl on carbon emissions from the perspective of city size. The results show that urban sprawl has the greatest impact on carbon emissions in small-scale cities in East China, followed by medium-sized cities and finally, large-scale cities. This is different from the research results of Li [6], who found that urban sprawl has the greatest impact on carbon emissions in small-scale cities, but there is little difference between medium-sized and large-scale cities. The reasons for this result are that it may be caused by different city characteristics, criteria for dividing city size and data sources. This study shows that urban sprawl has a more obvious difference in the impact on carbon emissions in large, medium and small-scale cities.
From the perspective of the impact of various control variables on carbon emissions:
(1)
Population size has no significant impact on carbon emissions, and it can play a certain inhibitory role for large and medium-sized cities because when the urban population is not concentrated and too dispersed, it increases the energy consumption and exhaust emissions caused by transportation. At the same time, energy consumption in urban construction, heating, lighting and other aspects will also increase accordingly, thereby increasing carbon emissions. However, for small-scale cities, the increase in population size will promote carbon emissions in small-scale cities to a certain extent because the increase in small-scale city size first manifests as an increase in population density, which in turn increases production and living energy consumption, and thus increases carbon emissions.
(2)
No matter the size of the city, economic development level has a positive and significant impact on carbon emissions, showing a positive promoting effect. Because urban economic prosperity and development require a lot of energy consumption, especially in fields such as industry, construction and transportation, energy consumption inevitably accompanies a lot of carbon emissions [57]. In addition, the urban economic development level will also affect the living standards and consumption habits of urban residents. Higher living standards and more consumption demand will also cause more energy consumption and carbon emissions.
(3)
Industrial structure has a positive promoting effect on carbon emissions for medium and small-sized cities but plays a certain inhibitory role for large-scale cities because industrial industries in large-scale cities are often accompanied by more advanced carbon emission treatment technologies, higher-level government control, etc.
(4)
Tertiary industry and fixed-asset investments can play a certain inhibitory role for large-scale cities, but they have a certain promoting effect on carbon emissions for medium and small-sized cities. For medium and small-sized cities, fixed-asset investments often accompany infrastructure construction, increasing consumption in lighting, heating and other aspects. They also affect the energy structure of cities and regions, thereby increasing carbon emissions. However, fixed asset investment in large-scale cities may introduce more energy-saving and emission-reduction technologies and equipment to promote energy-saving and emission-reduction and reduce carbon emissions.
(5)
Urban greening can play a positive inhibitory role for large-scale and small-scale cities, but it is not obvious for medium-sized cities. Most medium-sized cities are undergoing fast economic growth and urbanization, with a predominance of secondary and tertiary industries, which will raise energy and resource consumption and result in higher carbon emissions. Enhanced urban greening is insufficient to offset the increased carbon emissions. Therefore, what really causes urban sprawl to play an important role in carbon emissions is not population size and greening level but urban economic development level and industrial structure. To alleviate the carbon emissions caused by urban sprawl, we should first consider optimizing the industrial structure of cities rather than rushing to control urban population size and expand urban green space.
In addition, there are some shortcomings in this paper that need to be improved in the follow-up work. The main problems are as follows: (1) This study uses NPP-VIIRS and DMSP-OLS night light data to measure urban sprawl. The spatial resolution of the images is low, which may limit the analysis at a finer scale. Future studies should use higher-resolution night light remote sensing images, such as Luojia 1-01 and Jilin-1 Guangxe-A, to achieve more accurate measurement and analysis of urban sprawl. (2) Although this paper constructs a relatively complete analysis framework for the impact of urban sprawl on carbon emissions and selects six representative control variables according to the previous studies, it is not comprehensive, for example, the impact of distance factors, policy factors and so on, are not considered. (3) This paper only conducts heterogeneity analysis from the perspective of city size; whether there is heterogeneity in other aspects remains to be further studied.

6. Conclusions

Based on DMSP-OLS and NPP-VIIRS nighttime light data from 2000 to 2020, the study focuses on the impact of urban sprawl on carbon emissions in East China. The main conclusions are as follows:
(1)
The impact of urban sprawl on carbon emissions in the East China region was positive from 2000 to 2020. After introducing external factors, this impact was more significant, indicating that the positive impact of urban sprawl on carbon emissions was the result of a comprehensive trade-off between various factors.
(2)
The external factors have different impacts on carbon emissions in the East China region, and the degree of impact is ranked from large to small as follows: urban economic level, industrial structure, tertiary industry and fixed investment have a positive effect on carbon emissions; urban greening and population size have a negative effect on carbon emissions.
