Spatial characteristics and coupling coordination between carbon emission efficiency and industrial structure in three metropolitan areas of Jiangsu Province,China

Abstract

It is of great practical significance to study the spatial characteristics of carbon emission efficiency, industrial structure, their coupling and coordination relationship for China's green development and industrial structure transformation in the new era. From the perspective of coupling, coordination and space, this paper analyzes and summarizes the spatial characteristics of carbon emission efficiency and industrial structure of 19 cities in three metropolitan areas of Jiangsu Province during 2009–2019 and their coupling and coordination relationship. The carbon emission efficiency in this study is represented by the carbon emission economic efficiency index and carbon emission social efficiency index. The results show that (a) the high-emission centers in the three metropolitan areas developed from “three centers” in 2009 to “five centers” in 2019. The continuous high-energy consumption of the secondary industry and the growth of the economic aggregate of the third industry kept the regional high carbon dioxide emissions. (b) The average value of carbon emission economic efficiency in 19 cities continued to increase, indicating that the contribution rate of the same amount of carbon emissions to economic income gradually increased; the growth range of carbon emission economic efficiency index is greater than that of carbon emission social efficiency index, indicating that carbon emission has a more significant effect on the improvement of regional economic development than on the improvement of public service level and residents’ living quality. (d) The solidification degree of carbon emission efficiency is greater than that of the industrial structure (solidification degree carbon emission social efficiency > carbon emission economic efficiency > industrial structure). The high-grade industrial structure in Xuzhou metropolitan area is closely related to the improvement of carbon emission economic efficiency and carbon emission social efficiency, and both are in moderate antagonism. The rationalizing industrial structure in Nanjing metropolitan area is closely related to the improvement of carbon emission economic efficiency, which is in high coordination run-in. The concentration degree of industrial structure in Suzhou-Wuxi-Changzhou metropolitan area is closely related to the improvement of carbon emission economic efficiency and carbon emission social efficiency, which are in polar coordination coupling and high coordination run-in, respectively. The proposed coupling path of carbon emission efficiency-industrial structure can not only alleviate the dynamic disharmony in different cities but also effectively improve the coupling degree in cities.

Introduction

With the rapid growth of urban economy and urbanization in China, metropolitan, as an important carrier of regional spatial organization, plays a leading role in regional economic development and becomes a representative of regional core competitiveness. At the same time, behind the rapid growth is the increase of energy consumption and carbon dioxide emissions year by year. In the next two to three decades, the industrial structure will contribute 60% to carbon dioxide emissions, which has seriously restricted the sustainable development of China's economy and national security. Therefore, it is of great significance to study the impact of industrial structure on China's carbon dioxide emissions to find ways to reduce emissions and achieve emission reduction targets.

Literature review

Measurement of urban carbon emissions

Cities occupy less than 2% of the earth's area but consume 75% of the energy. They have small geographical areas, complex economic activities, and high energy consumption intensity. Therefore, a more comprehensive and systematic statistical analysis of urban energy consumption has always been a difficulty, and it is also the biggest challenge facing urban energy consumption and carbon dioxide accounting. Existing scholars have studied the accounting methods of urban carbon emissions from the perspectives of input–output table, urban carbon emission inventory rule, energy balance table, primary energy consumption, terminal energy consumption, direct and indirect emissions.^1–6 In summary, accounting accuracy and data acquisition difficulty should be considered comprehensively in urban carbon emission accounting methods.

Carbon emissions and industrial structure

To solve the problem of carbon emission, it is necessary to clarify the influencing factors of carbon emission. Scholars around the world have conducted a large number of studies on the influencing factors of carbon emissions (urbanization, foreign direct investment, economic growth, industrial structure, etc). Most scholars focus on the impact of energy structure, industrial structure, and economic structure on carbon emissions. Among them, industrial structure is an important factor affecting urban carbon emissions. Relevant scholars have demonstrated that changes in energy structure, economic structure, and industrial structure will have a significant impact on carbon emissions by studying oil and non-oil countries in the south of the Sahara, ⁷ countries at different levels of development,^8–12 Austria and Czechoslovakia. ¹³

Compared with other countries, China is in a critical period of transforming its economic development mode, there are more abundant researches on the relationship between industrial structure and carbon emission, the indicators to measure industrial structure are more diverse. Relevant studies show consistency in the presentation of conclusions.(a) There is a long-term equilibrium relationship between industrial structure and carbon emission. ¹⁴ Industrial structure is one of the main driving factors of carbon emission.^15,16 (b) Different industries have different impacts on carbon emissions.^17,18 There is a positive correlation between the proportion of the secondary industry and carbon emission, but the adjustment of the internal structure of the secondary industry has a positive effect on carbon emission reduction. ¹⁹ At the same time, different industries also have different impacts on carbon emissions. For example, the transportation industry and commerce have a small impact, while the construction industry and manufacturing industry have a larger impact.^16,18 (c) The influence of industrial structure on carbon emissions is different among regions, and the spatial spillover effect of high-grade industrial structure on carbon emission reduction between regions is heterogeneous.^20,21 At present, relevant scholars mainly study from the spatial perspectives of the country,^8,9,22 region,^12,23 province,^12,24 economic belt, and city.^4,14 From an international perspective, for middle and higher-development level countries, the carbon emission reduction effect brought by high-grade industrial structure is significantly higher than that of extremely high-development level countries. ²³ From the perspective of China, in provinces with reasonable economic structure and good economic foundation, industrial structure adjustment makes an obvious contribution to regional energy efficiency improvement. ²⁵ Specifically, high-grade industrial structure in eastern China has an inhibitory effect on carbon emissions, while it has an opposite effect in central and western China. ²⁶ At the same time, both high-grade industrial structure and rationalizing have significant inhibition effect on carbon productivity in the study area and adjacent areas, the inhibition effect of high-grade industrial structure is greater.^27,28

At the same time, some scholars have introduced the coupling and coordination model into the study of the relationship between industrial structure and carbon emission, mainly from the coupling and coupling coordination degree analysis. ²⁹ Most research results show that there is not only a one-way influence between industrial structure adjustment and carbon emission efficiency, but a coupling relationship of mutual influence. ³⁰ From the perspective of region, the coupling degree of China's carbon emission efficiency and high-grade industrial structure, rationalizing is low, there is obvious dynamic disharmony between them, showing a step distribution of “high in the east and low in the west,” which is basically consistent with the distribution of China's inter-regional economic development level, presents the characteristics of agglomeration distribution. ³¹ A large number of studies have proved that high-grade industrial structure is an important way to achieve low carbon emission reduction.

Carbon emission index—measurement of carbon emission efficiency

In existing studies, carbon emission efficiency is one of the main carbon emission indicators to measure the value of carbon emission. ³¹ Due to different research purposes, its connotation and measurement methods are different. Existing studies can be generally divided into two categories: Narrow carbon emission efficiency and broad carbon emission efficiency. The narrow carbon emission efficiency is mostly quantified by GDP/carbon emission, ³² while the broad carbon emission efficiency not only measures carbon emission value based on economic income but also evaluates carbon emission value based on other types of income created by carbon emission. ³³ At present, it is generally accepted by the academic community that the value of carbon emissions should be quantified by the ratio of social benefits reflecting residents’ living well-being to carbon emissions, the Human Development Index (HDI) model constructed by the United Nations Development Program (UNDP) is usually adopted to evaluate urban social benefits. In other words, urban carbon emission income can be measured from two perspectives: Economic income and social income created by carbon emission. ³³

To sum up, existing literature has conducted a gradual and in-depth discussion on the relationship between carbon emission and industrial structure, but there are two aspects worth improving or enriching: (a) Current studies on carbon emission efficiency mainly focus on the relationship between carbon emission efficiency and economic development, less on the relationship between carbon emission efficiency and the improvement of social welfare. In essence, a low-carbon economy should maximize social welfare with the minimum amount of carbon emissions. ³² Therefore, further research needs to study the coupling and coordination relationship between economic efficiency, social efficiency, and industrial structure on the basis of constructing the coupling and coordination degree model. (2) Most relevant studies are carried out from the spatial scale of national or provincial areas, but few studies are conducted from the perspective of metropolitan areas. In fact, the metropolitan area has gradually become an important geographical unit for regional participation in global or domestic economic division and competition. Taking metropolitan area as the research perspective rather than simply taking the administrative unit as the boundary is more in line with the status quo of economic and social development.

Based on this, this paper draws a comparative statistical map to analyze the spatial distribution characteristics of carbon dioxide emissions, and relies on the trend distribution map to analyze the spatial characteristics of the industrial structure of the three metropolitan areas in Jiangsu Province from the perspectives of high-grade industrial structure index, rationalizing industrial structure index and concentration degree of industrial structure index. On this basis, the degree, type, and characteristics of spatial coupling between carbon emission efficiency (carbon emission economic efficiency (CEE), carbon emission social efficiency (CSE)) industrial structure are revealed, and targeted coupling paths are designed under the framework of distribution dynamics to promote sustainable social and environmental development of metropolitan areas.

