Analysis on urban scaling characteristics of China's relatively developed cities.

China is undergoing rapid urbanization, but the speed and stage of urban development are quite heterogeneous among different regions and city types. Understanding the urban scaling characteristics of China's relatively developed cities is important for addressing environmental and social challenges. Within the scope of 114 third-tier-and-above Chinese cities, the research calculate the scaling parameters of various urban development variables with respect to urban population and urban GRP in different city types based on urban scaling quantitative models. Also, univariate and multivariate regression analyses were performed on the factors affecting urban electricity consumption. The research results show that the urban scaling characteristics of Chinese cities differ between different types of cities, industrial cities show unique scaling features compared to commercial cities and mixed-economy cities. Additionally, urban electricity consumption is found to be closely related to urban population, urban construction land area and street lamp number. The results can help different types of cities make targeted policies and provide insights for reducing resource consumption during the urbanization process.

1. Introduction

At present, China's urbanization is rapidly progressing. By 2018, China's urbanization rate had reached 59.58% (From the National Bureau of Statistics). The sizes and numbers of Chinese cities are both growing rapidly [1]. Although China's urbanization rate is quite fast in the world, its urbanization process still lags behind other countries [2, 3]. Besides, the complex Chinese national conditions give China's urbanization unique characteristics [4]. Due to this excessive urbanization speed, Chinese cities’ industrial structure, resource allocation, and technological progress do not match with their degree of development [5]. Mastering the process of urbanization in China needs quantitative models in the urban scaling study area.

City is the principal place of human life, and people have always maintained great interest in the development of urban systems [6]. However, due to the existence of various ever-changing systems such as society, economy, and infrastructure, cities can seem very complicated on the surface [7]. The explosive growth and rapid expansion of urban systems have led to fierce competition for space and resources between different urban systems, so the sizes and shapes of cities follow specific rules [8-10]. In fact, city systems correspond with life characteristics of biological systems [11, 12]. In biological systems, there is a sub-linear power-law relationship between the quality of mammalian species and the metabolic rate of organisms [13, 14]. This allometric growth scale law is modeled as a common feature observed by all biological systems [15]. The similarity between biological systems and urban systems makes it possible to conclude universal applicable urban scaling laws based on the general allometric growth scale models in biology [16].

By following the growth model of biological system, Bettencourt and his colleagues modeled the general scaling of urban systems: most characteristics of urban wealth creation and material energy use show index extensions with the increase of urban population and population interactions [16, 17]. In different geographies or different city scales, the scaling parameters of homogeneous indicators remain consistent [18, 19].

The urban indicators depicting the development of urban systems can be classified into three categories: innovative wealth indicators related to social wealth and social nature, such as inventions, crime rates and so on; physical energy indicators related to individual needs, such as water consumption, electricity consumption and so on [3, 20]; urban infrastructure indicators, such as the number of street lights, the length of the water supply pipeline, and so on [21]. Different categories of urban indicators present different characteristics as a city expands [22, 23].

A city's innovative wealth indicators tend to exhibit super-linear growth proportionality with city expansion [24]. Cities promote urban economic growth, wealth creation, and new ideas by attracting creative and innovative individuals [22, 25]. With the growth of urban talents and innovation, a city's socio-economic performance will exceed the proportional growth of the urban population [26]. The per capita invention and creativity of larger cities are significantly higher than those of smaller cities, and the gap is further increasing, which indicates that a city's innovative inventions have super-linear proportional relationships with the population growth [27-30].

The material and energy indicators of cities tend to be linearly proportional to the expansion of cities due to the close correlations with individual needs [31]. Scholars such as Kennedy explored the material and energy flows of 27 megacities with a population of more than 10 million to verify the consistency between the laws of resource flows in megacities and the general laws of urban scaling [32]. Further, the material and energy flow research on Chinese cities provides supporting evidence for the linear relationships between material energy indicators and city scaling [33].

Urban infrastructure indicators tend to obey sub-linear scaling laws as cities expand, that is, as the city population grows, the physical network usually grows more slowly than the city’s scale growth [34, 35]. This is mainly because of the existence of economies of scale [36].

Among the studies of urban scaling laws, how to determine the geographical extent of cities has always been a focus of discussion [37]. The criteria used to classify cities makes a big difference in the effectiveness of urban scaling models [6, 9]. Urban scale parameters are sensitive to urban partition and population size in the process of urban scaling [35, 38]. Research using public census data will continue to dominate the mainstream [39].

China's urban development is hugely unbalanced. Cities of different development levels in different regions show disordered states for both geographical and policy reasons. Based on numerous previous studies, the urban scaling laws can be more obvious in more developed cities. This study focuses on relatively developed cities in China, i.e., cities of the third tier and above on a five-tier scale. These cities have urban administrative units that are subject to high levels of urbanization and are thus more likely to belong to the same “urban system” [7].

With the development of the urban economy, the importance of primary industry will generally decline, while the proportion of secondary and tertiary industries will rise rapidly [40, 41]. The differences between the proportions of the first, second and third industries in different cities could lead to different scaling characteristics; thus, exploring the differences in scaling laws between different types of cities is taken into account in our work, while previous studies have ignored the impact of city type.

In term of variable selection, we have included many more indicators. As independent variables, both urban population and urban GRP are used to describe the scaling characteristics of cities. A broader variety of indicators concerning sustainable urban development are also included as response variables and are analyzed at finer levels.

China's urbanization process consumes a lot of energy, and electricity is an essential component [33, 42]. Electricity is not only the necessary energy directly needed in the commercial and industrial development of cities, but also urban residents’ most vital energy in daily life [43, 44]. Most importantly, electricity consumption is one of the primary sources of CO2 emissions [45]. To analyze the electricity consumption during urban scaling, univariate and multivariate regression analysis were conducted on the factors affecting urban electricity consumption in different types of cities. According to the analysis results, policies are recommended.

The main objectives of this research include the following aspects:

Calculation of the scaling parameters of urban development indicators as a function of urban population and urban GRP within the scope of 114 Chinese third-tier-and-above cities, and analysis of whether the scaling characteristics of different types of indicators are consistent with Bettencourt's conclusions;

Exploring the differences in urban scaling laws between industrial cities, commercialized cities and mixed-economy cities, and analyzing the reasons for the differences;

Carrying out univariate and multivariate regression analysis on the factors affecting urban electricity consumption of different types of cities and providing some suggestions according to the research results.

2. Materials and methods

2.1 Research data

2.1.1 Data sources

The research data mainly come from the China Urban Statistical Yearbook—2017 issued by the Department of Urban Social and Economic Investigation and the China Urban Construction Statistical Yearbook—2016 issued by the Ministry of Housing and Urban-Rural Development of the People's Republic of China [46, 47].

The data from the two yearbooks was cross-checked for data revision, and the China economic and social development statistical database was searched for the remaining missing data [48]. The processed data table containing 51 development variables of 263 Chinese cities was assembled as the original research data.

2.1.2. Selection of research cities and description of variables

As the development of China at this stage is unbalanced and insufficient, the development status varies significantly from city to city. As the level of urban development becomes higher, the laws followed by urban development are more pronounced. Therefore, selecting Chinese cities with better development could help to improve the pertinence of the research.

"2016 China Business Charm Ranking" was published by the "New First tier City Research Institute", a data news project of China Business Weekly, which ranked 338 Chinese prefecture-level cities on five dimensions of plasticity, including the concentration of business resources, urban hubs, urban people's activity, lifestyle diversity and future. According to the ranking results, there are 4 first tier cities, 15 new first tier cities, 30 second tier cities, 70 third tier cities, 90 fourth tier cities and 129 fifth tier cities. Based on the original data table and city classification results of the ranking, 114 cities, including 4 first tier cities, 15 new first tier cities, 27 second tier cities and 68 third tier cities, were selected for the research. The 114 selected cities are all third-tier-and-above cities. Their urban development is relatively mature and the construction of urban infrastructure is relatively better. The urban population, urban GRP and urban development resource consumption of the 114 cities account for the vast majority of Chinese cities. Exploring the scaling laws of these cities could help in understanding the overall urban development pace in China.