(3)
The impact of urban sprawl on carbon emissions in the East China region showed differences in terms of urban size, and the impact was ranked from large to small as follows: small-sized cities, medium-sized cities, large-sized cities. The external factors also have different impacts on carbon emissions in different-sized cities. The improvement of economic development level in large-sized cities will significantly increase carbon emissions, while the enhancement of urban greening will significantly reduce carbon emissions. The increase in economic development level and industrial structure in medium-sized cities will significantly promote carbon emissions, but urban greening is not obvious. The increase in economic development level and industrial structure in small-sized cities will significantly increase carbon emissions, while the enhancement of urban greening will also significantly reduce carbon emissions. Meanwhile, in the impact of urban sprawl on carbon emissions, the factor of population size is not obvious for cities of any scale, which is contrary to people’s experiential judgment on this issue.
Based on the main conclusions of the research, this paper puts forward the following suggestions: First, urban sprawl in the East China region is significant. To prevent uncontrollable urban sprawl from increasing carbon emissions, concepts such as “compact city”, “smart growth” and new urbanization strategy should be learned from to control the disorderly expansion of cities and pursue the low-carbon, healthy and sustainable development of cities. Second, promote the optimization and upgrading of urban industrial structure in the East China region, and carry out technological upgrading and transformation of the secondary industry. Achieve the dual implementation of industrial development and energy conservation and emission reduction strategies. Third, for large-sized and small-sized cities, urban greening conditions need to be improved to achieve the goal of reducing carbon emissions; for medium-sized cities, industrial structure should be optimized as soon as possible, investment enterprise entry threshold should be raised, advanced technology should be introduced, energy utilization rate should be improved and energy structure should be changed to achieve the goal of reducing carbon emissions.

Author Contributions

Conceptualization, L.Z. and J.Y.; Methodology, L.Z. and J.Z.; Software, J.Z., X.L. and K.Z.; Validation, J.Z., X.L. and K.Z.; Formal analysis, L.Z. and J.Y.; Investigation, J.Z., X.L. and K.Z.; Resources, L.Z., J.Z. and J.Y.; Data curation, J.Z., X.L. and K.Z.; Writing—original draft, L.Z. and J.Z.; Writing—review & editing, L.Z. and J.Y.; Visualization, J.Z. and K.Z.; Supervision, L.Z. and J.Y.; Project administration, L.Z. and J.Y.; Funding acquisition, L.Z. and J.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Chengdu University of Technology’s 2020 Young and Middle-aged Key Teacher Development Funding Program (No. 10912-JXGG2021-02120) and Research Center of Sichuan County Economy Development Funding Program (No. XY2020020).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

DMSP-OLS and NPP-DNB data were processed and distributed by the National Oceanic and Atmospheric Administration (NOAA) and are freely available (https://www.ngdc.noaa.gov/eog/, accessed on 1 June 2023). No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Zhang, Q.; Zhang, H.X.; Zhao, D.; Cheng, B.D.; Yu, C.; Yang, Y.L. Does Urban Sprawl Inhibit Urban Eco-Efficiency? Empirical Studies of Super-Efficiency and Threshold Regression Models. Sustainability 2019, 11, 5598. [Google Scholar] [CrossRef] [Green Version]
  2. Shi, K.F.; Shen, J.W.; Wang, L.; Ma, M.G.; Cui, Y.Z. A multiscale analysis of the effect of urban expansion on PM2.5 concentrations in China: Evidence from multisource remote sensing and statistical data. Build. Environ. 2020, 174, 106778. [Google Scholar] [CrossRef]