Overview of the study area

Jiangsu Province is strategically located at the umbilical cord of China's eastern coastal region, where the Belt and Road Initiative and the Yangtze River Economic Belt. With the largest manufacturing cluster in China, the per capita GDP of Jiangsu Province has consecutively ranked first in China since 2009, and it is one of the most economically active provinces in China (Figure 1). It has three metropolitan areas under its jurisdiction. The Nanjing metropolitan area is an economic zone centered on Nanjing, located in the core area of the urban zone along the middle and lower reaches of the Yangtze River, spanning Jiangsu and Anhui provinces. It is the first inter-provincial metropolitan area planned and constructed in China. The members of the Nanjing metropolitan area are Nanjing, Zhenjiang, Yangzhou, Huai ‘an, Ma'anshan, Chuzhou, Wuhu, and Xuancheng, covering a total area of 66,000 square kilometers. The Xuzhou metropolitan area is located in the north of Jiangsu Province, based on Huaihai Economic Zone, with Xuzhou as the central city. The Xuzhou metropolitan area, like Nanjing metropolitan area, is a metropolitan area spanning four provinces, covering parts of northern Jiangsu province, as well as parts of neighboring Henan, Anhui, and Shandong provinces. The Suzhou-Wuxi-Changzhou metropolitan area is located in the southern part of Jiangsu Province with the best overall development status. It is mainly composed of Suzhou, Wuxi, and Changzhou. The comprehensive development degree of Suzhou-Wuxi-Changzhou metropolitan area is the highest among the three metropolitan areas in Jiangsu Province and its urbanization rate is also the highest, and its population density, industrial density are relatively higher. Considering the development level and stage of Jiangsu Province is leading in China, this study is also exemplary and typical in China. ³⁴ (Table 1).

Figure 1.

Location of the study area.

Table 1.

Basic information of the three metropolitan areas in Jiangsu province.

Name of metropolitan area	Proportion of area in China	Proportion of population in China	Proportion of GDP in China
Nanjing metropolitan	0.6754%	2.5299%	4.0360%
Xuzhou metropolitan	0.6858%	3.2962%	1.6851%
Suzhou-Wuxi-Changzhou metropolitan	0.1825%	1.8265%	3.8836%

Note: The year of data was 2019, and the data were obtained from the statistical yearbooks of all cities in 2019.

Research methods and data sources

Total urban carbon dioxide emissions (Ce, 10,000 tons)

City main sources of carbon dioxide emissions into the carbon dioxide emissions of life and production of carbon dioxide emissions, combined with the current of carbon dioxide emissions measurement method and data acquisition, feasible degree, in this paper, the three metropolis area emissions of carbon dioxide in Jiangsu province is calculated to divide data method,^12,18 fuel variety, namely to measure different kinds of energy consumption carbon emissions, respectively, to accumulate. The calculation formula for carbon emissions is as follows：

$$\mathrm{CE}=(\mathrm{C}{\mathrm{E}}_1\times {\gamma }_1+\mathrm{C}{\mathrm{E}}_2\times {\gamma }_2+\mathrm{C}{\mathrm{E}}_3\times {\gamma }_3)\times 2.69$$

(1)where $\mathrm{CE}$ represents the total carbon dioxide emission of the city, $\mathrm{C}{\mathrm{E}}_1$ , $\mathrm{C}{\mathrm{E}}_2$ , $\mathrm{C}{\mathrm{E}}_3$ represent the annual electricity consumption (10,000 KWH), natural gas consumption (10,000 m³) and liquefied petroleum gas consumption (ton) of the city respectively, ${\gamma }_1$ , ${\gamma }_2$ , ${\gamma }_3$ respectively represent the electric standard coal coefficient (1.229 tons standard coal/10,000 KWH), natural gas standard coal coefficient (13.3 tons standard coal/10,000 m³), liquefied petroleum gas standard coal coefficient (1.7143 tons standard coal/ton). 2.69 is the amount of carbon dioxide emitted by one ton of standard coal (carbon dioxide emitted by one ton of standard coal is about 2.66–2.72 tons and the middle value is 2.69 tons in this paper). (Table 2).

Table 2.

Indicator data and sources.

Index level	Index	Units	Data source
Energy consumption	Annual urban electricity consumption	Thousands of kilowatt-hours	《China City Statistical Yearbook》
	Total natural gas consumption	10,000 m3
	Total consumption of liquefied petroleum gas	Tons
	Electric discount standard coal coefficient		Announcement on emission factors of China regional power grid base line
	Conversion coefficient of natural gas to standard coal		《Urban Greenhouse Gas Inventory Study》
	Conversion coefficient of liquefied petroleum gas to standard coal
Regional health level	Number of health care institutions		《China City Statistical Yearbook》
	Number of beds in ward		《China City Statistical Yearbook》
	Number of health technicians		《China City Statistical Yearbook》
Regional education level	Number of schools		《China City Statistical Yearbook》
	The number of students in school	10,000 people	《China City Statistical Yearbook》
	Number of teachers	10,000 people	《China City Statistical Yearbook》
Regional living standard	Total volume of post and telecommunications services	Hundred million yuan	《China City Statistical Yearbook》
	Total highway miles	Kilometer	《China City Statistical Yearbook》
	Per capita housing area	Square meter	《China City Statistical Yearbook》
Industrial structure	GDP	Hundred million yuan	《China City Statistical Yearbook》
	Value-added of the primary industry	Hundred million yuan	《China City Statistical Yearbook》
	Value-added of the secondary industry	Hundred million yuan	《China City Statistical Yearbook》
	Value-added of third industry	Hundred million yuan	《China City Statistical Yearbook》
	Primary industry workers	10,000 people	《China City Statistical Yearbook》
	Secondary industry workers	10,000 people	《China City Statistical Yearbook》
	Third industry workers	10,000 people	《China City Statistical Yearbook》

Carbon dioxide emission efficiency index measurement

According to the expected revenue types generated by urban energy consumption ⁴ and the current statistical data of Chinese cities, this study improved the measurement method of urban carbon dioxide emission efficiency. In this study, the concept of generalized carbon emission efficiency is adopted to measure the value of carbon emission from the perspectives of economic and social benefits created by carbon dioxide emission. The value is characterized by the CEE index and the CSE index.

Specifically, the economic efficiency index of carbon emissions measures the value of carbon emissions by the economic benefits obtained by cities and quantifies it by the ratio of economic benefits to the total amount of carbon dioxide emissions. In this study, the total GDP is chosen as the evaluation index of urban economic benefits. The social efficiency index of carbon emissions measures the value of carbon emissions by the social benefits obtained by residents and quantifies it by the ratio of social benefits to total carbon emissions (TCEs). Based on relevant studies at home and abroad, this paper chooses the HDI model constructed by the UNDP as the calculation model of social income, which is composed of single indicators from three dimensions: Health, education and living standard.^35–37 However, the current statistical data of small and medium-sized cities in China cannot fully meet the data requirements of the HDI model. Therefore, considering the availability and rationality of data, this paper improved the measurement method of the social efficiency of carbon emission based on the existing research according to the characteristics of small and medium-sized cities, the status quo of the data statistical system, and selected the three single indicators in the original model instead.^30,38 Specifically, the number of various medical and health institutions, the number of beds, and the number of health technicians can be used as representative indicators to evaluate the regional medical conditions, which can comprehensively reflect the regional health level. ³⁹ As indicators to measure educational resources, the number of various schools, students and teachers can comprehensively reflect the regional education level. ³⁹ The total amount of posts and telecommunications business reflects the level of public services, the total mileage of highways reflects the construction of the transportation system in small and medium-sized cities, and the per capita housing floor area of residents reflects the living conditions. The three indicators can comprehensively reflect the regional living standards. ⁴⁰ The CEE index and the CSE index are calculated according to the above formula. Before calculation, the median method was used to standardize the economic returns (Bc), social returns (Bcs), and TCEs of all the sample cities in the statistical year to eliminate the non-uniform dimensions. Secondly, the entropy method is used to calculate the weight of each social income index, eliminate the influence of human factors on the weight setting, and add all income indicators to get the total social income of each sample city. Finally, the CEE index and CSE index were calculated. ⁴¹ The CEE index is calculated using the following:

$$\mathrm{CEE}={\displaystyle \frac{B_{CE}}{\mathrm{TCE}}={\displaystyle \frac{I_{\mathrm{i}}}{I_{TCE}}}}$$

(2)where $I_{\mathrm{i}}$ is the standardized value of GDP, $I_{TCE}$ is the standardized value of carbon dioxide emission.

The carbon dioxide emission social efficiency index (CSE) is calculated using the following formula：

$$\mathrm{CSE}={\displaystyle \frac{B_{CS}}{\mathrm{TCE}}={\displaystyle \frac{\sum P_i\times I_i}{I_{TCE}}}}$$

(3)where $I_i$ is the social benefits to carbon index number for all kinds of medical institutions, medical institutions the total number of beds, health technical personnel number per one thousand people, all kinds of schools, the number of students per one thousand people, number of teachers per 1000 people, of land for public facilities accounted, every 1000 people of highway mileage, per capita housing area of the standardization of numerical, $P_i$ is the weight of each index in social income.