The 114 cities were divided into three different types of cities by the classification criteria proposed by Nelson [49]. Specifically, cities in which the proportion of secondary industry GRP is higher than the national average (the average of 263 cities) plus one standard deviation (58.00%) were classified as industrial cities; cities in which the proportion of tertiary GRP is higher than the average level plus one standard deviation (57.15%) were classified as commercial cities, and the rest were classified as mixed-economy cities. The classification results gave 14 industrial cities, 27 commercial cities and 73 mixed-economy cities in a total of 114 cities.

The 25 variables related to urban scaling selected in the research include urban population, GRP, total urban gas supply, total urban water supply, total urban electricity consumption, and so on. The administrative areas of the selected variables are municipal districts. The municipal district usually has high-level urbanization, massive population density and higher urban GRP. The municipal districts in China best fit the definition of cities in similar research studies.

2.2 Urban scaling model

The calculation model commonly used in urban scaling research is the quantitative model proposed by Bettencourt and his colleagues: N(t) represents a measure of the size of the urban population at time t; Y(0) is a normalized constant; Y(t) can represent a measure of material resources or social activity (e.g., wealth, patents and water consumption); the index β represents the general scaling parameter of urban development indicators with respect to population size. The model applies to cities in different years and different regions.

The leading urban development indicators were divided into infrastructure categories, individual demand categories and innovative wealth categories. The three types of urban indicators showed different scaling characteristics in the process of urban expansion. Between the different types of urban scaling indicators as the population size expands, the main differences are the general scaling parameter β values: β≈1 usually correspond to individual demands; β≈1.1–1.5>1 is usually related to social innovation wealth; β≈0.85<1 is usually associated with urban infrastructure. The differences in β values indicate different categories of indicators and show different scaling ratios as the urban population changes.

As a direct variable reflecting the degree of urban economic development, urban GRP plays a vital role similar to that of the urban population in the expansion of cities; thus, GRP was introduced into the model as a reactive indicator. Similar to model (1), the quantitative model of urban development indicators as a function of urban GRP can be described as: where Y(t) indicates the material resource or social activities; Y(0) is the normalization constant; the general scaling parameter β represents how different urban development indicators vary with GRP.

In order to facilitate the calculation, formula (2) is generally paired in the actual calculation process with: where y represents the urban development indicators that need to be explained, such as urban electricity consumption; x represents the urban scaling variables used to explain y, and in the model of this study x is urban population and urban GRP; c is the normalization constant; b represents the general scaling parameters of urban development indicators as a function of urban population or urban GRP.

The exploration of urban electricity consumption mainly uses multivariate regression analysis, and its calculation formula can be expressed as: where electricity represents the city's electricity consumption, which includes the city's total electricity consumption, urban industrial electricity consumption and urban residents' electricity consumption; c is the normalization constant; indicatori represents the variables affecting electricity consumption; bi serve as the parameters of explanatory variables of urban electricity consumption.

3. Results and discussion

The urban scaling parameters of various urban development indicators were calculated using the urban scaling models and the 2016 urban development yearbook data. A few urban development indicators that show better fitting effects are shown in our results. In addition, univariate and multivariate regression analyses were conducted on the factors affecting electricity consumption of different types of cities. Based on the results, some suggestions are made for reducing urban electricity consumption.

3.1 On urban scaling laws

Section 3.1.1 explores the overall scaling characteristics of China's third-tier-and-above cities and section 3.1.2 analyses the differences in the scaling laws between three different types of cities.

3.1.1 Scaling laws of all third-tier-and-above cities

Fig 1 shows the unitary regression results of urban GRP and urban population on a logarithmic scale.

Fig 1

Logarithmic regression of urban population and urban GRP.

(A) The blue bubble corresponds to the logarithmic population and GRP of each city; (B) The blue slash represents the logarithmic regression line of urban population and urban GRP.

As can be seen from Fig 1, for China's 114 third-tier-and-above cities, urban GRP shows a super-linear scaling relationship with the urban population (b = 1.111, R2 = 0.752), which is mainly due to the bidirectional positive feedback between the two. Cities with higher GRP are more mature and have more employment opportunities, thus attracting more urban population. More urban population could promote the further increase of urban GRP. As a result, urban GRP expands in a super-linear manner with urban population increase.

Table 1 shows the scaling parameters of different urban development variables relative to urban population and urban GRP.

Table 1

Scaling parameters of urban variables with urban population and urban GRP.

LN(RV)	With respect to LN(POP): First, Second and Third tier cities(n = 114)			With respect to LN(GRP): First, Second and Third tier cities(n = 114)
	b	Linearity	Adj-R2	b	Linearity	Adj-R2
TGS	1.247	Super-L	0.508	1.004	L	0.552
TWS	1.047	L	0.709	0.894	L	0.848
WSEC	0.994	L	0.669	0.853	L	0.81
FAI	0.947	L	0.667	0.807	Sub-L	0.794
CLA	0.856	L	0.748	0.717	Sub-L	0.853
DPL	0.916	L	0.635	0.784	Sub-L	0.765
RA	0.868	Sub-L	0.68	0.737	Sub-L	0.803
PA	0.788	Sub-L	0.503	0.669	Sub-L	0.594
GCA	0.932	L	0.637	0.816	Sub-L	0.802
SLN	0.761	Sub-L	0.548	0.675	Sub-L	0.707
LN(GDP)~LN(POP):b = 1.111, Adjusted R2 = 0.752

RV: Response variable

TGS: Total gas supply; TWS: Total water supply; WSEC: Whole society electricity consumption; FAI: Fixed asset investment; CLA: Construction land area; DPL: Drainage pipe length; RA: Road area; PA: Park area; GCA: Green coverage area; SLN: Street lamp number.

Super-L: Super-Linear; L: Linear; Sub-L: Sub-Linear; Adj-R: Adjusted R2.

In Table 1, it can be seen that urban development indicators related to individual needs, including total urban water supply, total urban electricity supply, fixed asset investment, built-up area and drainage pipeline length, scale linearly with the urban population.

Urban development indicators related to urban infrastructure including road area, park area and street light number expand sub-linearly with the urban population, mainly due to the existence of economies of scale. The construction and operation of the same infrastructure at higher density are more efficient, more economically viable and often result in higher quality services and solutions that are not possible in smaller locations, therefore often leading to economies of scale, which in turn leads to slower urban infrastructure construction speed. It’s worth noting that green coverage area expands linearly with population. Green coverage area includes not only park green area, but also residential green area and transportation green area, thus green coverage area is closely related to individual needs which leads to the linear relationship with population.

In general, the scaling parameter values of different types of urban development indicators are consistent with the conclusions that Bettencourt and colleagues have presented.

In Table 1, indicators related to urban energy consumption, including urban total gas supply, urban total water supply, and urban total electricity supply, scale linearly with urban GRP. Other indicators including fixed asset investment, construction land area, drainage pipeline length, road area, park area and street light number show sub-linear scaling with urban GRP, and these indicators can be collectively referred to as urban construction indicators. Urban GRP development is accompanied by energy consumption, while the restriction of material energy use efficiency makes the urban energy consumption follow certain linear proportional relationships with GRP increases. Also, the R2 values of the fitting equations between urban GRP and the indicators are significantly larger than those of the fitting equations between urban population and the indicators. This indicates that compared with the urban population, urban development indicators show stronger correlations with urban GRP, which means Chinese urban scaling characteristics could be better measured by urban GRP than the urban population.