  3. Cheng, Y.H.; Lu, J. A review of urban sprawl research. Urban Dev. Stud. 2016, 23, 45–50. [Google Scholar]
  4. Liu, X.H. Research on the influence of urban sprawl on haze pollution. Urban Probl. 2020, 4, 90–96. [Google Scholar] [CrossRef]
  5. Li, C.L. Exploring the Effect of Urban Sprawl on Carbon Emissions Based on the Multi-Source Nighttime Light Remote Sensing Data in Southwest China. Ph.D. Thesis, Southwest University, Chongqing, China, 2021. [Google Scholar]
  6. Seevarethnam, M.; Rusli, N.; Ling, G.H.T.; Said, I. A Geo-Spatial Analysis for Characterising Urban Sprawl Patterns in the Batticaloa Municipal Council, Sri Lanka. Land 2021, 10, 636. [Google Scholar] [CrossRef]
  7. Wu, H.; Fang, S.M.; Zhang, C.; Hu, S.W.; Nan, D.; Yang, Y.Y. Exploring the impact of urban form on urban land use efficiency under low-carbon emission constraints: A case study in China’s Yellow River Basin. J. Environ. Manag. 2022, 311, 114866. [Google Scholar] [CrossRef]
  8. Lee, C. Metropolitan sprawl measurement and its impacts on commuting trips and road emissions. Transp. Res. Part D Transp. Environ. 2020, 82, 102329. [Google Scholar] [CrossRef]
  9. Shi, K.; Xu, T.; Li, Y.; Chen, Z.; Gong, W.; Wu, J.; Yu, B. Effects of urban forms on CO2 emissions in China from a multi-perspective analysis. J. Environ. Manag. 2020, 262, 110300. [Google Scholar] [CrossRef] [PubMed]
  10. Hien, P.D.; Menb, N.T.; Tan, P.M.; Hangartner, M. Impact of urban expansion on the air pollution landscape: A case study of Hanoi, Vietnam. Sci. Total Environ. 2020, 702, 134635. [Google Scholar] [CrossRef]
  11. Ding, Y.H.; Ren, G. China’s National Assessment Report on Climate Change (I): Climate change in China and the future trend. Adv. Clim. Chang. Res. 2007, 3, 1–5. [Google Scholar]
  12. Cao, X.; Chen, J.; Hidefumi, I.; Osamu, H. A SVM-based method to extract urban areas from DMSP-OLS and SPOT VGT data. Remote Sens. Environ. 2009, 113, 2205–2209. [Google Scholar] [CrossRef]
  13. Meinshausen, M.; Meinshausen, N.; Hare, W.; Raper, S.C.B.; Frieler, K.; Knutti, R.; Frame, D.J.; Allen, M.R. Greenhouse-gas emission targets for limiting global warming to 2 °C. Nature 2009, 458, 1158–1162. [Google Scholar] [CrossRef] [PubMed]
  14. Bereitschaft, B.; Debbage, K. Urban form, air pollution, and CO2 emissions in large US metropolitan areas. Prof. Geogr. 2013, 65, 612–635. [Google Scholar] [CrossRef]
  15. Parry, I.W.H.; Williams, R.C.; Goulder, L.H. When Can Carbon Abatement Policies Increase Welfare? The Fundamental Role of Distorted Factor Markets. J. Environ. Econ. Manag. 1999, 37, 52–84. [Google Scholar] [CrossRef] [Green Version]
  16. IEA. World Energy Outlook 2008; International Energy Agency: Paris, France, 2008; pp. 382–386. Available online: https://iea.blob.core.windows.net/assets/89d1f68c-f4bf-4597-805f-901cfa6ce889/weo2008.pdf (accessed on 1 January 2023).
  17. Downs, A. Some realities about sprawl and urban decline. Hous. Policy Debate 1999, 10, 955–974. [Google Scholar] [CrossRef]
  18. Sun, C.Y.; Zou, Y. The lmpact of Urban Sprawl on Land Use Efficiency: Based on the Panel Data Analysis of Some Central Cities in Chengdu—Chongqing Economic Circle. J. Sichuan Adm. Inst. 2020, 5, 50–62. [Google Scholar]
  19. Ewing, R.; Rong, F. The impact of urban form on US residential energy use. Hous. Policy Debate 2008, 19, 1–30. [Google Scholar] [CrossRef]
  20. Hong, S.J.; Zhang, J.X. Study on Definition And Measurement Of Urban Sprawl: A Case Study On Yangtze River Delta Region. City Plan. Rev. 2013, 37, 42–45+80. [Google Scholar]
  21. Squires, G.D. Urban Sprawl: Causes, Consequences, and Policy Responses; Urban Institute Press: Washington, DC, USA, 2002. [Google Scholar]