The research method of industrial structure

To show the spatial characteristics of the industrial structure of the three metropolitan areas in more detail, from the three aspects of optimization, rationalization, and concentration analysis.^42,43

The high-grade industrial structure. The high-grade industrial structure represents the process in which the center of gravity of a country or region's industrial structure shifts from the primary industry to the secondary industry and the third industry. ⁴⁴ It marks the stage and direction of the economic development level of a country or region within a certain period of time. ⁴⁵ Based on the theory that the proportion vector of three industries and the angle of the corresponding coordinate axes will change with the proportion of industries, this paper calculates the advanced index of industrial structure ⁴⁵ :

$$IH={\theta }_1+{\theta }_2$$

(4)

$${\theta }_1=\pi -{\mu }_1-{\mu }_2$$

(5)

$${\theta }_2={\displaystyle \frac{\pi }{2}-{\beta }_1}$$

(6)where ${\theta }_1$ is the transfer effect of the primary industry to the secondary and third industries. ${\theta }_2$ is the transfer effect from the secondary industry to the third industry. ${\mu }_1,{\mu }_2$ are the angles between vector (x₁, x₂, x₃) and vector (0, 1, 0),(0, 0, 1), x₁,x₂,x₃ represent the proportion of added value of primary, secondary and third industries in GDP. ${\beta }_1$ is the angle between vector (x₂, x₃) and vector (0, 1). The higher the values of ${\theta }_1$ , ${\theta }_2$ the higher the level of industrial transfer. The higher the value of IH, the higher the level of advanced.

The rationalizing industrial structure.The rationalization of industrial structure means that in order to improve economic benefits, it requires adjusting the initially unreasonable industrial structure according to the level of science and technology, the structure of consumption demand, the basic quality of population, and resource conditions in a certain stage of economic development, so as to realize the reasonable allocation of production factors and make the coordinated development of various industries. The rationalizing industrial structure is the aggregation quality between industries, which reflects the coordination degree between industries on the one hand, and the effective utilization degree of resources on the other hand, that is to say, it is a measure of the coupling degree of factor input structure and output structure energy efficiency and industrial structure deconstruction, spatial distribution and coupling analysis under the constraint of carbon emissions. ⁴⁶ Theil index TL is used for analysis in this paper, and the calculation formula：

$$\mathrm{TL}={\sum }_1^K({\displaystyle \frac{Y_i}{Y})}\ln ({\displaystyle \frac{Y_{\mathrm{i}}}{L_{\mathrm{i}}}}/{\displaystyle \frac{Y}{L}})$$

(7)where $Y_{\mathrm{i}}$ , Y represent the three industries and the total added value respectively. $L_{\mathrm{i}}$ , L are respectively the three industries and the total employment number. The higher the value of TL is, the more unreasonable the allocation of production factors is, and the lower the level of rationalization is.

The concentration of industrial structure.The spatial allocation of resources is usually realized according to the industrial layout. China has a vast territory, and the differences in social and economic development among different regions lead to the differences in industrial layout of different regions. At the same time, the difference in regional economic development leads to the phenomenon of industrial agglomeration and structural convergence. Within the regional scope, the greater the economic output gap of different industries, the higher the degree of industrial concentration, and vice versa. Centralization index is an index describing the degree of centralization of geographic data distribution. Similarly, the industrial concentration index is an important quantitative index used to analyze and measure the degree of specialization (or concentration) of industry or economic sector in a region. ⁴⁷ The calculation formula is:

$$I=(A-P)/(M-P)$$

(8)where $A,P,M$ , respectively represent the cumulative percentage sum of the actual data, uniform distribution, and centralized distribution of the three industries. The higher the value of I, the higher the level of concentration.

Evaluation model of coupling coordination degree

Coupling degree is a physical concept, which is usually used to describe the degree of interaction and mutual influence between systems. There is an interaction between industrial structure adjustment and carbon emission efficiency, showing that carbon emission efficiency has a constraint effect on industrial structure adjustment, and carbon emission efficiency is under the stress of industrial structure adjustment. ⁴⁸ Therefore, the principle of the capacity coupling and capacity coupling coefficient model in physics can be used for reference to analyze the coupling degree of the two. On the one hand, industrial structure adjustment is constrained by carbon emission efficiency. On the other hand, the improvement of carbon emission efficiency is stressed by industrial structure adjustment, and there is an interactive coupling relationship between them. ⁴⁹ In this paper, the coupling degree (C) model in physics is used to evaluate the coupling relationship between carbon emission efficiency and industrial structure,^50,51which can be expressed as:

$$C={\left\{{\displaystyle \frac{EMX(\mathrm{x})I_{\mathrm{i}}(\mathrm{y})}{{\left[{\displaystyle \frac{EMX(\mathrm{x})+I_{\mathrm{i}}(\mathrm{y})}{2}}\right]}^2}}\right\}}^K$$

(9)where $EMX(\mathrm{x})$ represents the efficiency of the carbon composite index, efficiency to carbon value. $I_{\mathrm{i}}(\mathrm{y})$ represents the composite index of industrial structure, and high-level respectively, rationalization index and concentration values and K is the discrimination coefficient, 2 ≤ K ≤ 5 and 4 is taken in this paper. ⁵²

Because the coupling degree can only express the strength of the correlation between industrial structure and carbon emission efficiency, it cannot reflect the comprehensive effect and “synergistic effect” between industrial structure and carbon emission efficiency. To better evaluate the degree of interaction and coupling consistency between industrial structure and carbon emission efficiency, this paper uses the coordination degree (R) index to reflect the overall interaction and “synergistic effect” between carbon emission efficiency and industrial structure. The higher the degree of coupling consistency, the stronger the overall correlation between industrial structure and carbon emission efficiency, and the stronger the overall development consistency. ⁵¹ The calculation formula is:

$$R=(C\times T{)}^{\frac{1}{2}}$$

(10)Values all belong to [0,1], In order to distinguish the spatial distribution of coordination degree between carbon emission efficiency and industrial structure, the following definitions are made：①When R∈[0,0.5), it is at a low coordination level, ②R∈[0.5,0.75], it is at a medium coordination level, ③R∈(0.75,1], it is at a high coordination level.

$$T=a\times EMX(x)+b\times I_i(y)$$

(11)where T is the comprehensive coordination index of carbon emission efficiency and industrial structure, a and b are the weights to be determined. Since carbon emission efficiency and industrial structure are equally important, a  =  b  =  0.5. To facilitate data analysis, before the coupling degree and coordination degree calculation, the original data need to be standardized accordingly. This paper mainly selects the min-max standardization method. The smaller the value of the rationalization of the industrial structure, the more favorable it is. Therefore, the negative index formula is selected for calculation, and the original values of other data are calculated by the positive index formula. ⁴⁸ The calculation formula is：

$$Z_{ij}={\displaystyle \frac{C_{ij}-\min \{C_j\}}{\max \{C_j\}-\min \{C_j\}}\mathrm{Positive}\mathrm{indicators}}$$

(12)

$$Z_{ij}={\displaystyle \frac{\max \{C_j\}-\{C_{ij}\}}{\max \{C_j\}-\min \{C_j\}}\mathrm{Negative}\mathrm{indicators}}$$

(13)where $C_{ij},Z_{ij}$ , respectively represent the original value of the jth index in the ith year and the value after standardized treatment, $\max \{C_j\},\min \{C_j\}$ are the maximum and minimum values of the Cth index, respectively. The higher the value of the coupling degree, the more harmonious the coupling relationship between carbon emission efficiency and industrial structure. The larger the coordination degree is, the better the coupling development is (Table 3).

Table 3.

The reference standard of coupling and coordinating stage.

Coupling degree	Coupling phase	Coordination degree	Coordinate stage	Integrated coupling stage
0 < C ≤ 0.3	Separation phase	0 < R ≤ 0.3	Low-coordination stage	Low-coordination separation stage
0.3 < C ≤ 0.5	Antagonism phase	0.3 < R ≤ 0.5	Middle coordination stage	Middle coordination antagonism stage
0.5 < C ≤ 0.8	Running-in stage	0.5 < R ≤ 0.8	High coordination stage	High coordination run-in stage
0.8 < C ≤ 1	Coupling phase	0.8 < R ≤ 1	Polar coordination stage	Polar coordination coupling stage

Markov chain model

In the traditional distributed dynamic model, Markov chain method usually only examines the situation where the time length (short for time) of a step is 1 a, while in economic and social development, since the formulation, release, and implementation of policies are often out of sync, the change of state is often lag, so it is not representative to only consider the characteristics of state transition under the short variable time. In this paper, the traditional model is extended, that is, a multi-year transfer probability matrix is constructed to investigate the transfer changes of regional carbon emission efficiency (industrial structure) over time, and to dig out more solidification information. Specifically, the construction method of multi-year time Markov transition probability matrix is as follows ⁵³ :