It is worth noting that in urban material energy indicators, the total urban gas supply shows significantly different scaling characteristics comparing to water supply and electricity supply. The total gas supply scales super-linearly with the urban population (b = 1.247), while the total urban water supply and the urban electricity supply show linear proportional characteristics with urban population changes (b = 1.047, b = 0.994).

In Table 1, the correlation between total urban gas supply and urban population is significantly weaker than that between the urban population and the total urban water supply or the total urban electricity consumption. In cities, the use of urban gas supply is applicable mainly for residential households. Urban gas supply is not a necessary choice for residents because urban households have more options for cooking and heating methods, while urban water and electricity are necessary conditions for residents' family life. Therefore, the correlation between urban gas supply and urban population is significantly weaker.

Urban gas mainly includes natural gas and liquefied petroleum gas. Table 2 shows the scaling parameter values of different types of urban gas.

Table 2

Regression results of urban gas supply concerning urban population and urban GRP.

LN(RV)	With respect to LN(POP): First, Second and Third tier cities(n = 114)			With respect to LN(GRP): First, Second and Third tier cities(n = 114)
	b	Linearity	Adj -R2	b	Linearity	Adj-R2
TGS	1.247	Super-L	0.508	1.004	L	0.552
TNGS	1.439	Super-L	0.44	1.183	Super-L	0.488
TNGS(FR)	1.280	Super-L	0.403	1.025	L	0.423
TGS-LPG	1.046	L	0.27	0.917	L	0.335
TGS-LPG(FR)	0.837	Sub-L	0.196	0.713	Sub-L	0.227

RV: Response variable

TGS: Total gas supply; TNGS: Total natural gas supply; TNGS(FR): Total natural gas supply for residents; TGS-LPG: Total gas supply of LPG (Liquefied Petroleum Gas); TGS-LPG(FR): Total gas supply of LPG for residents

Super-L: Super-Linear; L: Linear; Sub-L: Sub-Linear; Adj-R: Adjusted R2.

According to Table 2, the urban natural gas supply scales super-linearly with urban population and urban GRP and the urban LPG supply scales linearly with urban population and urban GRP, but the household LPG consumption scales sub-linearly with urban population and urban GRP. Generally speaking, urban natural gas in cities is mainly transported by natural gas pipelines, while liquefied petroleum gas is mainly supplied by gas tanks. With the increase in city scale, the urban gas supply gradually shifts from liquefied petroleum gas, with lower safety and combustion efficiency, to natural gas, with higher safety and combustion efficiency. The urban residents also gradually abandon the use of liquefied petroleum gas and accept natural gas with unified transportation, so the super-linear scaling relationship between urban gas supply and urban population is mainly due to the increase in urban natural gas use.

Fig 2 shows that the increasing urban scale drives the construction of urban infrastructure, which leads to a considerable increase in the length of urban natural gas transmission pipelines, so the consumption of urban natural gas is greatly promoted. In conclusion, the super-linear scaling relationship between urban gas supply and urban population is mainly due to the rapid increase in the length of urban natural gas pipelines.

Fig 2

Trend of urban natural gas pipeline length with the urban population.

(A) The blue bubble corresponds to the logarithm of the population and the length of the gas pipeline; (B) The blue oblique line represents the logarithmic regression line between urban population and urban natural gas pipeline length.

3.1.2 Research on the scaling laws of different types of cities

According to the classification results of 114 cities, the urban scaling laws of the 14 industrial cities, 27 commercial cities, 73 mixed-economy cities are explored separately.

Fig 3 shows the proportional amounts of the average values of urban development indicators in three different types compared with the total average.

Fig 3

Proportions of average values of development indicators for different types of cities compared with the total average.

(A) The dark blue bars, light blue bars and gray blue bars respectively represent the proportions of the average values of urban development indicators in mixed economy cities, industrial cities, commercial cities compared with the total average.

As can be seen from Fig 3, all average values of development indicators in industrial cities are much lower than the total average, indicating that the third-tier-and-above industrial cities are relatively poorly developed. The number of commercial cities accounts for 23.7%, but all average values of development indicators in commercial cities are much lower than the total average, indicating that commercial cities are more attractive to Chinese people and have better development. The number of mixed-economy cities accounts for 64.0%, and the average values of their indicators are slightly under this value, but the gaps are smaller than industrial cities, indicating that Chinese mixed-economy cities are still in a period of development and transformation with no distinctive scaling characteristics.

Table 3 shows the scaling parameter calculation results of several urban development indicators of three different types of cities.

Table 3

Scaling parameters of urban development indicators with urban population and GRP in different city types.

City type	LN(RV)	With respect to LN(Population)			With respect to LN(GRP)
		b	Linearity	Adj-R2	b(Std.Error)	Linearity	Adj -R2
City-I (n = 14)	TGS	-1.236	Negative	0.063	-0.368	-	-0.046
	TWS	-0.769	-	-0.002	1.218	Super-L	0.502
	WSEC	0.924	L	0.029	1.487	Super-L	0.674
	FAI	-0.287	-	-0.072	0.062	-	-0.082
	CLA	0.078	-	-0.08	0.750	Sub-L	0.601
	DPL	0.432	-	-0.06	0.982	L	0.269
	RA	-0.374	-	-0.031	0.751	Sub-L	0.523
	PA	-0.127	-	-0.079	0.380	Sub-L	0.025
	GCA	-1.241	Negative	0.200	0.622	Sub-L	0.119
	SLN	0.536	Sub-L	0.031	0.326	Sub-L	0.038
	LN(GDP)~LN(POP):b = 0.133, Adjusted R^2 = -0.076
City-C (n = 27)	TGS	1.125	Super-L	0.689	0.876	L	0.646
	TWS	1.067	L	0.875	0.880	L	0.901
	WSEC	1.067	L	0.927	0.853	L	0.9
	FAI	0.907	L	0.834	0.765	Sub-L	0.9
	CLA	0.873	L	0.875	0.725	Sub-L	0.902
	DPL	0.826	Sub-L	0.743	0.740	Sub-L	0.908
	RA	0.840	Sub-L	0.78	0.727	Sub-L	0.888
	PA	0.750	Sub-L	0.61	0.657	Sub-L	0.712
	GCA	0.985	L	0.756	0.880	L	0.921
	SLN	0.723	Sub-L	0.607	0.649	Sub-L	0.746
	LN(GDP)~LN(POP):b = 1.148, Adjusted R^2 = 0.868
City-M (n = 73)	TGS	1.296	Super-L	0.458	1.067	L	0.543
	TWS	1.042	L	0.649	0.889	L	0.823
	WSEC	0.900	L	0.515	0.827	Sub-L	0.763
	FAI	1.024	L	0.633	0.883	L	0.823
	CLA	0.888	L	0.73	0.707	Sub-L	0.818
	DPL	0.988	L	0.621	0.802	Sub-L	0.714
	RA	0.912	L	0.638	0.743	Sub-L	0.739
	PA	0.802	Sub-L	0.446	0.657	Sub-L	0.523
	GCA	0.900	L	0.600	0.754	Sub-L	0.735
	SLN	0.810	Sub-L	0.509	0.721	Sub-L	0.705
	LN(GDP)~LN(POP):b = 1.077, Adjusted R2 = 0.663

City-I: Industrial cities; City-C: Commercial cities; City-M: Mixed-economy cities; RV: response variable; TGS: Total gas supply; TWS: Total water supply; WSEC: Whole society electricity consumption; FAI: Fixed asset investment; CLA: Construction land area; DPL: Drainage pipe length; RA: Road area; PA: Park area; GCA: Green coverage area; SLN: Street lamp number. Super-L: Super-Linear; L: Linear; Sub-L: Sub-Linear; Adj-R:Adjusted R2.