  22. Jiang, F.; Liu, S.H.; Yuan, H.; Zhang, Q. Measuring urban sprawl in Beijing with geo-spatial indices. Acta Geogr. Sin. 2007, 17, 469–478. [Google Scholar] [CrossRef]
  23. Anas, A.; Rhee, H.J. Curbing excess sprawl with congestion tolls and urban boundaries. Reg. Sci. Urban Econ. 2006, 36, 510–541. [Google Scholar] [CrossRef] [Green Version]
  24. Li, Q.; Gao, N. Study on the Eco-environmental Effects of Urban Sprawl—Based on the Panel Data of 34 Large and Medium-sized Cities. Chin. J. Popul. Sci. 2016, 06, 58–67+127. [Google Scholar]
  25. Tong, L.Y. A review on definitions and measurements for urban expansion. World Reg. Stud. 2020, 29, 762–772. [Google Scholar]
  26. Pan, H.; Page, J.; Zhang, L.; Cong, C.; Ferreira, C.; Jonsson, E.; Nsstrm, H.; Destouni, G.; Deal, B.; Kalantari, Z. Understanding interactions between urban development policies and GHG emissions: A case study in Stockholm Region. Ambio 2020, 49, 1313–1327. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  27. Cheng, Y.H.; Wei, Q.; Jie, F.B.; Wang, K. Geographical transformations of urban sprawl: Exploring the spatial heterogeneity across cities in China 1992–2015. Cities 2020, 105, 102415. [Google Scholar]
  28. Jaeger, A.J.A.G.; Bertiller, R.; Schwick, C.C.; Kienast, D.F. Suitability Criteria for Measures of Urban Sprawl. Ecol. Indic. 2010, 10, 397–406. [Google Scholar] [CrossRef]
  29. Li, G.D.; Li, F. Urban sprawl in China: Differences and socioeconomic drivers. Sci. Total Environ. 2019, 673, 367–377. [Google Scholar] [CrossRef]
  30. Zhu, Q.Q.; Zeng, M.Z.; Jia, P.F.; Guo, M.Q.; Liang, X.; Guan, Q.F. Measuring the urban sprawl based on economic-dominated perspective: The case of 31 municipalities and provincial capitals. Geo-Spat. Inf. Sci. 2023, 2023, 2202201. [Google Scholar] [CrossRef]
  31. Feng, Y.C.; Qu, S.Y.; Peng, K.; Wang, J.T. Quantifying Urban Sprawl and Its Driving Forces in China. Discret. Dyn. Nat. Soc. 2019, 2019, 2606950. [Google Scholar]
  32. Hui, L.; Kai, W.C.; Guo, C. The Heterogeneity and Influencing Factors of Unban Sprawl in China’s Large Cities: A Panel Data Analysis of 35 Cities. Urban Dev. Stud. 2017, 11, 118–124. [Google Scholar]
  33. Glaeser, E.L.; Kahn, M.E. The greenness of cities: Carbon dioxide emissions and urban development. J. Urban Econ. 2010, 67, 404–418. [Google Scholar] [CrossRef] [Green Version]
  34. Ehrlich, M.V.; Hilber, C.A.L.; Schni, O. Institutional settings and urban sprawl: Evidence from Europe. J. Hous. Econ. 2017, 42, 4–18. [Google Scholar] [CrossRef]
  35. Xiao, W.X.; Ting, S.R.; Ying, Z. Dynamics of urban sprawl and sustainable development in China. Socio-Econ. Plan. Sci. 2019, 70, 100736. [Google Scholar]
  36. Ji, H.Z.; Hua, W.Y.; Fei, H.C. Urban land expansion under economic transition in China: A multi-level modeling analysis. Habitat Int. 2015, 47, 69–82. [Google Scholar] [CrossRef]
  37. Ioppolo, G.; Gao, J.; Yenneti, K.; Chen, W.; Wei, Y.D. Urban Land Expansion and Structural Change in the Yangtze River Delta, China. Sustainability 2015, 7, 10281–10307. [Google Scholar]
  38. Wei, Y.D.; Li, H.; Yue, W. Urban land expansion and regional inequality in transitional China. Landsc. Urban Plan. 2017, 163, 17–31. [Google Scholar] [CrossRef]
  39. Shan, B.G.; Shao, X.; Yu, S.; He, S.W. Analysis on Phase Characteristics and Driving Factors of Urban Sprawl in China. J. Geo-Inf. Sci. 2018, 20, 302–310. [Google Scholar] [CrossRef]
  40. Kowalczyk, C.; Kil, J.; Kurowska, K. Dynamics of development of the largest cities—Evidence from Poland. Cities 2019, 89, 26–34. [Google Scholar] [CrossRef]
  41. Marais, L.; Denoon-Stevens, S.P.; Cloete, J.; Moskalenko, N. Mining towns and urban sprawl in South Africa. Land Use Policy 2020, 93, 103953. [Google Scholar] [CrossRef]
  42. Lu, Q. The Current Situation of Urban Sprawl in China and Its Impact on Carbon Dioxide Emissions; Jinan University: Guangzhou, China, 2016. [Google Scholar]