The probability value of the year-long Markov transition probability matrix is: $P_{ij}^{t,t+d}=\{X_{t+d}=j|X_t=i\}$ . It represents the probability that a region with t level of i type in i year belongs to j type after d year. The probability is estimated as follows ⁵³ :

$$P_{ij}^d={\sum }_{t=t_0}^{T-d}{n}_{ij}^{t,t+d}/{\sum }_{t=t_0}^{T-d}{n}_t^t$$

(14)

where represents the sum of all the cities that belong to i type in tyear and transfer to j type in $t+d$ years during the whole study period, and n represents the total number of cities whose carbon emission efficiency (industrial structure) belongs to i type in t year. Different types of transition probabilities are estimated respectively, and then d year long Markov transition probability matrix is obtained ⁵³ :

$$\left[\begin{array}{ccccc}{\displaystyle \frac{n_{11}^d}{n_{1.}^d}}& \cdots & {\displaystyle \frac{n_{1j}^d}{n_{1.}^d}}& \cdots & {\displaystyle \frac{n_{1k}^d}{n_{1.}^d}}\\ {\displaystyle \frac{n_{21}^d}{n_2^d}}& \cdots & {\displaystyle \frac{n_{2j}^d}{n_{2.}^d}}& \cdots & {\displaystyle \frac{n_{2k}^d}{n_2^d}}\\ \cdots & \cdots \cdots & \cdots & & \cdots \\ {\displaystyle \frac{n_{k1}^d}{n_k^d}}& \cdots & {\displaystyle \frac{n_{kj}^d}{n_{k.}^d}}& \cdots & {\displaystyle \frac{n_{kk}^d}{n_{k.}^d}}\end{array}\right]=\left[\begin{array}{ccc}{p}_{11}^d\cdots p_{1j}^d& \cdots & p_{1k}^d\\ p_{21}^d\cdots p_{2j}^d& \cdots & p_{2k}^d\\ \cdots \cdots \cdots & \cdots & \cdots \\ p_{k1}^d\cdots p_{k2}^d& \cdots & p_{kk}^d\end{array}\right]$$

(15)where the scale of class i horizontal region is $n_i^d$ , is $\sum _{t=t_0}^{T-d}{n}_i^t$ ; $P_{ii}^d$ represents the transfer probability of i type regions remaining i type after d year, and the larger the value, the higher the degree of regional differences in carbon emission efficiency (industrial structure) is solidified, which also indicates the existence of club convergence.

Measures of regional solidification degree-Club convergence index. In order to accurately measure the solidification degree of carbon emission efficiency (industrial structure), the club convergence index is further constructed. The index takes into account both the size of different types of regions (clubs) and the degree of convergence within each club. Specifically, the formula for calculating the club convergence index of the year is as follows ⁵³ :

$$\mathrm{CC}{\mathrm{L}}^d=p_{11}^d\times {\displaystyle \frac{n_1^d}{\sum n_i^d}+p_{22}^d\times {\displaystyle \frac{n_2^d}{\sum n_i^d}+\cdots +p_{kk}^d\times {\displaystyle \frac{n_k^d}{\sum n_i^d}}}}$$

(16)where p is the diagonal element in the formula, which represents the convergence degree of k-type clubs in d years, and $n_{kg}^d/\sum n_{ig}^d$ represents the scale proportion of k-type clubs.

In order to compare the curing degree of different indexes, the transfer probability matrix needs to be tested statistically. In this paper, likelihood ratio statistics are constructed in the following way ⁵³ :

$$Q=-2\log \left\{\prod _{l=1}^2\prod _{i=1}^k\prod _{j=1}^k{\left[{\displaystyle \frac{P_{ij}^d}{P_{ij}^d(l)}}\right]}^{n_{ij}(l)}\right\}$$

(17)where $P_{ij}^d(l)$ and $n_{ij}(l)$ , respectively represent the number of elements and corresponding cities in the transfer probability matrix under the condition that the tested duration is d years; k represents the number of horizontal types divided; $P_{ij}^d$ represents the transfer probability value obtained after data combination; the statistic Q obeys a chi-square distribution, and its degree of freedom is k × (k-1) minus the sum of all transfer probabilities with values of zero.

Results and analysis

Characteristics and spatial evolution pattern of carbon dioxide emissions in the three metropolitan areas

To show and analyze the spatial pattern and evolution process of carbon dioxide emissions in the three metropolitan areas of Jiangsu Province, the Cartogram ⁵⁴ was used to show the carbon dioxide emissions in the main years from 2009 to 2019 (Figure 2). On the premise that the topological relationship of the region in the map and the total area of the whole study area remain unchanged. The comparative statistical map can strongly express the observed value information by using the distorted changes of the sub-regional area (the size of the sub-regional area is proportional to the size of the attribute value). It can be seen that the carbon dioxide emission of the three metropolitan areas in Jiangsu Province generally presents an obvious distribution pattern of high carbon dioxide emission from the southeast to the northwest, and low carbon dioxide emission in the middle. As time goes by, the center of gravity of this pattern shifts more and more to the southeast. The inner part of the metropolitan area shows the spatial distribution pattern of center-periphery, that is, the high-emission area is centered and surrounded by the low-emission area. The carbon dioxide emissions of the “three centers” in 2009 and 2019 accounted for about 32.97% and 38.69%, separately, while the carbon dioxide emissions of the “five centers” in 2009 and 2019 accounted for about 51.41% and 55.22%, separately. The proportion of “three centers” and “five centers” in total emissions increased, and the carbon dioxide emissions in Jiangsu metropolitan areas showed a strong central-multi-point divergent growth.

Figure 2.

Comparative statistical map of carbon dioxide emissions in three metropolitan areas of Jiangsu Province from 2009 to 2019 ((a) standard maps of administrative divisions, (b) distorted map of carbon dioxide emissions).

During the study period, the proportion of the third industry in Suzhou, Wuxi, Xuzhou and Jining successively exceeded the proportion of secondary industry (Nanjing had achieved this in 2008), it was gradually strengthened as a regional leader. Although the third industry is characterized by low carbon emissions, the increase of its economic aggregate will promote the increase of carbon dioxide emissions. ⁴ At the same time, although the proportion of the secondary sector in these cities declined year by year, it still accounted for half of the majority of cities. For example, the proportion of secondary industry in Suzhou and Wuxi in 2019 was still close to 50%, and in Changzhou reached 50.2%, the energy consumption of secondary industry still accounted for a large proportion. From this point of view, the continuous high-energy consumption of the secondary industry and the growth of the economic aggregate of the third industry in the “five centers” maintain a high regional carbon dioxide emission.

Distribution characteristics of carbon emission efficiency

Using the natural break point method, the carbon emission efficiency is divided into five categories (Figure 3). The evolution chart of carbon emission efficiency from 2009 to 2019 shows the dynamic changes in the economic efficiency of carbon emission (hereinafter referred to as CEE) and social efficiency of carbon emission (hereinafter referred to as CSE) of each city in the three metropolitan areas of Jiangsu Province in selected years. The study found that from 2009 to 2019, the average CEE of 19 cities in the three metropolitan areas of Jiangsu Province continued to increase, indicating that the contribution rate of the same amount of carbon emissions to economic income gradually increased, and the development status of low-carbon economy in the three metropolitan areas was improving. In addition, CSE values in all 19 cities showed an increasing trend, although the degree of CSE increase varied greatly between cities. Comparing the changes of the two types of efficiency in the same metropolitan area, it is found that the growth range of CEE index is greater than that of the CSE index within the study scope, the trend distribution map (drawn by ArcGIS) shows that the high point of CEE gradually concentrates in the eastern and southern regions, while it is precisely the low point of CSE. These findings indicate, to some extent, that the improvement effect of carbon emission on the regional economic development level is more significant than the improvement of public service level and residents’ living quality. (Figure 4).

Figure 3.

Carbon emission efficiency of three metropolitan areas in Jiangsu Province: CEE (a) and CSE (b) from 2009 to 2019. CEE: carbon emission economic efficiency; CSE: carbon emission social efficiency.

Figure 4.

Trend distribution of carbon emission efficiency of three metropolitan areas in Jiangsu Province: CEE (a) and CSE (b) from 2009 to 2019. CEE: carbon emission economic efficiency; CSE: carbon emission social efficiency.

Distribution characteristics and spatial pattern evolution of industrial structure

From the numerical distribution. In terms of the high-grade industrial structure (Figure 5-(a)), the high-grade industrial structure index of the Nanjing metropolitan area showed an upward trend during the whole study period, and the overall difference was decreasing (coefficients of variations were: 7.20%, 7.41%, and 4.62%, respectively). Among them, Nanjing's high-grade industrial structure index is above 2.4, which belongs to the developed area of industrial structure. The high-grade industrial structure indexes of Ma'anshan, Wuhu, Chuzhou, and Xuancheng located in the western area of the Nanjing metropolitan area are all lower than the average level of the Nanjing metropolitan area (2.132, 2.217, 2.367 respectively from 2009 to 2019), which belong to backward areas of the industrial structure. The possible reasons are the strong foundation of urban agriculture and the large proportion of the labor-intensive manufacturing industry. The other three cities (Zhenjiang, Yangzhou, Huai'an) are all higher than the average level of the metropolitan area, and belong to the advanced development of industrial structure, the high-grade industrial structure index of the Xuzhou metropolitan area showed an upward trend, and the overall gap narrowed sharply (coefficient of variations were 5.83%, 4.00%, 1.89%). Among them, Xuzhou, Suzhou, Huaibei, Jining, and Suqian are all higher than the average level of the Xuzhou metropolitan area, and belong to the advanced development area of industrial structure, which is at the same level as the same type of cities in the Nanjing metropolitan area. The high-grade industrial structure indexes of Shangqiu and Zaozhuang are all lower than the average level of the Xuzhou metropolitan area, which belongs to backward areas of the industrial structure, it is at the same level as the same type of cities in the Nanjing metropolitan area. This is partly due to its relative location in the west. China's reform and opening up began in the coastal areas and then extended to the inland, the industrial changes also showed a similar law. The high-grade industrial structure index of Suzhou-Wuxi-Changzhou metropolitan area showed an overall upward trend, with a small overall gap and trend of narrowing (coefficient of variations were 0.610%, 0.042%, and 0.037%, respectively), which belonged to the area of advanced industrial structure development. This has a significant relationship with the industrial direction of the three cities aimed at strategic emerging industries and service industries.

Figure 5.

Spatial distribution changes of industrial structure in the three metropolitan areas of Jiangsu Province from 2009 to 2019 ((a) high-grade industrial structure, (b) rationalizing industrial structure, (c) concentration of industrial structure).

Industrial structure rationalization aspects (Figure 5(b)), the entire study period of the Nanjing metropolitan area and the Xuzhou metropolitan area industrial structure rationalization indexes average presents a downward trend after rising first, to some extent, indicates that the Nanjing metropolitan area and the Xuzhou metropolitan area industrial structure and balanced state deviation degree decreased, began to tend to a reasonable industrial structure. The variation within the Nanjing metropolitan area was small (coefficient of variations were 28.27%, 24.43%, 30.78%), while the variation within the Xuzhou metropolitan area was large (coefficient of variations were 52.05%, 54.69%, 70.37%). The reason may be that the industrial structure rationalization index of Zaozhuang keeps increasing (0.21, 0.23, 0.35), which is much higher than the average level of Xuzhou metropolitan area (0.15, 0.19, 0.15), which belongs to the area with a low reasonable degree of industrial structure. This is because Zaozhuang industrial structure is over-dependent on the traditional coal and chemical industries. The average level of the industrial structure rationalization index in Suzhou-Wuxi-Changzhou metropolitan area has been in a declining stage, and the variation of the regional difference has been sharply reduced (coefficient of variations were 45.02%, 19.89%, 18.40%).

In terms of concentration of industrial structure (Figure 5(c)), the average level of industrial structure concentration index in the three metropolitan areas of Jiangsu Province showed an upward trend (the average level of the Nanjing metropolitan area was 0.26, 0.28, 0.42. The average level of the Xuzhou metropolitan area was 0.18, 0.20, 0.37. The average level of Suzhou-Wuxi-Changzhou metropolitan area was 0.39, 0.42, and 0.50) inter-regional disparities are also narrowing (The coefficient of variation in the Nanjing metropolitan area was 44.96%, 46.12%, 19.71%. The coefficient of variation in the Xuzhou metropolitan area was 50.17%, 47.64%, 11.85%. The coefficient of variation in the Suzhou-Wuxi-Changzhou metropolitan area was 36.57%, 26.32%, 25.92%).

From the perspective of spatial distribution:

In 2009, the high-grade industrial structure of the three metropolitan areas in Jiangsu Province showed the distribution characteristics of a “C-shaped mirror structure,” with Jining, Xuzhou, Suqian, Huai'an, Nanjing, Zhenjiang and Changzhou as the advanced developed areas, while the advanced backward areas were mainly concentrated in the western region. In 2019, the high-grade industrial structure of the three metropolitan areas showed the distribution characteristics of a “northwest to southeast” spatial structure with Xuzhou, Nanjing, Suzhou, Wuxi, and Changzhou as the core. Compared with the distribution characteristics of high-grade industrial structure of the three metropolitan areas in 2009, the overall gap was narrowing (Figure 6).

From 2009 to 2019, the rationalizing industrial structure showed the distribution characteristics of the “three-step ladder” from the northwest (high rationalization) to southeast (low rationalization), high-value areas are concentrated in the east and south (Figure 6) and the deviation degree between industrial structure and equilibrium state narrowed in space. Industrial structure not only represents the state of economic development in each region but also presents the corresponding resource and environment basis. ²⁴ The rationalizing industrial structure represents the rational utilization of resources, and the differences in the rationalization of industrial structure in various metropolitan areas also reflect the reasonable utilization of energy resources to a certain extent. ¹⁶

Generally, the concentration of the industrial structure shows the distribution characteristics of a “C” shape mirroring structure with Xuzhou-Yangzhou-Zhenjiang-Changzhou-Nanjing as the high concentration area, which is similar to the distribution characteristics of the optimization of industrial structure. Similarly, areas with a low concentration of industrial structure are also concentrated in the western region (Figure 6).

Figure 6.

Trend distribution of industrial structure in the three metropolitan areas of Jiangsu Province from 2009 to 2019 ((a) the high-grade industrial structure; (b) rationalizing industrial structure, (c) concentration of industrial structure).

Coupling and coordination analysis of carbon emission efficiency and industrial structure optimization

The coupling and coordination degree evaluation model was used to calculate the coupling coordination of carbon emission efficiency (CEE, CSE) and industrial structure (optimization, rationalization, concentration) in the three metropolitan areas of Jiangsu Province (Figure 5). On the whole, the coupling and coordination degree of CEE-industrial structure in the three metropolitan areas are higher than that of CSE-industrial structure.

Carbon emission efficiency—high-grade industrial structure coupling, coordination degree

On the whole, the coupling coordination distribution of the high-grade industrial structure and CEE in Xuzhou metropolitan area is significantly better than the other industrial structure indicators, with slightly higher coupling points. In addition, from the coupling values corresponding to each coupling point, the mean values of the coupling and coordination degree of the high-grade industrial structure with CEE and CSE are also higher than the industrial structure rationalization and concentration of industrial structure. This indicates that the optimization level of industrial structure in the Xuzhou metropolitan area is closely related to the improvement of CEE and CSE, both of which are in the stage of coordination antagonism. The high-grade industrial structure is the key link affecting the improvement of carbon emission efficiency, the implementation of energy conservation, and emission reduction policies in the Xuzhou metropolitan area.

From the perspective of time evolution, during the study period, the coordination degree of CEE-high-grade industrial structure in the three metropolitan areas of Jiangsu Province from 2009 to 2019 remained at 0.14–0.83, 0.110.87,and 0.11-0.95, respectively (Figure 7(a1)). The coordination degree of CSE—high-grade industrial structures is maintained at 0.01–0.82, 0.1–0.87, and 0.006–0.94, respectively (Figure 7(a2)), which covers the four stages from low coordination stage to polar coordination stage, with big differences. From a regional perspective, the CEE—high-grade industrial structure coupling coordination in the Nanjing metropolitan area takes Nanjing as the dividing line, showing a spatial distribution pattern of “high coordination coupling” in the east and “low coordination separation” in the west. Among them, the coupling and coordination degree of Nanjing CEE—high-grade industrial structures are in the rising stage, from the high coordination run-in stage in the early stage of the study to the polar coordination coupling stage. The coupling and coordination degree of Zhenjiang, Yangzhou, and Huai'an, located in the east of Nanjing, are in a downward trend, which seems to be contrary to the rising trend of carbon emission efficiency and high-grade industrial structure in the three cities during the study period, reflecting the declining effect of optimization of the industrial structure in the three cities on carbon emission reduction.

Figure 7.

Coupling and coordination distribution of carbon emission efficiency and industrial structure in three metropolitan areas of Jiangsu province. (a1) CEE-the high-grade industrial structure coupling degree, coordination degree. (a2) CSE-the high-grade industrial structure coupling degree, coordination degree. (b1) CEE-industrial structure rationalization coupling degree, coordination degree. (b2) CSE-industrial structure rationalization coupling degree, coordination degree. (c1) CEE-concentration of industrial structure coupling degree, coordination degree. (c2) CSE-concentration of industrial structure coupling degree, coordination degree. (I) Nanjing metropolitan. (II) Xuzhou metropolitan. (III) Suzhou-Wuxi-Changzhou metropolitan from left to right, each group of points represents 2009, 2014, and 2019, respectively). CEE: carbon emission economic efficiency; CSE: carbon emission social efficiency.

Xuzhou-the heart city of the Xuzhou metropolitan, CEE- high-grade industrial structure on the stage of high coordination coupling, the coordination degree has increased, the rest of the city shows that the polarization, Lianyungang, Shangqiu coupling, coordination degree increased, both from low coordinate separation phase into the middle coordination coupling stage, Suqian by middle coordination separation stage entered the stage of high coordination coupling stage. The coordination degree of Suzhou decreased from the high coordination stage to the middle coordination stage, the coupling and coordination degree of Huaibei and Jining showed a downward trend, respectively, from the middle coordination running-in stage, the high coordination coupling stage to the middle coordination separation stage and the middle coordination antagonism stage. Suzhou decreased from the high coordination stage to the middle coordination stage, and coupling and coordination degree of Huaibei and Jining showed a downward trend, respectively, from the middle coordination running-in stage, the high coordination coupling stage to the middle coordination separation stage, and the middle coordination antagonism stage.

Similar trends in the change of CEE—high-grade industrial structure coupling and coordination in the three cities of the Suzhou-Wuxi-Changzhou metropolitan area, which rapidly declined from the high coordination run-in stage and the polar coordinated/high coordinated coupling stage to the high coordinated antagonistic stage and high coordination run-in stage (Changzhou), respectively.

The spatial distribution pattern of CSE—high-grade industrial structure coupling coordination is similar to that of CEE—high-grade industrial structure coupling coordination. Among them, the coordination degree of the Nanjing metropolitan area and the Xuzhou metropolitan area is generally on the rise, although the increase in the degree of CSE among cities in the metropolitan area is very different, while the coordination degree of the Suzhou-Wuxi-Changzhou metropolitan area shows a downward trend. The high-grade industrial structure and CSE are both in the rising stage. The deviation of the two trends reflects that the optimization of the industrial structure has not brought the simultaneous growth of social welfare to a certain extent.

Carbon emission efficiency-industrial structure rationalization coupling, coordination degree

Generally, the coupling coordination distribution between CEE—industrial structure rationalization in the Nanjing metropolitan area is significantly better than the other two industrial structure indicators, the coupling points are evenly distributed and there are slightly higher coupling points. In addition, according to the corresponding coupling values of each coupling point, the mean values of the coupling and coordination degree of CEE- industrial structure rationalizations are also higher than those of optimization of the industrial structure and concentration of industrial structure. This indicates that the rationalization of industrial structure in the Nanjing metropolitan area is closely related to the improvement of CEE and is in the stage of high coordination run-in. The industrial structure rationalization is the key link affecting the improvement of carbon emission efficiency, the implementation of energy conservation, and emission reduction policies in the Nanjing metropolitan area.

From the perspective of time evolution, during the study period, the coordination degree of CEE—industrial structure rationalization in the three metropolitan areas of Jiangsu Province from 2009 to 2019 remained at 0.29–0.90, 0.11–0.88, and 0.10–0.82, respectively (Figure 7(b1)), covering four stages from low coordination stage to polar coordination stage, with large differences. The coordination degree of CSE-industrial structure rationalization remained at 0.11–0.68, 0.15–0.61, and 0.03–0.68, respectively (Figure 7(b2)), covering the three stages from low coordination stage to high coordination stage, with slightly smaller differences. From a regional perspective, the coupling degree of CEE-industrial structure rationalization in Nanjing is in the first rapid, then slow decline stage, the coordination degree is in the first small decline, then rapid rise stage and enters the high coordination run-in stage from the middle coordination coupling stage in the early stage of the study. During the study period, Zhenjiang, Yangzhou, Huai'an, Ma'anshan, and Wuhu were all at or above the high coordination run-in stage, reflecting that the rationalization of industrial structure in the five cities had a high impact on the improvement of CEE. Although the coupling degree of CEE-industrial structure rationalization in Chuzhou and Xuancheng is in the rising stage, the coordination degree is in the declining stage. The opposite trend of the two shows that although the correlation of CEE-industrial structure rationalization in Chuzhou and Xuancheng is strengthened, the overall interaction and “synergistic effect” are in the weakening stage. How to enhance the overall interaction and “synergistic effects” should be the exact problem to be solved in both places.

Xuzhou metropolitan CEE-industrial structure rationalization coupling overall weak in the Nanjing metropolitan, among them, Xuzhou CEE-industrial structure rationalization in a polar coordination coupling phase shifts to high coordination coupling phase, coordination degree decline, Lianyungang, Suqian coupling rose, coordination degree down, both from high coordinate separation phase into middle coordination coupling phase. The degree of coupling and coordination degree in Suzhou decreased from the stage of middle coordination antagonism to the stage of low coordination separation, while the degree of coupling in Huaibei increased from the early stage of the study, and the degree of coordination decreased from the stage of middle coordination separation to the stage of low coordination antagonism. Shangqiu coupling and coordination degree is in a “V” type trend, but overall in the low level (in middle coordination separation phase), middle coordination of Zaozhuang rises to high coordination phase, a high level in the circles of Xuzhou city.

The CEE-industrial structure rationalization coupling and coordination trend of the three cities in Suzhou–Wuxi–Changzhou metropolitan area is quite different. The coupling and coordination degree of Suzhou is in the declining stage, from the stage of high coordination coupling to the stage of low coordination antagonism. The coupling and coordination degree of Wuxi is in the V-shaped trend and the coordination degree is much higher than the coupling degree. In Changzhou, the coordination degree rises, and the coupling degree drop from the coupling stage to the run-in stage.

The spatial distribution patterns of CSE-industrial structure rationalization coupling, coordination, and CEE-industrial structure rationalization coupling, and coordination are different, and the spatial distribution shows obvious proximity. Among them, the coordination degree of Nanjing, Zhenjiang, and Yangzhou in the east of the Nanjing metropolitan area has increased, and the coordination and coupling degree of other cities is generally in the “inverted V-shaped” decline or rapid decline. The change trend of Lianyungang and Suqian in the eastern part of Xuzhou metropolitan area is consistent, from the middle coordination running-in stage to the middle/low coordination separation stage. Suzhou and Huaibei in the south of Xuzhou are both in the middle coordination run-in stage, while Zaozhuang and Jining in the north of Xuzhou have similar change trends, the coupling, coordination degree of the former is stronger than the latter and the coupling, coordination degree of Shangqiu City is in an “inverted V” downward trend, at a low stage level (low coordination separation stage). The variation trend of the coupling degree in Suzhou and Wuxi is similar, while the coordination degree in Wuxi decreases first and then increases slightly.

Carbon emission efficiency-concentration of industrial structure coupling, coordination degree

In addition, from the coupling values corresponding to each coupling point, the mean values of the coupling and coordination degree of concentration of industrial structure with CEE and CSE are also higher than those of optimization of the industrial structure and industrial structure rationalization. This indicates that the concentration of industrial structure is closely related to the improvement of CEE in Suzhou-Wuxi-Changzhou metropolitan area, which is in a polar coordination coupling stage. The concentration of industrial structure is the key link affecting the improvement of carbon emission efficiency, implementation of energy conservation, and emission reduction policies in Suzhou-Wuxi-Changzhou metropolitan area.

From the perspective of time evolution, during the study period, the coordination degree of CEE-concentration of industrial structure in the three metropolitan areas of Jiangsu Province from 2009 to 2019 remained at 0.15–0.92, 0.20–0.93, and 0.12–0.86, respectively (Figure 7-c1). The coordination degree of CSE-concentration of industrial structure remained at 0.12–0.91, 0.10–0.71, and 0.13–0.92, respectively (Figure 7(c2)), which all covered the four stages from low coordination stage to polar coordination stage, with great differences. The spatial distribution pattern of CEE-concentration of industrial structure coordination degree in Nanjing metropolitan area and Xuzhou metropolitan area is similar to the spatial distribution pattern of CEE-industrial structure rationalization. The coordination degree of CEE-industrial structure concentration in Xuzhou metropolitan area is weaker than that in Nanjing metropolitan area.

The spatial distribution patterns of CSE-industrial structure concentration coupling, coordination, and CEE-industrial structure concentration coupling, and coordination are different. Among them, the coordination degree of Nanjing, Zhenjiang, and Yangzhou in the east of Nanjing metropolitan area has increased, and coupling and coordination degree of Huai'an, Ma'anshan, and Xuancheng are generally in an “inverted V-shaped” decline or rapid decline stage. The trend of Lianyungang and Suqian in the eastern part of Xuzhou metropolitan area is consistent, which decreases from low/middle coordinated separation/run-in stage to be low/middle coordinated separation/antagonism stage. The variation trend of Zaozhuang and Jining in the north of Xuzhou is similar, both of which are “inverted V”, and the coupling coordination degree of Shangqiu and Huaibei is in the “V” upward trend. The variation trend of coupling degree in Suzhou, Wuxi, and Changzhou metropolitan areas is not consistent. Suzhou and Changzhou are in the rising stage, while Wuxi's coupling degree shows a declining trend.