In Table 3, it can be seen that almost all urban development indicators fail to fit well with the urban population, and it seems that the development of China's industrial cities does not follow the general urban scaling laws. The rapid development of the secondary industry in industrial cities is worsening the ecological environment of cities, thus leading to the migration of urban residents, which can offset the immigration of urban population attracted by economic growth, so the population growth in industrial cities did not increase significantly with urban development. However, the development of industrial cities will lead to the increase of GRP inevitably, so the urban development indicators of industrial cities show good correlations with urban GRP.

The urban scaling characteristics of commercial cities conform to the general urban scaling laws basically and their urban scaling laws are the most apparent. Urban development indicators show perfect fitting effects with urban population and GRP in commercial cities. Commercial cities mainly depend on the development of the tertiary industry. The production and consumption of goods and the existence of consumers are critical factors for commercial urban development. Therefore, for commercial cities, more urban population and more potential consumers will bring faster urban development and higher urban GRP. In China, the most developed cities are all commercial cities. On the whole, the development of Chinese commercial cities is relatively mature and their scaling laws are more visible.

The urban development of mixed-economy cities combines the development characteristics of the other two types of cities. Although the urban development indicators have discernable correlations with urban population and urban GRP, the correlation intensity is weaker than that of commercial cities and stronger than that of industrial cities. For mixed-economy cities, balancing the development of the secondary industry and tertiary industry is an important issue. The ambiguity of the urban type attribute will affect the formulation of urban policies, thus reducing the attractiveness and development potential of cities, which will ultimately affect the health of urban development.

3.2 Research on influencing factors of urban electricity consumption

Among the indicators reflecting the state of urban development, urban electricity consumption is an essential one. Electricity is an indispensable resource in the process of industrial development and urban residents' living, and the massive consumption of electricity will inevitably exacerbate the destruction of the ecological environment. The univariate and multivariate regression analysis of influencing factors of urban electricity consumption for different types of cities can help us put forward some advice for efficient electricity use.

3.2.1 Univariate regression analysis of factors affecting urban electricity consumption

Table 4 lists the univariate regression results of the urban area, population, construction land area, street lamp number and per capita GRP concerning total electricity consumption, industrial electricity consumption, and household electricity consumption.

Table 4

Univariate regression results of electricity consumption in Chinese third-tier-and-above cities.

LN(IV)	With respect to LN(electricity use): First, second and third cities(n = 114)
	City-wide(WSEC)			Industry			Household
	b	Linearity	Adj-R2	b	Linearity	Adj-R2	b	Linearity	Adj-R2
UA	0.473	Sub-L	0.153	0.459	Sub-L	0.104	0.487	Sub-L	0.18
POP	0.994	L	0.669	1.033	L	0.532	1.003	L	0.755
CIA	1.051	L	0.738	1.087	L	0.574	1.011	L	0.763
SLN	0.961	L	0.64	1.021	L	0.533	0.941	L	0.684
P-GRP	0.908	L	0.193	0.985	L	0.167	0.841	Sub-L	0.18

IV: Independent variable; WSEC: Whole society electricity consumption

UA: Urban area; Pop: Population; CLA: Construction land area; SLN: Street Lamp number; P-GRP: Per capita GRP

Super-L: Super-Linear; L: Linear; Sub-L: Sub-Linear; Adj-R: Adjusted R2.

Correlation analysis of the urban electricity consumption reveals the population size effect, urban form effect and urban infrastructure effect, that is, urban population, construction land area and street lamp number positively correlate with urban electricity consumption. However, urban area and per capita GRP have little impact on urban electricity consumption. Urban area includes not only construction land but also land to be developed. The land area to be developed in different cities is very different from and consumes less electricity than construction land, so the urban area is weakly related to urban electricity consumption. The per capita GRP can be used to measure the affluence of the residents in different regions, but the affluence of urban residents does not have much effect on consumption of essential living resources such as electricity, so the per capita GRP and electricity consumption present a weak correlation.

Table 5 lists the univariate regression results of total electricity consumption, industrial electricity consumption and household electricity consumption relative to the urban area, population, construction land area, street lamp number and per capita GRP in different types of cities.

Table 5

Univariate regression results of urban electricity consumption in different types of cities.

City type	LN(IV)	With respect to LN(EU)
		City-wide EU(WSEC)			Industry			Household
		b	Linearity	Adj-R2	b	Linearity	Adj-R2	b	Linearity	Adj-R2
City-I (n = 14)	UA	0.447	Sub-L	0.114	0.517	Sub-L	0.097	0.114	Sub-L	-0.068
	POP	0.924	L	0.019	0.883	L	-0.019	1.532	Super-L	0.324
	CLA	1.706	Super-L	0.805	2.094	Super-L	0.828	0.494	Sub-L	0.049
	SLN	0.978	L	0.230	1.105	Super-L	0.189	0.822	Sub-L	0.281
	P-GRP	0.983	L	0.164	1.287	Super-L	0.206	-0.096	Negative	-0.087
City-C (n = 27)	UA	0.689	Sub-L	0.274	0.739	Sub-L	0.187	0.735	Sub-L	0.348
	POP	1.067	L	0.927	1.254	Super-L	0.800	1.010	L	0.903
	CLA	1.105	Super-L	0.86	1.264	Super-L	0.704	1.044	L	0.862
	SLN	1.106	Super-L	0.765	1.339	Super-L	0.701	1.038	L	0.730
	P-GRP	1.782	Super-L	0.364	2.050	Super-L	0.296	1.842	Super-L	0.430
City-M (n = 73)	UA	0.293	Sub-L	0.055	0.250	Sub-L	0.028	0.325	Sub-L	0.082
	POP	0.900	L	0.515	0.840	Sub-L	0.375	0.943	L	0.644
	CLA	0.990	L	0.668	0.959	L	0.525	0.981	L	0.737
	SLN	0.848	Sub-L	0.586	0.834	Sub-L	0.475	0.847	Sub-L	0.667
	P-GRP	0.662	Sub-L	0.146	0.690	Sub-L	0.132	0.616	Sub-L	0.142

City-I: Industrial cities; City-C: Commercial cities; City-M: Mixed-economy cities

IV: Independent variable; EU: Electricity use; WSEC: Whole society electricity consumption; UA: Urban area; POP: Population; CLA: Construction land area; SLN: Street lamp number; P-GRP: Per capita GRP

Super-L: Super-Linear; L: Linear; Sub-L: Sub-Linear; Adj-R: Adjusted R2.

The compositions of electricity consumption in different types of cities are different. According to Fig 4, it can be found that the power consumption of industrial cities is mainly concentrated in industrial electricity consumption, while household electricity consumption and other electricity consumption are relatively small. Compared with industrial cities, commercial cities and mixed-economy cities use significantly less industrial electricity, and household electricity and other types of electricity consume relatively more. The different compositions of electricity consumption may affect the scale characteristics of electricity consumption in different types of cities.

Fig 4

Electricity consumption compositions of different types of cities.

(A) The blue bars, brown bars and gray bars represent the proportions of industrial electricity, commercial electricity and other types of electricity respectively.

According to Table 5, the total electricity consumption of industrial cities has strong super-linear scaling relationships with urban construction land area and weak correlation with the urban population. The industrial electricity consumption of industrial cities accounts for a large proportion, so that the electricity consumption of industrial cities is mainly affected by industrial development. The development of industrial cities depends on the production of industrial enterprises, which are engaged in the exploitation and processing of natural resources. For industrial cities, the increase of construction land area usually means the development of urban industry, which requires further exploitation of resources. Natural resource exploitation requires a lot of electricity. As a result, the urban electricity consumption will rise super-linearly as construction land area increases. In the process of developing from a coal mining city to a comprehensive industrial city with petroleum, iron, steel and other industries, industrial cities will further aggravate urban electricity consumption.

The electricity consumption of commercial cities and mixed-economy cities reflects the impact of the urban population size effect, urban form effect and urban infrastructure effect, and is little affected by urban area and per capita GRP. The differences of the two types of cities are the R2 values of regression models.

3.2.2 Multivariate regression analysis of factors affecting urban electricity consumption

Table 6 shows the multivariate regression results of factors influencing the electricity consumption in Chinese third-tier-and-above cities.

Table 6

Multivariate regression results of urban electricity consumption.

LN(EU)	City type	LN(Independent variable)
		b					Adj- R2
		Urban area	Population	Construction land area	Street lamp number	Per capita GRP
City-wide EU (WSEC)	City-A	-	0.322**	0.769**	-	-	0.753
	City-I	-	0.793*	1.682***	-	-	0.870
	City-C	-0.180*	0.750***	0.471**	-	-	0.952
	City-M	-	-	0.702***	0.323*	-	0.694
Industrial EU	City-A	-	0.582***	-	0.578***	-	0.599
	City-I	-	-	2.094***	-	-	0.828
	City-C	-	1.254*	-	-	-	0.800
	City-M	-	-	0.644***	0.352*	-	0.549
Household EU	City-A	-	0.441***	0.371**	0.293***	-	0.827
	City-I	-	1.282*	-	0.669*	-	0.515
	City-C	-	0.881***	-	-	0.626***	0.939
	City-M	-	-	0.657***	0.360***	-	0.776

City-A: All cities; City-I: Industrial cities; City-C: Commercial cities; City-M: Mixed-economy cities; EU: Electricity use; WSEC: Whole society electricity consumption; Adj-R: Adjusted R2.

* Indicates whether the urban scaling parameter value is different from 0 (***p < 0.001; **p < 0.01; *p < 0.05;).

According to the multivariate regression results in Table 6, urban area and per capita GRP have little impact on the urban electricity consumption of commercial cities, and urban area shows a negative influence. The scaling of the urban area and the growth of urban per capita GRP reflect the rapid transformation of urban commercialization. A large number of labor-intensive urban enterprises may move their production plants to less-developed cities with abundant human resources and low labor costs. Besides, the optimization of urban electricity efficiency in more-developed cities may also alleviate the city's electricity burden. Therefore, the scaling of the urban area and the growth of per capita GRP could slow down the increase in urban electricity consumption for commercial cities.

Urban population, urban construction land area and street lamp number have significant positive influences on urban electricity consumption. The urban population and urban construction land area reflect the development trend of cities indirectly. The increase of urban population and construction land indicates the positive development of the urban economy and thus promotes the increase of urban electricity consumption. Street lamp number represents the city’s urban infrastructure construction level. Urban infrastructure includes energy facilities, transportation facilities and communication facilities, all of which are pure consumers of electricity resources. More street lamps correlate with better infrastructure construction, thus street lamp numbers show positive relationships with urban electricity consumption.

According to the b values in the multivariate regression models, when urban area, population and construction land area increase by 10%, the total electricity consumption of commercial cities will decrease by 1.80%, increase by 7.50% and increase by 4.71% respectively. When population increases by 10%, the industrial electricity consumption of commercial cities will increase by 12.54%. When other parameters remain unchanged, the per capita GRP in cities will increase by 10%, and the household electricity consumption in commercial cities will increase by 6.26%.

Comparing the results of the electricity consumption regression analysis between Table 5 and Table 6, there are many differences in the values of urban scale factors affecting urban electricity consumption. This indicates that the influence factors of urban electricity consumption are sensitive to the inclusion of other variables.

On the whole, the results of multivariate regression analysis of the factors affecting urban electricity consumption are consistent with the results of univariate regression analysis. The regression results show that Chinese urban electricity consumption is mainly affected by urban population, urban construction land area, and street lamp number, while urban area and per capita GRP have little impact on electricity consumption.

4. Conclusion

In the context of 114 Chinese third-tier-and-above cities, our results show that the overall development patterns in China are consistent with the general urban scaling laws. Previous studies have rarely been conducted within Chinese cities, and our research can help readers understand the scaling characteristics of them. Urban innovation wealth indicators scale super-linearly with urban population changes, urban infrastructure indicators scale sub-linearly with urban population changes, and urban material and energy indicators related to individual demands, except urban gas supply, scale linearly with urban population changes. The urban gas supply shows super-linear scaling with the urban population because of the rapidly increasing provision of urban natural gas transmission pipelines. In addition, the study analyzed the scaling of urban development indicators with GRP which has rarely used as independent variable. The goodness of fit with urban GRP as the independent variable appears better than that with the urban population, which manifests that the urban scaling characteristics of Chinese cities could be better modeled by urban GRP. The consistency between urban scaling laws and Chinese urban scaling characteristics makes it possible for China to promote urban development by referring to the experience of other countries. In the process of expanding the size of Chinese cities, in addition to considering the needs of the urban population for various development indicators, it should also be considered that the rapid development of the urban economy will also lead to increasing requirements for various indicators. Urban economy should play a more important role in the urban planning and construction process. In the context of all cities in China vigorously introducing talents, it should be considered whether the city’s economy is sufficient to support the city’s sustainable development. The development of Chinese cities should be based on people and more on economy.

For different types of cities, the differences between the values of scaling parameters indicate different development characteristics. The influence of city type on the characteristics of urban scaling has always been ignored. In industrial cities, the urban development indicators have no apparent correlation with urban population, but correlate strongly with urban GRP. Although the development of secondary industry in industrial cities can promote the development of urban GRP, it is challenging to attract talent. Also, the development of industry causes the deterioration of the ecological environment. While the Internet economy and service economy are taking up a greater and greater proportion of the national economy, the traditional industrial economy seems to show more weakness. How industrial cities can balance the relationships between economic development and ecological environment will be a question needing careful thought. For mixed-economy cities, their unclear urban attribute makes their development behave in a less straightforward way. The mixed-economy cities need to refer to the development experience of other types of cities for further development. The Chinese government should think about the differences between different types of cities and adjust measures to local conditions when making decisions on urban development. Industrial cities can properly transfer economic development centers to commercial development, and at the same time need to coordinate the relationship between the ecological environment and industrial development. Commercial cities need to pay more attention to building livable cities and attract more residents. Mixed-economy cities need clarify the center and direction of development, so as to improve the speed and quality of development.

Univariate and multivariate analyses of factors influencing electricity consumption show that urban electricity consumption is mainly affected by urban population, urban construction land area and street lamp number. To reduce the cities’ electricity consumption, urban residents should pay more attention to saving electricity, and more power-saving facilities should be adopted under urban infrastructure construction. In addition, urban area and per capita GRP have little impact on electricity consumption. Although the correlations are weak, increasing the per capita GRP of urban residents can still play a significant role in slowing down the increase of urban electricity consumption. To fundamentally solve the environmental pollution problems of urban electricity use, cities need to rely on new technologies to improve the efficiency of urban electricity and develop cleaner energy sources such as solar energy and wind energy.

In the future, further analysis on the development characteristics of China's industrial cities could help build urban scaling models with more generality and utility. Furthermore, the impact of urban development on the biophysical environment is also worthy of further investigation.

Detailed information of various indicators.

(DOC)

Click here for additional data file.

20 Apr 2020

PONE-D-20-08745

Analysis on urban scaling characteristics of China’s relatively developed cities

PLOS ONE

Dear Mr. Xingchao,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by Jun 04 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Bing Xue, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

1. We suggest you thoroughly copyedit your manuscript for language usage, spelling, and grammar. If you do not know anyone who can help you do this, you may wish to consider employing a professional scientific editing service.

Whilst you may use any professional scientific editing service of your choice, PLOS has partnered with both American Journal Experts (AJE) and Editage to provide discounted services to PLOS authors. Both organizations have experience helping authors meet PLOS guidelines and can provide language editing, translation, manuscript formatting, and figure formatting to ensure your manuscript meets our submission guidelines. To take advantage of our partnership with AJE, visit the AJE website (http://learn.aje.com/plos/) for a 15% discount off AJE services. To take advantage of our partnership with Editage, visit the Editage website (www.editage.com) and enter referral code PLOSEDIT for a 15% discount off Editage services.  If the PLOS editorial team finds any language issues in text that either AJE or Editage has edited, the service provider will re-edit the text for free.