  43. Ting, W.J.; Yu, X.; Fu, M.H.; Yuan, C.S. Measuring China’s Urban Sprawl in The Rapid Urbanization Period by Multi-Indicator Index Method. City Plan. Rev. 2019, 43, 11. [Google Scholar]
  44. Sun, X.; Crittenden, J.C.; Li, F.; Lu, Z.; Dou, X. Urban expansion simulation and the spatio-temporal changes of ecosystem services, a case study in Atlanta Metropolitan area, USA. Sci. Total Environ. 2018, 622–623, 974–987. [Google Scholar] [CrossRef]
  45. Tan, R.H.; Liu, Y.L.; Zhou, K.H.; Jiao, L.M.; Tang, W. A game-theory based agent-cellular model for use in urban growth simulation: A case study of the rapidly urbanizing Wuhan area of central China. Comput. Environ. Urban Syst. 2015, 49, 15–29. [Google Scholar] [CrossRef]
  46. Kovács, Z.; Farkas, Z.J.; Egedy, T.; Kondor, A.C.; Szabó, B.; Lennert, J.; Baka, D.; Kohán, B. Urban sprawl and land conversion in post-socialist cities: The case of metropolitan Budapest. Cities 2019, 92, 71–81. [Google Scholar] [CrossRef]
  47. Gregg, J.S.; Andres, R.J.; Marland, G. China: Emissions pattern of the world leader in CO2 emissions from fossil fuel consumption and cement production. Geophys. Res. Lett. 2008, 35. [Google Scholar] [CrossRef]
  48. Wang, P.; Pu, Y.R. Does urbanization impact the municipal infrastructure operation GHG emission? According to a systematic estimation framework. Int. J. Urban Ences 2021, 25, 501–521. [Google Scholar] [CrossRef]
  49. Ghaffar, A.; Puminjomnong, N. Valuation of Carbon Sources Vs Sinks and Validation through Land Use Land Cover Analysis: A Case Study of Bangkok Metropolitan Area. Jokull 2017, 63, 86–102. [Google Scholar]
  50. Jing, L.I.; Kevin, L.O.; Pingyu, Z.; Meng, G. Relationship Between Built Environment, Socio-Economic Factors and Carbon Emissions from Shopping Trip in Shenyang City, China. Chin. Geogr. Sci. 2017, 27, 722–734. [Google Scholar]
  51. MarcyOverman, B.; GPuga, H.; DiegoTurner, A.M. Causes of Sprawl: A Portrait from Space. Q. J. Econ. 2006, 121, 587–633. [Google Scholar]
  52. Carruthers, J.I. The Impacts of State Growth Management Programmes: A Comparative Analysis. Urban Stud. 2002, 39, 1959–1982. [Google Scholar] [CrossRef]
  53. Wang, J.T.; Zang, J.X. Different Effects of Urban Private Transport Public Transit on Urban Sprawl: An Empirical Test Based on the Panel Data of 65 Large and Medium-Sized Cities in China. Econ. Geogr. 2018, 38, 74–81. [Google Scholar] [CrossRef]
  54. Yang, L.; Wang, Y.Q.; Lian, Y.J.; Dong, X.; Liu, J.H.; Liu, Y.Y.; Wu, Z.H. Rational planning strategies of urban structure, metro, and car use for reducing transport carbon dioxide emissions in developing cities. Environ. Dev. Sustain. 2022, 25, 6987–7010. [Google Scholar] [CrossRef]
  55. Liu, Y.; Wang, J.; Han, Y.Q.; Liu, S.; Sun, Y.Y. Urban transport carbon dioxide (CO2) emissions by commuters in rapidly developing Cities: The comparative study of Beijing and Xi’an in China. Transp. Res. Part D Transp. Environ. 2017, 68, 65–83. [Google Scholar]
  56. Andong, R.F.; Sajor, E. Urban sprawl, public transport, and increasing CO2 emissions: The case of Metro Manila, Philippines. Environ. Dev. Sustain. 2017, 19, 99–123. [Google Scholar] [CrossRef]
  57. Sarkodie, S.A.; Owusu, P.A.; Leirvik, T. Global effect of urban sprawl, industrialization, trade and economic development on carbon dioxide emissions. Environ. Res. Lett. 2020, 15, 034049. [Google Scholar] [CrossRef]
  58. Ali, G.; Pumijumnong, N.; Cui, S. Valuation and validation of carbon sources and sinks through land cover/use change analysis: The case of Bangkok metropolitan area. Land Use Policy 2018, 70, 471–478. [Google Scholar]
  59. Shi, F.C.; Liao, X.; Shen, L.Y.; Meng, C.H.; Lai, Y.Y. Exploring the spatiotemporal impacts of urban form on CO2 emissions: Evidence and implications from 256 Chinese cities. Environ. Impact Assess. Rev. 2022, 96, 106850. [Google Scholar] [CrossRef]