Coupling path of carbon emission efficiency and industrial structure in Jiangsu metropolitan area

In this paper, coupling coordination between inter-regional carbon emission efficiency and industrial structure is investigated from the perspective of internal dynamics of distribution. ⁵³ The club convergence index constructed on the framework of distribution dynamics is used to measure and compare the difference between the high and low-level solidification degree of the two in a long time. Considering that the implementation of regional coordination policy requires a certain amount of time, the club convergence index under different time accumulation is calculated, and considering that different discrimination methods may obtain different dynamic distribution results, the club convergence index in this paper is calculated on the basis of mean and median discrimination, so as to make the conclusion more robust. The discrimination method divided regional divisions into four types (clubs) by taking 50%, 100%, and 150% of the mean or median of each year as the demarcation points. ⁵³ The larger the value is, the higher the degree of solidification will be. The government's adjustment of inter-regional intensity is not enough, and the adjustment intensity should be strengthened to achieve inter-regional coordination. ⁵³ Considering that the sample size in a single metropolitan area is too small, which will cause great interference with the scientific results, the club convergence study is conducted from the perspective of the three metropolitan areas as a whole. The results are shown in Table 4.

Table 4.

Club convergence index of carbon efficiency and industrial structure for different durations.

Time	Mean			Median
	CEE	High-grade industrial structure	Difference duration	CEE	High-grade industrial structure	Difference duration
k  =  1	0.911	0.860	0.051	0.881	0.776	0.105
k  =  2	0.892	0.783	0.109	0.831	0.695	0.136
k  =  3	0.900	0.697	0.203	0.867	0.671	0.196
k  =  4	0.803	0.652	0.151	0.811	0.627	0.184
k  =  5	0.733	0.582	0.151	0.761	0.584	0.177
Mean value	0.848	0.715	0.133	0.830	0.671	0.160
	CEE	Industrial structure rationalization	Difference duration	CEE	Industrial structure rationalization	Difference duration
k  =  1	0.911	0.842	0.069	0.881	0.810	0.071
k  =  2	0.892	0.751	0.141	0.831	0.743	0.088
k  =  3	0.900	0.716	0.184	0.867	0.683	0.184
k  =  4	0.803	0.620	0.183	0.811	0.616	0.195
k  =  5	0.733	0.628	0.105	0.761	0.621	0.140
Mean value	0.848	0.711	0.136	0.830	0.695	0.136
	CEE	Concentration of industrial structure	Difference duration	CEE	Concentration of industrial structure	Difference duration
k  =  1	0.911	0.847	0.064	0.881	0.826	0.055
k  =  2	0.892	0.770	0.122	0.831	0.756	0.075
k  =  3	0.900	0.734	0.166	0.867	0.687	0.180
k  =  4	0.803	0.685	0.118	0.811	0.644	0.167
k  =  5	0.733	0.655	0.078	0.761	0.621	0.140
Mean value	0.848	0.738	0.110	0.830	0.707	0.123
	CSE	High-grade industrial structure	Difference duration	CSE	High-grade industrial structure	Difference duration
k  =  1	0.939	0.842	0.097	0.894	0.810	0.084
k  =  2	0.914	0.751	0.163	0.887	0.743	0.144
k  =  3	0.896	0.716	0.180	0.844	0.683	0.161
k  =  4	0.863	0.620	0.243	0.814	0.616	0.198
k  =  5	0.828	0.628	0.200	0.785	0.621	0.164
Mean value	0.888	0.711	0.177	0.845	0.695	0.150
	CSE	Industrial structure rationalization	Difference duration	CSE	Industrial structure rationalization	Difference duration
k  =  1	0.939	0.847	0.092	0.894	0.826	0.068
k  =  2	0.914	0.770	0.144	0.887	0.756	0.131
k  =  3	0.896	0.734	0.162	0.844	0.687	0.157
k  =  4	0.863	0.685	0.178	0.814	0.644	0.170
k  =  5	0.828	0.655	0.173	0.785	0.621	0.164
Mean value	0.888	0.738	0.150	0.845	0.707	0.138
	CSE	Concentration of industrial structure	Difference duration	CSE	Concentration of industrial structure	Difference duration
k  =  1	0.939	0.842	0.097	0.894	0.810	0.084
k  =  2	0.914	0.751	0.163	0.887	0.743	0.144
k  =  3	0.896	0.716	0.180	0.844	0.683	0.161
k  =  4	0.863	0.620	0.243	0.814	0.616	0.198
k  =  5	0.828	0.628	0.200	0.785	0.621	0.164
Mean value	0.888	0.711	0.177	0.845	0.695	0.150

CEE: carbon emission economic efficiency; CSE: carbon emission social efficiency.

It can be seen from Table 4 that no matter what kind of decentralization method is adopted, the CSE club convergence index of the three metropolitan areas in Jiangsu Province is greater than that of CEE. The degree of solidification of the former is greater than that of the latter, and both of them are greater than that of the industrial structure club convergence index, and this difference tends to increase with the accumulation of time. This indicates that, from a dynamic perspective, CEE, CSE, and industrial structure between regions are obviously incongruous. The former shows the characteristics of high-low level solidification, namely “the high one is always high, the low one is always low,” while the latter has relatively high mobility, that is, with the change of time, some “disadvantaged regions” may perform better in industrial structure, showing a certain trend of catch-up. But these regions may face an “inefficiency trap” of carbon emissions. This may be related to the fact that the “feasible measures” taken by the government to the former are stronger than those taken by the latter, and the effect is faster. At the same time, it is related to the idea that China is still in the developing state. The further revelation is that, when improving the coupling coordination degree of carbon emission efficiency and industrial structure in the metropolitan area, the government should focus on the areas with the low coupling coordination degree caused by low carbon emission efficiency, and make adjustments by improving the carbon emission efficiency of these areas. Therefore, while improving the coupling coordination degree within the region. It can also alleviate the problem of curing inter-regional carbon emission efficiency.

Coupling coordination type classification of carbon emission efficiency and industrial structure

According to the research of relevant scholars ⁵³ and discussion in this paper, the coupling and coordination characteristics of “carbon emission efficiency and industrial structure” system can be divided into the following four types:

Coupling coordination types with high coupling degree and coupling coordination degree. Such cities generally include Nanjing (CEE-high-grade industrial structure, concentration), Suzhou, Changzhou (CEE-industrial structure concentration). The coupling characteristics of CEE and industrial structure in these cities are high coupling degree and coordination degree, which belong to the benchmark area of low-carbon economic development in the metropolitan area.

Coupling coordination types with high coupling degree but low coupling coordination degree. It can be further divided into two subcategories (II-1 and II-2). The low coupling coordination degree of II-1 is caused by the lag of carbon emission efficiency, while the low coupling coordination degree of II-2 is mainly caused by the industrial structure. For example, cities belonging to the former type include Zhenjiang and Yangzhou in Nanjing metropolitan area, Xuzhou and Lianyungang in Xuzhou metropolitan area, while cities belonging to the latter type include Huai ‘an, Ma ‘anshan and Suqian.

Coupling coordination types with low coupling degree and coupling coordination degree. It can also be further divided into III-1 and III-2 subcategories. The lower coordination degree of the former is mainly caused by the lag of carbon emission efficiency, while the latter is mainly caused by the industrial structure. Wuhu and Chuzhou are III-1 cities, while Shangqiu and Huaibei are III-2 cities. These areas are “vulnerable areas” in the development of low-carbon economy in China and have great potential and space for improvement of coupling degree.

Coupling coordination types with low coupling degree but high coupling coordination degree. Typical cities include Xuancheng and Wuxi (CEE—Industrial structure rationalization). It can be seen that the “carbon emission efficiency—industrial structure” system in Jiangsu metropolitan area still has a “low level coordination” phenomenon. There may be two reasons for the above situation: First, due to the mutual influence between urban carbon emission efficiency and industrial structure, the situation of “double low” is very likely to occur, that is, low carbon emission efficiency, low level of industrial structure upgrading, and the coupling coordination degree of “double low” area can reach a high level, resulting in the situation of “low level of coordination.” Measures need to be taken to break this low-quality coordination. Second, these “low-level coordination” cities often lack the motivation to break the low-level coordination “deadlock.” This is because when these regions break the “low level of coordination” in a phased and step-by-step way, worse results may occur, that is, not only does the development degree not reach the required height, but also lead to the poorer coordination degree.

The coupling coordination path of carbon emission efficiency and industrial structure

Considering that the carbon emission efficiency of metropolitan area is more solidified than that of industrial structure, the government should focus on supporting areas with long-term low carbon emission efficiency, such as II-1 and III-1 cities, to assist them to improve their carbon emission efficiency, so as to enhance the coupling and coordination level of the “carbon emission efficiency–industrial structure” system of these cities. In addition, as the improvement of carbon emission efficiency can also promote the improvement of system coupling level, it is hopeful that these cities can realize the coupling lifting path from II-1 to I or III-1 to I. This strategy also has the following two meanings: First, it can alleviate the problem of regional carbon emission efficiency solidification in metropolitan areas, avoid regions with low carbon emission efficiency falling into the “low level trap,” and realize the dynamic that coordinated development of carbon emission efficiency and industrial structure in metropolitan areas. Second, it can have an incentive effect on IV cities, prompting them to actively break the coordination state and improve the coupling degree step by step under the current situation of “low level coordination.” Specifically, the level of industrial structure should be first upgraded to enter II region or III region, and finally into I with high coupling degree and coupling coordination degree, so as to take a path of IV→II→I or IV→III→I. In addition, cities in other types of regions can also take targeted measures to improve the coupling coordination level of carbon emission efficiency and industrial structure according to their own actual conditions. For example, II-2 cities have high-carbon emission efficiency, but the coupling and coordination degree between them is poor due to the level of industrial structure. This kind of city should start from the industrial structure, optimize and upgrade the industrial structure, improve and enhance the quality of economic development, enhance the coupling degree of carbon emission efficiency and industrial structure, and then enter into the I with a high degree of coupling.

Conclusion

Firstly, this paper calculates the carbon emission efficiency (CEE, CSE) of 19 cities in the three metropolitan areas of Jiangsu Province from 2009 to 2019. Based on the trend distribution map, this paper calculates it from the perspectives of high-grade industrial structure, rationalization, and concentration, and analyzes the carbon emission efficiency and industrial structure from the perspective of space. Then the coupling and coordination model is used to investigate the coupling degree, coupling coordination degree, and coupling type of carbon emission efficiency and industrial structure of each city. Finally, under the framework of distribution dynamics, the promotion path of the coupling coordination between urban carbon emission efficiency and industrial structure is proposed from the perspective of the dynamic coordination between the two. The results are as follows:

The carbon emission of metropolitan area shows the center-multi-point divergent growth. The continuous high-energy consumption of the secondary industry and the growth of the economic aggregate of the third industry in the “five centers” maintain a high regional carbon dioxide emission.

From 2009 to 2019, the average CEE of 19 cities in the three metropolitan areas continued to increase, indicating that the contribution rate of the same amount of carbon emissions to economic income gradually increased, and the development status of low-carbon economy of cities in the metropolitan area was improving. The growth range of CEE index was greater than that of CSE index, indicating that the effect of carbon emission on the improvement of regional economic development was more significant than the improvement of public service level and the improvement of residents’ quality of life.

Optimization of the industrial structure index of the three metropolitan areas showed an upward trend, the overall gap within the metropolitan area in narrow, whole space from Jining-Xuzhou-Suqian-Huaian-Nanjing-Zhenjiang-Changzhou to optimization degrees developed areas inverted “C” shape structure shift to Xuzhou-Nanjing-Suzhou-Wuxi-Changzhou, the core of “northwest-southeast” spatial structure distribution. Three metropolitan areas present on the rationalization of industrial structure as a whole northwest rationalization (high)-southeast of rationalization (low) “three steps” distribution, industrial structure and equilibrium degree space narrow, Nanjing metropolitan area and Xuzhou metropolitan area industrial structure rationalization indexes average presents the downward trend after rising first. To some extent, it shows that the deviation degree between the industrial structure and the equilibrium state of Nanjing metropolitan area and Xuzhou metropolitan area becomes smaller, and the industrial structure tends to be reasonable. The average level of industrial structure concentration index in the three metropolitan areas showed an upward trend, and the intra-regional gap was also narrowing. On the whole, the distribution characteristics of the inverted “C” shape structure in the high concentration areas of Xuzhou, Yangzhou, Zhenjiang, Changzhou and Nanjing are similar to the distribution characteristics of the optimization of the industrial structure.

High-grade industrial structure in the Xuzhou metropolitan area is closely related to the improvement of CEE and CSE. The high-grade industrial structure is a key link affecting the improvement of carbon emission efficiency, the implementation of energy conservation, and emission reduction policies in the Xuzhou metropolitan area. The coordination degree of CSE- high-grade industrial structure in the Suzhou-Wuxi-Changzhou metropolitan area shows a downward trend, while both the high-grade industrial structure and CSE are in the rising stage. The deviation in the two trends reflects that the high-grade industrial structure has not brought the simultaneous growth of social welfare to a certain extent. The industrial structure rationalization in the Nanjing metropolitan area is closely related to the improvement of CEE. The industrial structure rationalization is the key link affecting the improvement of carbon emission efficiency, the implementation of energy conservation, and emission reduction policies in Nanjing metropolitan area. The concentration of industrial structure in the Nanjing metropolitan area is closely related to the improvement of CSE.

The curing degree of carbon emission efficiency in metropolitan area is greater than that of industrial structure (curing degree CSE > CEE > industrial structure). Specifically, the government should focus on supporting cities with long-term low carbon emission efficiency (II-1 and III-1) to improve their carbon emission efficiency, so as to enhance the coupling coordination level of the “carbon emission efficiency and industrial structure” system of these cities. The significance of this strategy is as follows: First, it can alleviate the problem of regional carbon emission efficiency solidification in the metropolitan area and avoid regions with low carbon emission efficiency falling into the “low level trap.” Second, it can generate an incentive effect for cities with low coupling degree and coupling coordination degree (IV), prompting them to actively break the coordination state and improve the coupling coordination degree step by step under the current situation of “low level coordination.” In addition, cities in other types of regions should take targeted measures to improve the coupling coordination level of carbon emission efficiency and industrial structure according to their own actual conditions. For example, for cities with high-carbon emission efficiency and industrial structure to be improved (II-2), the quality of economic development should be improved by optimizing high-grade industrial structure, and the coupling coordination between carbon emission efficiency and industrial structure should be enhanced.

Policy recommendations

The period of the 14th Five-Year Plan and the 15th Five-Year Plan is an important opportunity for China to achieve carbon peak and carbon neutrality. To promote the improvement of energy consumption structure in metropolitan areas with “high concentration, high consumption and high emission,” the government should respect the heterogeneity of local development when formulating policies, and formulate and implement low-carbon policies according to local conditions is an important measure to achieve sustainable development and continuously improve social welfare. From the actual overall situation of each metropolitan area, the current industrial development among Jiangsu metropolitan areas is not balanced, there is a large gap in the degree of development between industries, competition between regions is greater than cooperation, reasonable selection of industrial path suitable for the development of this region is the only way to the coordinated development of the three metropolitan areas in Jiangsu Province to improve the energy structure. From the perspective of each metropolitan area, Xuzhou metropolitan area, located in the economic depression in northern Jiangsu, currently needs to focus on the high-level development of industrial structure, actively break the low-level coordination state, break the positioning of the old industrial base and resource-based city, promote the regional industrial collaborative development of high quality, get rid of the “low level equilibrium trap,” and improve the coupling level within the metropolitan area step by step. We will achieve balanced and coordinated development in various areas. Nanjing metropolitan area and Suxi-Changzhou Metropolitan area located in the economically developed south of Jiangsu should formulate targeted policies and measures to promote the rationalization of the industrial structure of each city within the area in order to promote the development of low-carbon economy and the promotion of social welfare. For example, the core city of Nanjing is currently supported by electronic information, petrochemical, automobile manufacturing, and steel, and supported by emerging industries such as software and service outsourcing, smart grid, wind power and photovoltaic, rail transit, etc. ³⁴ It should speed up the construction of an industrial pattern of coordinated development of high value-added advanced manufacturing and modern services, and gradually phase out energy-intensive industries. Promote the rationalization of the city's industries, and at the same time drive the industrial economy as the pillar within the metropolitan area, Zhenjiang, Yangzhou, Wuhu, Maanshan, and other cities with a large proportion of capital-intensive manufacturing, as well as Chuzhou, Huai “an and Xuancheng, which are dominated by labor intensive manufacturing, to realize division of labor and cooperation, and gradually realize the rationalization of regional industrial layout. The rationalization of industrial structure is the key link for energy saving and emission reduction and the development of low-carbon economy. It is also an internal requirement of urban development and transformation. For Suzhou, Wuxi, and Changzhou, the GDP density is more than five times that of other metropolitan areas, and strategic emerging industries account for nearly half of their economic activities. Therefore, in the face of the high concentration of industrial space distance and industrial structure, the three cities need to further strengthen industrial coordination and division of labor, so as to promote overall low-carbon economic development and the improvement of social welfare.

Discussion

As a public health emergency, coronavirus disease 2019 (COVID-19), which began to spread widely in early 2020, has significant characteristics different from traditional financial crises and has caused severe impact and great interference with regional economic development and industrial structure. Therefore, it is more scientific to take 2020 as the dividing line between industrial structure and carbon emission research. At the same time, the data involved in this paper covers 13 categories of 19 cities in 4 provinces. Due to the differences and changes of the statistical caliber in different regions and considering the availability and scientific city of statistical data, the research period is selected from 2009 to 2019. The research covers the less-mentioned interrelationships between carbon emission efficiency, industrial structure upgrading, and social welfare improvement in metropolitan areas.

It is worth noting that (a) there are many factors affecting urban carbon emissions and there are also many evaluation indicators for industrial structure and social well-being. The research results obtained under different index combinations are often different or even contrary, so the indicators considered in the research need to be further improved. (b) Traditionally, the indicators obtained (or the indicators that can be counted) are often quantitative. The “quality” behind the quantitative indicators is usually difficult to quantify, but it is equally important for research. For example, the social benefits studied in this paper, such as the number of various medical and health institutions, the number of various schools, and other indicators, are often more influenced by the quality of these indicators than by the quantity, and the former is difficult to accurately reflect due to various reasons. (c) The coupling types and coupling paths of carbon emission efficiency and industrial structure obtained in this paper need to be further studied.