Upon resubmission, please provide the following:

The name of the colleague or the details of the professional service that edited your manuscript

A copy of your manuscript showing your changes by either highlighting them or using track changes (uploaded as a *supporting information* file)

A clean copy of the edited manuscript (uploaded as the new *manuscript* file)

2. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

3. Thank you for including the following funding information within the acknoweldgements section of your manuscript; "The research was supported by the National Natural Science Foundation of China (No. 51478025)"

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

"The author(s) received no specific funding for this work."

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: Yes

Reviewer #4: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: I Don't Know

Reviewer #4: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

Reviewer #4: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: General comments:

1. 114 cities are not a decent sample size, especially when the sample is further classified into three groups. At such small sample size, especially the 14 industrial cities, how do you make sure the regression results are robust?

2. why use supply instead of gas and water consumption data while the electricity consumption data is used? Are there factors such as import of gas, water resource allocation would affect supply and consumption?

3. A table showing more details of the data used in the study would help the readers to know better about your research.

4. I don’t think park area is a good example for urban infrastructure, especially when explaining its sub-linear relationship with population or GRP.

5. The results in Table 5 should be explained with more details, for example the difference between industry consumption and household consumption. Different city types may have different constitution of industry and household consumption and whether it causes the super-linear scaling in industrial cities, but not commercial and mixed cities.

6. if you can access the electricity consumption data by different economic sector in each city, that would be an interesting perspective to see the scaling.

Specific comments:

line 180: how the 263 cities were selected, what are the relationship between 114 third-tier-and-above cities, 338 prefecture-level cities and 263 cities?

line 205: why compare with 263 cities? rather than 338 cities or 114 cities.

line 213: does the yearbook you mentioned provides all data at municipal districts level?

line 291: the sentence should be re-written to clarify which variables are individual needs, for example the drainage pipeline is individual needs or urban infrastructure?

line 336: “while urban water and water”

Reviewer #2: Based on urban scaling quantitative models, this paper analyzes the correlation among various index variables in urban and population and GRP, as well as the factors affecting urban electriurban consumption, taking 114 third-tier-and-above Chinese urban as examples. The amount of research data is large, the content is substantial, and the research conclusion has certain practical significance, but there are still the following problems.

(1)It is suggested that the introduction should be reorganized. On the one hand, the content of urban crime mentioned in the introduction is not highly related to the article. On the other hand, the introduction discusses the size and scale of the urban, but at the end of the article, it is mentioned that the analysis is based on the type of urban, and the relationship between the two is not high.

(2)In this paper, the urban scaling characteristics are mentioned and the third-tier-and-above urban are selected for analysis. However, the analysis process of the full text is mostly based on the urban type, and there is less research on the urban scale. It is suggested to add comparative analysis of the characteristics of different scale of urban in the analysis, and the research value may be better.

(3)The percentages of various indicators in different types of cities are compared and analyzed in Fig. 3, but it can be seen that there are great differences in the number of commercial cities, industrial cities and mixed-economy cities. Comparing the percentages of various cities with the percentages of consumed resources directly, the reliability and reliability of the results are not high. For comparative analysis, it is recommended to compare the average values of various cities.

(4)The article uses a lot of space to analyze the influencing factors of urban electricity consumption, which is not closely related to the theme of this article. Electricity consumption is an important indicator of urban scaling. The indicators ,such as the scale of urban construction land, urban population, water consumption, and transportation, can be used as key indicators that affect the size of the city. What are the reasons and particularities of analyzing the power consumption of the city?

(5)The conclusions mentioned in this paper can be used to formulate effective targeted policies for cities and to provide suggestions for reducing resource consumption in the process of urbanization. However, the end of this paper describes this part too little, and it is suggested to elaborate on different types or cities of different scales.

It is recommended to review after modification.

Reviewer #3: This manuscript analyse the urban scaling characteristics of China’s cites, the topic is interesting. However the novelty of this manuscript is lack. I did not see new findings on the urban scaling. The resluts of in figure 1 and 2 are not fresh knowledges.

For the indicators selection. the street lamp number was eclected as urban infrastructure indicator. In my opinions, the building and food supply are more important infrastructure than street lamp.

figure 2 shows the results of relationship of natural gas supply pipe line with population, but in some provinces in China, there have no such natural gas pipe due to lack of this kind of resource.

Reviewer #4: Urban scaling is an important research direction in geography and economics. Within the scope of 114 third-tierand-above Chinese cities, this article calculate the scaling parameters of various urban development variables with respect to urban population and urban GRP in different city types based on urban scaling quantitative models.

1. Introduction is long enough, and it is suggested that a part of literature review can be added. such as "1. Introduction; 2.Literature review;…"

2.Research data needs descriptive statistical analysis of the data and more detailed explanation of the data source and processing.

3.According to "2016 China Business Charm Ranking", there are 119 third-tier-and-above cities, Why this article selected 114 third-tier-and-above cities ?

4.What are the 25 variables related to urban scaling and why are they chosen?

5.If the research results can be compared with the existing research, the research contribution of this article will be enhanced.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Long Chen

Reviewer #2: No

Reviewer #3: No

Reviewer #4: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

3 Jun 2020

Dear Editors and Reviewers:

Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled “Analysis on urban scaling characteristics of China’s relatively developed cities” (ID: PONE-D-20-08745). Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval. Revised portions are marked in the paper.

The main corrections in the paper and the responds to the reviewer’s comments are as following.

Responds to the reviewer’s comments:

Reviewer #1:

General comments:

1. 114 cities are not a decent sample size, especially when the sample is further classified into three groups. At such small sample size, especially the 14 industrial cities, how do you make sure the regression results are robust?

Response:

The number of cities is not an important factor affecting the regression results. It can be found from Table 2 that the number of commercial cities is only 27, and the number of mixed-economy cities is 73. However, the regression effect of commercial cities is significantly better than that of mixed economy cities, which to some extent indicates that the number of different types of cities will not affect the reliability of the results. In fact, we have also calculated the scaling index of different urban development indicators with urban population and urban GRP under more industrial cities (including the fourth-tier and below industrial cities, a total of 42), and the results are generally consistent with the results of this study, so I think that even if the number of industrial cities is relatively small, it will not have a big impact on the reliability of the regression results. It is the great difference between different industrial cities that leads to the less obvious scaling characteristics of industrial cities.

2. Why use supply instead of gas and water consumption data while the electricity consumption data is used? Are there factors such as import of gas, water resource allocation would affect supply and consumption?

Response:

The total amount of urban gas supply refers to the total amount of gas supplied by gas enterprises to users during the reporting period, including the amount of sales and losses. Total urban water supply refers to the total amount of water supplied by water supply units during the reporting period, including effective water supply and leakage water. Leakage of gas and water mainly refers to the leakage caused by damage to pipelines and ancillary facilities, theft or failure of counting tables in the process of water supply. Leakage can help us judge whether the construction of urban infrastructure has been improved to some extent with the expansion of city scale. Therefore, the total water supply and gas supply were used in our study, rather than the consumption index after the leakage was removed. Urban power loss is usually caused by electricity transportation, which is an inevitable quantitative technical loss. Therefore, the city statistical yearbook usually only collects electricity consumption data and does not include electricity loss data. Therefore, we also use electricity consumption data instead of electricity supply data in our research.

3. A table showing more details of the data used in the study would help the readers to know better about your research.

Response:

According to your suggestion, we have made a table to help readers understand the specific information of the research indicators, including the indicator name, unit, specific classification and data source. This table has been included in the supporting information.

4. I don’t think park area is a good example for urban infrastructure, especially when explaining its sub-linear relationship with population or GRP.

Response:

Urban park is an important part of China's urban construction planning, and the park area has always been an important indicator to measure the level of China's urban modernization. The park plays an important role in helping to improve the city's ecological environment and residents' living environment. Therefore, we believe that park area should be an important part of urban infrastructure.

Of course, we carefully considered your opinion and decided to increase the "green area coverage" as one of the urban infrastructure indicators. The urban green space coverage area covers a wider range. In addition to the park green area, it also includes the city's residential green space and traffic green space, so it may be a better indicator of urban infrastructure. The two indicators of comprehensive park area and green area coverage.

5. The results in Table 5 should be explained with more details, for example the difference between industry consumption and household consumption. Different city types may have different constitution of industry and household consumption and whether it causes the super-linear scaling in industrial cities, but not commercial and mixed cities.

Response:

Based on your suggestions, we calculated the urban electricity consumption proportions in different types of cities and analyzed its possible impact on the scale of urban electricity consumption. The modifications are shown in the article.

6. If you can access the electricity consumption data by different economic sector in each city, that would be an interesting perspective to see the scaling.

Response:

We have worked hard to search the electricity consumption data of different economic sectors in various cities. This is indeed a very difficult thing because many data are missing. I'm sorry that this part of the content may not be supplemented in this research, but your suggestion is indeed very enlightening. It is a topic that is worth writing an article for in-depth research. We will try to fill this gap in the future.

Specific comments:

line 180: how the 263 cities were selected, what are the relationship between 114 third-tier-and-above cities, 338 prefecture-level cities and 263 cities?

Response:

China’s business charm rankings have divided 338 prefecture-level cities in China. We hope to conduct research within these cities. However, the China City Statistical Yearbook and China City Construction Statistical Yearbook do not include all 338 prefecture-level cities. At the same time, there are many data missing in the two yearbooks. After the cross-comparison and supplement of the data, only the data of 263 cities is relatively complete, so the data table containing 263 cities is used as the original data. 263 cities include all prefecture-level cities of different levels, and we only want to select more developed cities of third-tier and above to conduct research, so only 114 cities of third-tier and above are taken as the final research scope.

line 205: why compare with 263 cities? Rather than 338 cities or 114 cities.

Response:

The city type division method used in the study was conducted by comparing the proportions of different industries in each city with the national average. The 263 urban development data obtained are defaulted to represent the national average. So our compare is within the scope of the 263 cities, but not 338 or 114 cities.

line 213: does the yearbook you mentioned provides all data at municipal districts level?

Response:

Yes, these two yearbooks can provide all the data at the municipal level, among which some data are missing. We have filled in the data by consulting the provincial and municipal yearbooks and online databases.

line 291: the sentence should be re-written to clarify which variables are individual needs, for example the drainage pipeline is individual needs or urban infrastructure?

Response:

The table including the indicator name, unit, specific classification and data source has been included in the supporting information.

line 336: “while urban water and water”

Response:

I'm sorry, it was my fault. It should be “while urban water and electricity”. It has been modified. Sorry again.

Reviewer #2:

1. It is suggested that the introduction should be reorganized. On the one hand, the content of urban crime mentioned in the introduction is not highly related to the article. On the other hand, the introduction discusses the size and scale of the urban, but at the end of the article, it is mentioned that the analysis is based on the type of urban, and the relationship between the two is not high.

Response:

Based on your suggestions, we have made the necessary cuts and modifications to the introduction.

1. The part of the city crime not discussed in this article has been deleted.

2. The impact of city size on city development is an important part of our research. Using the urban population to represent the size of the city is an important prerequisite for the Bettencourt scale model. On this basis, we have increased the urban GRP to represent the economic scale of the city. Therefore, the scaling indexes of various urban development indicators with the urban population and urban GRP reflects the impact of urban size on urban development. City type is an important dimension of our research. Through separate analysis of different types of cities, it can help us understand the differences in scalig characteristics between different types of cities. Therefore, city size and city type are two dimensions of research, rather than two variables with low correlation.

2. In this paper, the urban scaling characteristics are mentioned and the third-tier-and-above urban are selected for analysis. However, the analysis process of the full text is mostly based on the urban type, and there is less research on the urban scale. It is suggested to add comparative analysis of the characteristics of different scale of urban in the analysis, and the research value may be better.

Response:

The answer to this question is similar to the answer to the previous question. The impact of city size on urban development is an important part of our research. Using the urban population to represent the size of the city is an important prerequisite for the Bettencourt scale model. On this basis, we have added urban GRP to represent the urban economic scale. Therefore, the scaling indexes of various urban development indicators along with urban population and urban GRP reflects the impact of urban size on urban development. City type is an important dimension of our research. Through separate analysis of different types of cities, it can help us understand the differences of scaling characteristics between different types of cities.

3. The percentages of various indicators in different types of cities are compared and analyzed in Fig. 3, but it can be seen that there are great differences in the number of commercial cities, industrial cities and mixed-economy cities. Comparing the percentages of various cities with the percentages of consumed resources directly, the reliability and reliability of the results are not high. For comparative analysis, it is recommended to compare the average values of various cities.

Response:

Based on your suggestions, we have modified this part of the content and compared the average of different types of cities with the overall average, which is indeed more intuitive.

4. The article uses a lot of space to analyze the influencing factors of urban electricity consumption, which is not closely related to the theme of this article. Electricity consumption is an important indicator of urban scaling. The indicators ,such as the scale of urban construction land, urban population, water consumption, and transportation, can be used as key indicators that affect the size of the city. What are the reasons and particularities of analyzing the power consumption of the city?

Response:

Urban electricity is not only a necessary energy for the development of urban commerce and industry, but also an important energy for the living of urban residents. More importantly, urban electricity consumption is also an important source of urban CO2 emissions. Therefore, electric energy consumption is closely related to the ecological environment. At the same time, compared with other indicators, the urban statistical yearbook provides a more detailed dimension of urban energy consumption data, which provides the possibility for in-depth analysis of urban energy consumption. In-depth analysis of urban electricity consumption can not only provide a reference for energy conservation, but also help relieve the pressure on the ecological environment.

5. The conclusions mentioned in this paper can be used to formulate effective targeted policies for cities and to provide suggestions for reducing resource consumption in the process of urbanization. However, the end of this paper describes this part too little, and it is suggested to elaborate on different types or cities of different scales.

Response:

According to your suggestion, we have supplemented the conclusion. For different types of cities, we have put forward different policy recommendations.

Reviewer #3:

1.This manuscript analyse the urban scaling characteristics of China’s cites, the topic is interesting. However the novelty of this manuscript is lack. I did not see new findings on the urban scaling. The resluts of in figure 1 and 2 are not fresh knowledges.

Response:

The law of urban expansion scale is the universal law of urban expansion established by Bettencourt based on the urban population, and it shows regularity without being restricted by space and time. China is rarely studied because of its complicated urbanization. We select more developed Chinese cities to try to verify whether the scaling law established by Bettencourt is applicable in China, and introduce urban GRP as an important independent variable. Although such research is not new, this universality provides a basis for China to master the urbanization process and make reference to the urbanization of other countries.

2.For the indicators selection. the street lamp number was eclected as urban infrastructure indicator. In my opinions, the building and food supply are more important infrastructure than street lamp.

Response:

The number of city street lights is closely related to the urban population and living environment, and is the most dense urban infrastructure. Urban street lights are often concentrated in areas with a high population density. The increase in the total number of street lights can reflect the population aggregation and infrastructure construction caused by urbanization. Therefore, the total number of urban street lights has been adopted by us as one of the infrastructure indicators. The city's architecture and food supply are relatively complex, which may involve many types of indicators. It may not be appropriate to conduct research directly as infrastructure, and it needs to be analyzed separately in subsequent studies.

3.Figure 2 shows the results of relationship of natural gas supply pipe line with population, but in some provinces in China, there have no such natural gas pipe due to lack of this kind of resource。

Response:

The data used in our research are all municipal districts. The natural gas pipeline facilities in more developed cities are relatively complete. China has also adopted the "west-to-east natural gas transmission" to complete the construction of pipeline facilities nationwide. According to our research results, it can be found that the urban natural gas pipeline shows a super-linear scaling relationship with the urban population, indicating that the construction of natural gas pipeline varies greatly among different cities, and that there is still much room for improvement in the construction of natural gas pipeline in Chinese cities.

Reviewer #4:

1. Introduction is long enough, and it is suggested that a part of literature review can be added. such as "1. Introduction; 2.Literature review;…"

Response:

We have cut and modified the introduction part of the article, but did not split it into two parts. In the introductory part, we hope to be able to tell the reader the background, research content, and significance of the article as if telling a story. Splitting the introductory part may cause the part to lose its integrity and be unclear. If you think there is still a problem with the revised introduction, we will modify it again according to your suggestions

2.Research data needs descriptive statistical analysis of the data and more detailed explanation of the data source and processing.

Response:

A table including the indicator name, unit, specific classification and data source has been included in the supporting information.

3.According to "2016 China Business Charm Ranking", there are 119 third-tier-and-above cities, Why this article selected 114 third-tier-and-above cities ?

Response:

According to the “2016 China Business Charm Ranking”, there are 119 third-tier cities and above in China, but the data of five of these cities (including three second-tier cities and two third-tier cities) are seriously missing or not included in China city statistical yearbook, so our study only adopt 114 third-tier cities and above.

4.What are the 25 variables related to urban scaling and why are they chosen?

Response:

The 25 variables associated with urban growth include:

population density, per capita road area, urban area, urban population, built-up area, urban construction land area, public facility land, fixed asset investment, street lights number, road area, park area, land area, GRP, per capita GRP, total water supply, electricity consumption for the whole society, industrial electricity, residential electricity, total gas supply, number of buses, operating vehicles, number of taxis, green area coverage and Three industries account for the proportion of GRP.

The 25 variables are all from China's urban statistical yearbook and China's urban construction statistical yearbook, covering most of the indicators of urban innovation wealth, material energy or infrastructure, which are also important factors to measure the level of urbanization. Therefore, we selected these indicators for analysis and calculation.

5.If the research results can be compared with the existing research, the research contribution of this article will be enhanced.

Response:

Thank you for your suggestions and we have supplemented the conclusion. Comparing our research results with previous studies can help readers understand the significance of the research.

We have tried our best to improve the manuscript and made some changes in the

manuscript. These changes will not influence the content and framework of the

paper. And here we did not list the changes but marked in revised paper.

We appreciate for editors and reviewers’ warm work earnestly and hope that the

correction will meet with approval.

Once again, thank you very much for your comments and suggestion.

Best regards,

Xingchao Liu and Zhihong Zou

Submitted filename: Response to Reviewers.doc

Click here for additional data file.

30 Jun 2020

PONE-D-20-08745R1

Analysis on urban scaling characteristics of China’s relatively developed cities

PLOS ONE

Dear Dr. Xingchao,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Aug 14 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Bing Xue, Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

Reviewer #4: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

Reviewer #4: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

Reviewer #4: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #4: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #4: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have addressed most of my comments in the first round of review, I appreciate the thorough revisions by the authors. Many parts of the manuscript were greately improved in this round. However, I don't think one of my concerns, which is the small sample size for the 14 industrial cities has been addressed properly. My concern is from the pespective of statistics, saying that you need a decent sample size to conduct any regression analysis and to deliver valid results. Apparently 14 is not a decent size for regressions. The negative Adj-R2 in Table 2 also indicates that you have small R-square and a small sample size. One possible solution maybe the inclusion of panel data for the cities to increase the sample size. Beyond that, I have no further comments.

Reviewer #2: The data source of this article is true, the research method is feasible, and the research conclusion is credible. The author responds to the expert's opinions one by one and provides a detailed explanation.

Reviewer #4: The authors s have adequately addressed my comments, and I feel that this manuscript is now acceptable for publication

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #4: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

7 Jul 2020

Dear Editors and Reviewers:

Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled “Analysis on urban scaling characteristics of China’s relatively developed cities” (ID: PONE-D-20-08745). Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval.

The responds to the reviewer’s comments are as following.

Responds to the reviewer’s comments:

Reviewer #1:

I don't think one of my concerns, which is the small sample size for the 14 industrial cities has been addressed properly. My concern is from the perspective of statistics, saying that you need a decent sample size to conduct any regression analysis and to deliver valid results. Apparently 14 is not a decent size for regressions. The negative Adj-R2 in Table 2 also indicates that you have small R-square and a small sample size. One possible solution maybe the inclusion of panel data for the cities to increase the sample size.

Response to Reviewer #1:

According to your suggestion that we should increase the sample number of industrial cities, we selected 42 industrial cities for the calculation of the scaling index of urban development indicators with the change of urban population and urban GRP. The 42 industrial cities include not only 14 third-tier and above industrial cities, but also 28 industrial cities from the fourth and fifth tier.

Table 1 shows the scaling index results of the development indicators of 42 industrial cities with the change of urban population and urban GRP, and Table 2 shows the scaling index results of 14 third-tier and above industrial cities with the change of urban population and urban GRP.

From Table 1 and Table 2, it can be found that, even if the selection scope of industrial cities is expanded and the sample number of industrial cities is increased, the Adj-R2 obtained by 42 industrial cities is not significantly improved compared with the previous 14 industrial cities. In the regression results with urban population, Adj-R2 is mostly within the range of 0.1-0.3, which is still not a reliable R2 value. In the regression results with urban GRP, except that the R2 value of fixed asset investment has been significantly improved, the other R2 values are still at a low range level.

In fact, we can't find any more industrial cities in the third-tier and above.If we want to increase the sample size, we may need to include some underdeveloped industrial cities. However, this conflicts with our research topic, and does not result in more reliable regression results (R2 values did not increase significantly in 42 industrial cities).

The number of industrial cities in China is relatively small, and the development level and urbanization level of industrial cities are relatively low. The unremarkable scaling characteristics of industrial cities in the research results is more due to the lag of the development models of industrial cities. The development of industrial cities is more dependent on industry. Compared with commercial cities and mixed economy cities with a larger commercial proportion, they are less attractive to residents. The ecological environment deterioration caused by industrial development further intensifies the outflow of local residents. Therefore, urban physical energy indicators, infrastructure indicators and other development indicators that are closely related to residents have failed to perform well in proportion to the urban population. Therefore, the number of third-tier and above industrial cities in the study is not an important factor affecting the scaling results, but the development characteristics of industrial cities themselves are the main reason.

The development characteristics of China's industrial cities are quite different from those of other types of cities, suggesting that the Chinese government should take more measures in accordance with local conditions when making urban development policies.

In our study, industrial cities show very different development characteristics, which are often neglected in previous studies.It is worth further in-depth study to explore the development characteristics of industrial cities. We also hope to explore and analyze the scaling laws of industrial cities separately in the future, which is really an interesting and worthwhile subject.

We appreciate for editors and reviewers’ warm work earnestly and hope that the response will answer the reviewer’s question.

Once again, thank you very much for your comments and suggestion.

Best regards,

Xingchao Liu and Zhihong Zou

Submitted filename: Response to Reviewers.doc

Click here for additional data file.

10 Jul 2020

Analysis on urban scaling characteristics of China’s relatively developed cities

PONE-D-20-08745R2

Dear Dr. Xingchao,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Bing Xue, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

15 Jul 2020

PONE-D-20-08745R2

Analysis on urban scaling characteristics of China’s relatively developed cities

Dear Dr. Xingchao:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Bing Xue

Academic Editor

PLOS ONE