  60. Li, C.L.; Li, Y.Q.; Shi, K.F.; Yang, Q.Y. A Multiscale Evaluation of the Coupling Relationship between Urban Land and Carbon Emissions: A Case Study of Chongqing, China. Int. J. Environ. Res. Public Health 2020, 17, 3416. [Google Scholar] [CrossRef]
  61. Jing, F.C.; Li, T.; Lin, Z.; Yi, H.D.; Yan, S.; Hui, Q.X.; Wei, L.J. Examining the impacts of urban form on air pollutant emissions: Evidence from China. J. Environ. Manag. 2018, 212, 405–414. [Google Scholar]
  62. Jin, G.; Chen, K.; Wang, P.; Guo, B.S.; Dong, Y.; Yang, J. Trade-offs in land-use competition and sustainable land development in the North China Plain. Technol. Forecast. Soc. Chang. 2019, 141, 36–46. [Google Scholar] [CrossRef]
  63. Li, X.Y.; Wang, L. Does Administrative Division Adjustment promote low-carbon city development? Empirical evidence from the “Revoke County to Urban District” in China. Environ. Sci. Pollut. Res. 2022, 30, 11542–11561. [Google Scholar]
  64. Fan, T.; Chapman, A. Policy Driven Compact Cities: Toward Clarifying the Effect of Compact Cities on Carbon Emissions. Sustainability 2022, 14, 12634. [Google Scholar] [CrossRef]
  65. Peiser, R. Decomposing Urban Sprawl. Town Plan. Rev. 2001, 72, 275–298. [Google Scholar] [CrossRef]
  66. Mills, E.S. Urban sprawl causes, consequences and policy responses: Gregory D. Squires, editor. Washington, D.C.: Urban Institute Press, 2002. Reg. Sci. Urban Econ. 2003, 33, 251–252. [Google Scholar] [CrossRef]
  67. Nechyba, T.J.; Walsh, R.P. Urban Sprawl. J. Econ. Perspect. 2004, 18, 177–200. [Google Scholar] [CrossRef] [Green Version]
  68. Wang, X.; Jiao, L.M. A Comparative Analysis of Urban Sprawl Characteristics of High-Density and Low-Density Cities Comparative Analysis of Large Cities in China and America. Econ. Geogr. 2020, 40, 70–78+88. [Google Scholar] [CrossRef]
  69. Liu, L.Y.; Tang, Y.L.; Chen, Y.Y.; Zhou, X.; Bernard, B.K. Urban Sprawl and Carbon Emissions Effects in City Areas Based on System Dynamics: A Case Study of Changsha City. Appl. Sci. 2022, 12, 3244. [Google Scholar]
  70. Yun, L.L.; Ling, X.L.; Yuan, W.Z.; Wan, L.H.; Yu, H.; Bedra, K.B. Peak Carbon Dioxide Emissions Strategy Based on the Gray Model between Carbon Emissions and Urban Spatial Expansion for a Built-Up Area. Appl. Sci. 2023, 13, 187. [Google Scholar]
  71. Jin, W.S.; Fang, L.C. The impact of urbanization on CO2 emissions in China: An empirical study using 1980–2014 provincial data. Environ. Sci. Pollut. Res. 2018, 25, 2457–2465. [Google Scholar]
  72. Carpio, A.; Ponce-Lopez, R.; Lozano-García, D.F. Urban form, land use, and cover change and their impact on carbon emissions in the Monterrey Metropolitan area, Mexico. Urban Clim. 2021, 39, 100947. [Google Scholar] [CrossRef]
  73. Wang, J.T.; Zhang, J.T. Measurement on the urban spreading in China: Empirical study based on the panel data of 35 large and middle cities. Economist 2010, 10, 56–63. [Google Scholar]
  74. Jiao, F.S.; Ke, Z. Study on the Impact of Urban Sprawl on Energy Carbon Emissions in the Yangtze River Economic Belt: Empirical Evidence from Night Light. Study Pract. 2022, 10, 30–39. [Google Scholar] [CrossRef]
  75. Liu, Y. Urban Sprawl in the Yangtze River Economic Belt and lts lmpact on Carbon Emissions. Spec. Zone Econ. 2021, 9, 33–36. [Google Scholar]
  76. Meng, Q.; Yan, L.X. Does Urban Sprawl Lead to Urban Productivity Losses in China? Empirical Study Based on Nighttime Light Data. J. Financ. Econ. 2015, 41, 28–40. [Google Scholar] [CrossRef]
  77. Zhang, J.Q.; Ji, X.J.; Xiu, C.L. Calculation of The Urban Sprawl Index Based on The Scale-Related Method: A Case Study of 230 Cities in China. City Plan. Rev. 2021, 45, 6. [Google Scholar]
Sustainability 15 11940 g001
Figure 1. General situation of East China (Note: The vector data of China‘s administrative regions in the figure are downloaded from the China Basic Geographic Information Center).
Figure 1. General situation of East China (Note: The vector data of China‘s administrative regions in the figure are downloaded from the China Basic Geographic Information Center).
Sustainability 15 11940 g001
Sustainability 15 11940 g002
Figure 2. Calibrated Nighttime Light Data for East China 2000–2020 (partial).
Figure 2. Calibrated Nighttime Light Data for East China 2000–2020 (partial).
Sustainability 15 11940 g002
Sustainability 15 11940 g003
Figure 3. Spatial and temporal distribution of urban sprawl in East China.
Figure 3. Spatial and temporal distribution of urban sprawl in East China.
Sustainability 15 11940 g003
Sustainability 15 11940 g004
Figure 4. Time variation of provincial total carbon emissions in East China.
Figure 4. Time variation of provincial total carbon emissions in East China.
Sustainability 15 11940 g004
Table 1. Comparison of DMSP-OLS and NPP-VIIRS data.
Table 1. Comparison of DMSP-OLS and NPP-VIIRS data.
Data NameDMSP-OLSNPP-VIIRS
Accessible Time Periods1992–2013April 2012–present
Orbital Height/km833824
Regression Period/min102102
Time Of Descending Node06:00 a.m./07:30 p.m.13:30 a.m./01:30 a.m.
Satellite Image Width/km30003040
Low light wavelength/μm0.4–0.950.5–0.9
Image Spatial Resolution/m1000500
Night Transit Time19:30–21:30 p.m.1:30 a.m.
Quantization Level6 bit14 bit
Data type/formataverage visible: *.TIFFvmflg: TIFF
Image CharacteristicsThe range of pixel values is from 0 to 63, which reflects the relative value of surface illumination and eliminates the interference of random noise such as moonlight and snow mountain reflections. However, there are benchmark differences between images collected by different sensors from the same series of satellites.The range of pixel values is from 0 to 255, and the sensor has a stronger ability to capture low light. Compared with DMSP-OLS data, the satellite sensor can more accurately reflect the relevant information on human economic activities and effectively improve the quality of night light data.
Table 2. Summary table of built-up area thresholds determined by nonlinear relationships for each year.
Table 2. Summary table of built-up area thresholds determined by nonlinear relationships for each year.
YearThreshold (DN)YearThreshold (DN)YearThreshold (DN)
200030200738201442
200132200838201543
200234200938201644
200334201038201746
200434201139201849
200536201240201951
200638201341202055
Table 3. Control variable set interpretation.
Table 3. Control variable set interpretation.
VariableAbbreviationVariable RepresentationReference
Population sizepopdenThe ratio of regional population to administrative area is used as the level of population size.Marais et al. [41],
Li [6]
Urban economic Development levelPGDPThe per capita GDP level is used to characterize the economic development level of a city.Liu [75],
Deng et al. [27]
Industrial structureGDP2The ratio of GDP of the secondary industry to GDP is used as the industrial structure.Liu [75],
Deng et al. [27]
Third industry structureGDP3The ratio of GDP of the tertiary industry to GDP is used as the industrial structure.Fang et al. [74],
Deng et al. [27]
Fixed investmentinvThe proportion of urban fixed asset investment in GDP is used to characterize fixed investment.Li [6],
Shi et al. [3]
The degree of urban greeningNDVINDVI index is used to represent the degree of greening in the area.Li [6],
Glaeser et al. [33]
Table 4. Multicollinearity test.
Table 4. Multicollinearity test.
Variable NameVariance Inflation FactorTolerance
popden1.7350.975
GDP3.4920.316
GDP24.7560.246
GDP33.1570.375
US2.0670.628
inv2.7310.548
NDVI1.9380.864
Mean VIF2.8390.565
Table 5. Impact of urban sprawl on carbon emissions: full sample estimation results.
Table 5. Impact of urban sprawl on carbon emissions: full sample estimation results.
Variable Name(1)
OLS		(2)
OLS		(3)
FE		(4)
FE
US2.158
(1.349)		3.342 **
(2.349)		0.388
(0.516)		0.795 **
(2.679)
popden −0.796
(−0.813)		 −0.156
(−0.846)
PGDP 0.679 ***
(10.167)		 0.368 ***
(5.893)
GDP2 1.726 ***
(3.498)		 0.156 ***
(4.698)
GDP3 0.358
(2.769)		 0.034
(2.136)
inv −0.178 ***
(−3.864)		 0.592 **
(2.458)
NDVI −4.235 ***
(−4.891)		 −0.998
(−5.691)
constant term1.806 ***
(3.711)		−2.459 ***
(−2.496)		2.894 ***
(10.498)		0.783 ***
(2.891)
Urban fixed effectNoNoYesYes
Year fixed effectNoNoYesYes
Number of observed samples1694161716941617
R20.0090.6810.9650.945
F1.689126.3426.8359.7
Note: (1) *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; (2) OLS represents the mixed panel regression model, FE represents the panel fixed-effect regression model; (3) the values in brackets are internal standard errors, the same below.
Table 6. Heterogeneity analysis of the whole sample city scale in East China.
Table 6. Heterogeneity analysis of the whole sample city scale in East China.
Variable NameLarge-Scale Urban Regression ResultsMedium-Scale Urban Regression ResultsSmall-Scale Urban Regression Results
(1)
OLS		(2)
FE		(3)
OLS		(4)
FE		(5)
OLS		(6)
FE
US2.371
(1.983)		0.284
(1.429)		2.694
(1.684)		0.681
(0.359)		4.059 **
(2.158)		1.351 **
(2.041)
popden−0.346
(0.659)		−0.159
(−0.973)		−1.087
(−1.659)		−0.157
(−0.648)		−2.598 *
(−2.584)		0.017
(0.354)
PGDP0.924 ***
(10.598)		0.265 ***
(2.594)		0.947 ***
(8.135)		0.248 ***
(2.149)		0.389 ***
(5.821)		0.314 ***
(5.321)
GDP20.159
(1.583)		−0.108 **
(−1.357)		0.247 ***
(1.584)		0.269 ***
(2.594)		1.508 ***
(3.241)		0.684 ***
(3.584)
GDP3−0.108
(−0.895)		−0.089
(−1.028)		0.548 **
(2.584)		0.128
(1.659)		−0.017 **
(−2.365)		0.158
(0.541)
inv0.002
(0.035)		−0.038
(−0.587)		0.002
(−0.001)		0.029
(0.821)		0.248 ***
(−4.215)		0.0358
(2.548)
NDVI−5.648 ***
(−8.157)		−1.852 ***
(−6.872)		−9.584 ***
(−10.658)		0.527
(0.821)		−3.165 ***
(−5.364)		−1.087 ***
(−3.541)
constant term−2.349 *
(−2.348)		2.138 ***
(10.487)		−3.158
(−2.654)		0.984
(1.548)		1.358
(0.541)		−0.367
(−2.514)
Urban fixed effectNoYesNoYesNoYes
Year fixed effectNoYesNoYesNoYes
Number of observed samples357357693693567567
R20.7470.9420.6210.8340.6960.975
F91.34306.5157.2692.5571.06179.8
Note: (1) *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively; (2) OLS represents the mixed panel regression model, FE represents the panel fixed-effect regression model; (3) the values in brackets are internal standard errors.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zhang, L.; Zhang, J.; Li, X.; Zhou, K.; Ye, J. The Impact of Urban Sprawl on Carbon Emissions from the Perspective of Nighttime Light Remote Sensing: A Case Study in Eastern China. Sustainability 2023, 15, 11940. https://0-doi-org.brum.beds.ac.uk/10.3390/su151511940

AMA Style

Zhang L, Zhang J, Li X, Zhou K, Ye J. The Impact of Urban Sprawl on Carbon Emissions from the Perspective of Nighttime Light Remote Sensing: A Case Study in Eastern China. Sustainability. 2023; 15(15):11940. https://0-doi-org.brum.beds.ac.uk/10.3390/su151511940

Chicago/Turabian Style

Zhang, Ling, Jiawei Zhang, Xiaohui Li, Kaidi Zhou, and Jiang Ye. 2023. "The Impact of Urban Sprawl on Carbon Emissions from the Perspective of Nighttime Light Remote Sensing: A Case Study in Eastern China" Sustainability 15, no. 15: 11940. https://0-doi-org.brum.beds.ac.uk/10.3390/su151511940

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop