Identification of the Characteristics of the Industrial System of the Multilevel Urban Agglomeration in the Pearl River Delta

Yao Song,Yanxu Yu,Jiansu Mao

State Key Joint Laboratory of Environment Simulation and Pollution Control,School of Environment,Beijing Normal University,Beijing 100875,China

Keywords Industrial system Urban agglomeration Pearl River Delt

Abstract

1 Introduction

With the rapid development of urbanizatio n and regional economic integration,the interaction among cities has gradually strengthened and become more complex. As an aggregation of independent cities, promoting the development of the urban agglomeration has become an effective way for a country to plan as a whole. In the Twelfth Five-Year Plan announced in China, the urban agglomeration is clearly defined as the main form of urbanization.

The urban agglomeration in the Pearl River Delta, composed of the nine cities of Guangzhou, Shenzhen,Zhuhai, Foshan, Huizhou, Dongguan, Zhongshan, Jiangmen and Zhaoqing, has the obvious geographical advantage of being adjacent to Hong Kong and Macao. In 2015,it created 9.20%of China’s gross national product(GDP)with only 0.57% of its area. This area is certainly one of the most vigorous economic zones in China,and its demonstration study can predict and guide the industrial development of other regions. On the other hand,although the degree of urban development in the Pearl River Delta is at the forefront in China,the developments of the nine component cities are unbalanced due to the inconsistencies of their resource endowments,urban scales and socio-economic levels(Zhu et al.,2011). There are one or two regional cores in this area,and many economic circles with significant different structural have been formed around them (Li and He, 2013).Moreover,there are obvious differences between the core cities and the marginal cities(Mei et al.,2012). Therefore, the study of this area is also of great practical significance for promoting its coordinated and sustainable development.

Based on the requirements of regional planning and management,Huang et al. (2016)constructed a framework of the multilevel structure of the urban agglomeration that provided a foundation for exploring the internal relationships and cluster effects of urban agglomerations. On this basis,most studies of the relationships between the urban agglomerations and their component cities in China in recent years have focused on the empirical study of the radiation effect of core cities(Zhao et al.,2017;Jin and Zhao,2016),the functional distribution within a specific urban agglomeration (Ding and Wang,2016; Lu,2015), the interaction intensity of cities and its influencing factors(He et al.,2017;Ye et al.,2018),and the quantitative analysis of the cluster effect(Li et al.,2018;Huang et al.,2018).

As an important link in transforming resources into services,the industrial system(IS)has strengthened the relationship among cities owing to its complexity in resource input,product processing and so on. The agglomeration economy formed by inter-city industrial agglomeration contributes to the development and maturity of urban agglomerations (Su, 2004). Combes et al. (2012) took France as an example and noted that industrial agglomeration is conducive to improving industrial productivity. Feng et al. (2018b) estimated the impact of agglomeration economies on carbon emissions in China and found that industrial agglomerations can promote carbon emissions reduction in local and surrounding cities. Zheng and Lin (2018) found that promoting an industrial agglomeration when it is low can improve energy efficiency. Previous studies have shown that the cluster effect of inter-city industrial agglomeration has a certain impact on the regional economy(Zhang,2014),industrial energy consumption (Liu et al., 2017; Zhao and Lin, 2019) and environmental pollution discharge(Shang and Mao, 2016; Chen et al., 2018). Therefore, it is of great significance to manage IS from the perspective of urban agglomerations for regional development. Current studies have quantitatively analyzed the impacts of an IS in a specific urban agglomeration (Feng et al., 2018a; Liu et al., 2018; Shen et al., 2019), as well as the changes and influencing factors of its industrial eco-efficiency(Li and Li,2011;Ma and Liu,2015).However,they all consider an urban agglomeration as a whole and seldom consider the differences in its internal composition.

At present, the methodology of deconstructing an IS is mostly based on the classification of primary, secondary and tertiary industries(Qi and Zhang,2015;Yu,2015)or of heavy and light industries(Liu et al.,2018).The methodology focuses on the evolution(Jiang and Deng,2015)of industrial structures,industrial impacts on externalities (Li et al.,2018),and the adjustment(Lin and Xu,2018)of industrial structures. Moreover,Mao et al. (2010,2011)analyzed the eco-efficiency of 41 industrial sectors in China and five major cities and proposed some suggestions to achieve sustainable development via structural adjustment. Song et al. (2018)reconstructed the IS according to the material flow process from resources to products and analyzed basic characteristics of the ISs of three major cities in Guangdong Province;however,overall,there are still few studies related to industrial sectors or industrial production phases.

Fig. 1 The framework of the relation between the industrial system and its externalities. EE: Energy exploitation; ME:Material exploitation;MP:Industrial material production;PM:Product processing and manufacture;MR:Product maintenance and recycling; E: energy consumption; R: material resource consumption; G: economic output; P: output of services;Q: pollutant discharge;superscript m,l,n,k: number of types.

To solve the problems mentioned above,this work divides the Pearl River Delta into several levels centered on the core city that constantly expand outward. At the same time, the IS will be reconstructed based on the material flow process from resources to products. To obtain the changes of the IS under the cluster effect as well as the influence of each component city on the characteristics of the urban agglomeration,a relation framework and an evaluation index system for the IS are developed, and the impacts of the IS on externalities will be established to quantitatively and comprehensively analyze the structure,the impacts,and their relations at each level. The research conclusions will provide a basis for the management of the IS in the Pearl River Delta.

2 Materials and methods

2.1 Industrial system framework

According to the material flow process from resources to products, a specific IS can be re-divided into four phases: resource exploitation (RE), industrial material production (MP), product processing and manufacture(PM)and product maintenance and recycling(MR).The RE phase can be further separated into energy exploitation (EE) and material resources exploitation (ME). While providing services for human society and bringing economic growth, all phases of an IS consume materials and energy and discharge a variety of types of pollutants into the environment during the process of production. Therefore, the IS has impacts on the economy,society, resources and environment. Based on this,a framework of relations among phases and between the IS and externalities is established(Song et al.,2018),as shown in Fig. 1.

G represents total profit,with a unit of 100 million Chinese Yuan(CNY),P represents the output of products,and the superscript l is the serial number of the types of products. R represents resource consumption, the superscript n is the serial number of the types of resources, and energy is mainly considered and studied as a single system in this study, as the red line in Fig. 1. E represents energy consumption, with a unit of tons of standard coal equivalent (tce), and the superscript m is the serial number of the types of energy. Q represents pollutant discharge,and the superscript k is the serial number of the types of pollutants. The grey dashed line in Fig. 1 shows the recycling of components or materials after a product is scrapped,although if this flow actually exists,it will not be considered in this study.

The composition of each phase is determined according to the classification of 41 industrial sectors in the national economic industry classification standard (GB/T 4754-2011), as well as their industrial activities. For example, the sector named mining and processing of ferrous metal ores (FMM), the main activity of which is mining ferrous metal ores from the natural environment, belongs to the RE phase. The sector named smelting and pressing of ferrous metals (FMS) belongs to the MP phase, with its main activity being the preliminary processing of mined ferrous metal ores into industrial materials. The codes of industrial sectors and phases to which they belong are summarized in Table A1.

2.2 Evaluation index system

Based on the framework of the relation among industrial components and between the IS and its externalities,the basic characteristics of an IS should be considered from both internal and external perspectives. At the same time,China is regarded as a basic standard in this study in regard to the comparison of different regions.

2.2.1 Internal characteristics

(1)Structure coefficient

To clarify the sectors that should be emphasized in regional industrial development, this study quantifies the internal structure by defining the ratio of the gross industrial output of a sector to the entire IS in a specific statistical period as this sector’s structural coefficient(fi),expressed as:where V represents the gross industrial output, with a unit of 100 million CNY,and the subscript i represents the serial number of industrial sectors, i=1,2,...,n. In a specific region,∑ni=1fi=1. For a specific sector, the higher its fi,the greater the share of total gross industrial output. It is considered that a sector is dominant in a region when the structural coefficient is greater than 0.05.

(2)Location quotient

Because the structure coefficient can only reflect whether a sector is dominant or not in a specific region, to further clarify its degree of specialization in the entire country, this study defines the ratio of the structural coefficient of a sector in a specific region to that in the entire country as this sector’s location quotient (LQi),which can be expressed as:where the subscript?represents China in Eq. (2). When the location quotient of a sector is greater than 1,this sector has an advantage in a specific region compared with the entire country, and its related products are exported to other regions. Moreover, when the location quotient of a sector is greater than 2, its capacity to export is considered to be very strong.

2.2.2 External characteristics

(1)External impacts

In this study, the impacts of the IS on the society, economy, resources and environment will be considered.Because the establishment of the framework of the relationships between the IS and its externalities is basedon the material flow process from resources to products, the main consideration in this study is the material change during the industrial production process with respect to the selection of indexes. To make different levels comparable, indicators characterized by absolute quantity are transformed into a series of relative-level indicators. Impacts on society are characterized by per capita product output,and impacts on economy,resources and environment are characterized by total profit per unit of gross industrial output,energy consumption per unit of gross industrial output and pollutant discharge per unit of gross industrial output, respectively, which is different from the analysis of a specific region(Song et al.,2018). The external indicators and their calculation methods are summarized in Table 1. Note that high levels of energy consumption and pollutant discharge are undesirable to achieve sustainable development.

Table 1 External indexes and their estimation formulas.

(2)Contribution rate

To clarify the main sources of each impact,this study also calculates the absolute quantity of each industrial sector in proportion to the entire IS.The industrial composition of each impact is characterized by the contribution rate(cxi),with a unit of%,which can be expressed as:where x represents the absolute quantity of a specific impact,which can be G,E and Q in Eq. (3).

2.2.3 Relational characteristics

Eco-efficiency is often used to measure the relation between the internal structure of a specific IS and the impacts on its externalities (Mao et al.,2010, 2013). Because of the limitation of data acquisition and the emphasis of research, this work only takes energy impact as an example and then defines the energy efficiency (e) as the gross industrial output per unit of energy consumption, with a unit of 10 thousand CNY·tce-1,which is shown in Eq. (4):

For a specific IS, the energy efficiency of a sector (ei) can also be estimated according to the equation ei=Gi/Ei. Combined with Eq. (1),Eq. (4)becomes:

As shown in Eq. (5),energy efficiency can reflect the influence of industrial structure on energy consumption. Increasing the structural coefficient of sectors with high energy efficiency can improve the energy efficiency of the IS.

2.3 Division of levels

Fig.2 Multilevel urban agglomeration in the Pearl River Delta.

This study takes the Pearl River Delta as the research area. Shenzhen, the industrial center of Guangdong province, contributed 22.28% of the gross domestic product (GDP) of Guangdong province in 2015, which ranked first among the nine cities in the Pearl River Delta. Guangzhou and Dongguan likewise play important roles in the industrial development of this area by virtue of their advantages as the provincial capital city and manufacturing sub-center,respectively. On this basis,this study divides the Pearl River Delta into the following three levels: Shenzhen,the core city,is the first level(I);the central city belt composed of Guangzhou,Shenzhen and Dongguan is the second level (II); and the urban agglomeration composed of nine cities is the third level(III).The composition and the extended relationships among the three levels are shown in Fig. 2.

2.4 Data and sources

This study takes China as the basic standard and chooses 2015 as the current year. The materials involved mainly include statistical data,national development policies and the actual regional situation. Statistical data,such as gross industrial output,product output,total profit,energy consumption and pollutant discharge, can all be obtained from the statistical yearbooks of the nine cities (SBGZ,2016; SBSZ,2016; SBDG,2016; SBFS,2016; SBZH, 2016; SBZS,2016; SBHZ,2016; SBJM, 2016; SBZQ,2016), Guangdong Statistical Yearbook 2016(SBGD,2016)and China Urban Statistical Yearbook 2016(NBSC,2016). National development policies can be obtained from the official website for the National Development and Reform Commission of the People’s Republic of China(http://www.ndrc.gov.cn),and regional industrial development policies can be obtained from the official website for the department of industry and information technology of Guangdong province(http://www.gdei.gov.cn). The actual situation of the Pearl River Delta can be obtained from local investigation and survey.

Fig. 3 Internal structure of the industrial system at each level. The y-axis represents the structural coefficient, and the x-axis represents codes of sectors or phases.

3 Results

3.1 Industrial structure

3.1.1 Structural coefficient

The structural coefficient of each industrial sector and phase at the three levels in 2015 are calculated. Sectors with higher structural coefficients or more obvious changes among the levels are selected and presented in order of level from high to low in Fig. 3.

As shown in Fig. 3, the lack of mineral resources in the Pearl River Delta has limited the development of the RE phase and the MP phase,the structural coefficients of which are all lower than that of China at the three levels. In contrast, the PM phase is the phase with the largest scale at each level, and its structure coefficient reaches 0.91 at the first level. Although it decreases with the enlargement of level,it is still higher than that of China. This is because the location advantage of being adjacent to Hong Kong and Macao as well as the policy advantage of opening-up has enabled the Pearl River Delta to develop manufacturing industry with the support of foreign capital. Subsequently, with the goal of building a modern industrial system, the electronic industry was vigorously developed in Shenzhen, Dongguan under the guidance of national policies. The sector of the manufacture of computers,communication and other electronic equipment(CEM)in this phase has become the most significant dominant sector, with structural coefficients of 0.59, 0.40 and 0.29 at the three levels in 2015,which are much higher than those of the other sectors. Moreover, some sectors are also dominant, such as the sector of the manufacture of electrical machinery and apparatus (EAM,0.08)at the first level;the manufacture of automobiles sector(AMM,0.08)and EAM(0.07)at the second level;and EAM(0.11),AMM(0.06)and the manufacture of raw chemical materials and chemical products sector (CMM,0.05) at the third level. It can be seen that the industrial development of all three levels obviously concentrates on the high-tech manufacturing industry and the large equipment manufacturing industry; however, the degree of centralization decreases with the enlargement of level and instead presents a more diversified characteristic.

3.1.2 Location quotient

The location quotient of each industrial sector and phase at the three levels in 2015 are also calculated. Dominant sectors and sectors with higher location quotient are summarized in Table 2.

As shown in Table 2,the RE phase and the MP phase of the Pearl River Delta are at a disadvantage in theentire country, especially the RE phase, whose location quotients at the three levels are all less than 0.25. In contrast,the PM phase of this area has an obvious competitiveness,although its location quotient decreases from 1.48 at the first level to 1.28 at the third level,which indicates the reduction of agglomeration degree and product output capability. Regarding the dominant sectors at each level,CEM not only contributes greatly to economic growth of the three levels but also has strong specialization and product output capacity. Its location quotient is as high as 7.10 at the first level. In addition,the manufacture of articles for the culture,education,arts and crafts,sport and entertainment activities sector (ASM) and the production and supply of the water sector (WPS) are the other two sectors with strong output capability at the three levels,although their location quotient decreases with the enlargement of level, as does the CEM sector. The number of sectors whose location quotients are greater than 2 is largest at the third level, which shows that the economic composition of the IS at this level is more diversified and that the nine cities within the urban agglomeration have formed a cross-complementary advantage in industrial development.

Table 2 Location quotient of industrial phases and sectors at each level.

Table 3 Impacts of the industrial system on its externalities at each level.

3.2 Impacts of the industrial system on its externalities

The relative level of each impact on the externalities at the three levels in 2015 are calculated and summarized in Table 3.

Fig.4 Industrial composition of total profit at each level. The y-axis represents contribution rate,and the x-axis represents codes of sectors or phases.

As shown in Table 3,although the industrial energy consumption and pollutant discharge of the three levels are far less than those of China,they all increase with the enlargement of level, which means that the negative impacts on the environment are more serious while the gross industrial output is the same. Especially for the discharge of waste solids and smoke and dust,their values at the third level are approximately 450 times and 30 times those at the first level,respectively. As for the impact on economy, the total profit of the IS is as high as 71.72 million CNY/billion CNY in Shenzhen,whereas it is only 58.86 million CNY/billion CNY at the second level,which is the lowest among the three levels and is lower than that of China as a whole.

Moreover, the specific phase and sector sources of each impact are also calculated and are presented in the following sections.

3.2.1 Impact on the economy

The contribution rates of each industrial sector and phase to the total profit at the three levels in 2015 are calculated. Sectors with higher contribution rates or more obvious changes among levels are selected and presented in order of level from high to low in Fig. 4.

As shown in Fig.4,the PM phase is the most profitable phase at each level but,with the enlargement of level,the contribution rate of the MP phase gradually increases and exceeds 20%at the third level,which is mainly due to the rapid increase of that of CMM from 1.29%to 6.37%. CEM is the sector that brings the most total profits to each level,especially to the first level,with a contribution rate of 58.45%,which is related to the large domestic and export demand of related products. Moreover, there are also some other sectors with higher contribution rates, such as AMM (10.86%) and the production and supply of electric power and heat power sector (EHP,8.60%)at the second level,and EAM(12.46%)at the third level,which all are more outstanding compared with China. Combined with the analysis of Fig.3,it is not difficult to find that the industrial distributions of economic impact are coincident with that of the structural coefficient at the three levels, which shows that the dominant sectors have formed not only a scale effect under the support of policy but also a strong market competitiveness.

3.2.2 Impact on society

To make the production of different products at the same level and the production of the same product at different levels both comparable,the ratio of per capita output of each product at each level to that of China is calculated and presented in Fig. 5, where the outputs of products with different units are transformed into dimensionless constants.

Fig. 5 Production of industrial products at each level. The y-axis represents names of industrial end products, and the x-axis represents the ratio of per capita output of each product to that of China.

As shown in Fig.5,the social impact at the first level is mainly manifested in the provision of semiconductor integrated circuits, micro-computers and electronic units for society. Their outputs are 14.57, 10.69 and 3.93 times that of China. Whereas the productivity of electronic units is as strong as that at the first level,the product types at the second and third levels are more diversified. The per capita outputs of some products with greater demand for labor or resources, such as furniture, clothing, machine-made paper and paperboards and plastic products, are far higher than those of China, especially for machine-made paper and paperboards, which have high energy consumption and pollution during production and a per capita output that reaches 5.32 times that of China. However, the per capita outputs of pig iron, crude steel and steel products, products related to the iron and steel industry,are less than 30%of China’s per capita outputs at each level.

3.2.3 Impact on resources

The contribution rate of each industrial sector and phase to the energy consumption at the three levels in 2015 are calculated. Sectors with higher contribution rate or more obvious changes among levels are selected and presented in order of level from high to low in Fig.6.

As shown in Fig. 6, the MP phase is the phase with the largest energy consumption at each level. Its contribution rate increases with the enlargement of level and reaches 80.84% at the third level, which exceeds that of China (77.87%). In this phase, as an indispensable sector to meet regional demands of electric power,EHP contributes 55.68%,45.40%and 32.96%of energy consumption at the three levels,which are much higher than that of China(5.24%). In contrast,some sectors that mainly depend on energy input,such as the smelting and pressing of ferrous metals sector(FMS),CMM and the manufacture of non-metallic mineral products sector(NMM),which have high proportions of energy consumption in China,are not outstanding in the first two levels.In addition, the contribution rate of the manufacture of petroleum,coking and processing of nuclear fuel sector(PCM)varies greatly among the three levels,with a change from 0.01%at the first level to 23.56%at the third level. The PM phase has a trend completely opposite to that of the MP phase; its dominant CEM sector is the main reason for its gradual reduction of the proportion to total energy consumption.

Fig.6 Industrial composition of energy consumption at each level. The y-axis represents contribution rate,and the x-axis represents codes of sectors or phases.

Table 4 Production and disposal of industrial waste at each level.

3.2.4 Impact on the environment

In this section, due to the lack of pollutant discharge data for each industrial sector, the production, treatment and utilization of sulfur dioxide and waste solids at each level are collected and calculated instead to reflect the ability of reducing the impacts of IS on the environment. All of them are summarized in Table 4.

As shown in Table 4, thanks to its lowest production of pollutants and strongest capacity of waste endtreatment and recycling, the IS at the first level causes the least environmental impacts among the three levels.The disposal rate of sulfur dioxide and the comprehensive utilization rate of waste solids at the first level reach 88.60% and 99.86%, respectively. The capacity of waste end-treatment and the production of pollutants show a completely opposite trend of change with the enlargement of level. It can be seen that the implementation of cleaner production during industrial processes, the improvement of pollutant end-treatment capacity,and the increase of the recycling of intermediate products or components by the dismantling of discarded products are all effective ways for reducing industrial pollutant discharge.

3.3 Relation between industrial system and environment

The energy efficiency of each industrial sector and phase at the three levels in 2015 are calculated. Sectors with higher energy efficiency or larger scale or higher energy consumption are selected and presented in order from high to low at each level in Fig. 7.

Fig.7 The energy efficiency of industrial phases and sectors at each level. The y-axis represents energy efficiency with a unit of 10 thousand CNY/tce,and the x-axis represents codes of sectors or phases.

As shown in Fig. 7, the energy efficiency of ISs and phases at the three levels are all higher than those of China, especially the PM phase, the energy efficiency of which reached 417.94 thousand CNY/tce at the first level, 351.57 thousand CNY/tce at the second level and 338.36 thousand CNY/tce at the third level. Although the energy efficiency of the MR phase is the highest at the second and third level,the driving effect of this phase in increasing energy efficiency of the entire IS is negligible because of its extremely small scale and low energy consumption. The energy efficiencies of the dominant sector CEM are as high as 632.11 thousand CNY/tce at the first level, 636.52 thousand CNY/tce at the second level and 588.97 thousand CNY/tce at the third level,which are much higher than those of the IS at the corresponding levels. There is no doubt that this sector has a strong driving effect in improving energy efficiency,with its high contribution of energy consumption at each level. Moreover, there are many other sectors with extremely high energy efficiency such as PCM,ASM, the smelting and pressing of non-ferrous metals sector (NFS)and the manufacture of tobacco sector (TOM)at the first level,AMM and TOM at the second level,and the mining and processing of non-ferrous metal ores sector(NFM)and TOM at the third level. However, the energy efficiency of EHP,whose energy consumption is the highest at each level,is much lower than that of China. The number of sectors with high energy consumption but low energy efficiency is increasing with the enlargement of level. The above two points show that there is great potential to achieve the sustainable development of the IS in the Pearl River Delta by the improvement of energy efficiency of each sector and structural adjustment. For different levels,the sectors that should be focused on are different.

3.4 Causal analysis of level change

To determine the reason for the change of the degree of external characteristics among the three levels under the cluster effect and to assign the industrial management in the Pearl River Delta into each component city, this study further calculated the external characteristic indexes of each city,presented in Fig. 8.

Combining Fig. 8 and the previous analysis, cities with both high contributions and ratios are the main sources of a specific impact and are the key points for changing this impact in the Pearl River Delta. Jiangmen,Zhaoqing and Huizhou, which are the main contributors of sulfur dioxide, waste solids, smoke and dust and energy consumption in the Pearl River Delta,have caused more serious negative impacts of the IS in the urban agglomeration compared with that in the core city. It can be seen that the reduction of the negative impacts of the IS in the Pearl River Delta should begin in these three marginal cities. Implementing cleaner production during industrial processes and improving pollutant end-treatment capacity are both effective ways to make the marginal cities keep up with the speed of industrial development and the upgrading of the entire urban agglomeration.On the other hand, the total profit of Dongguan, one of the major cities in the central city belt, is far lower than that of Shenzhen and Guangzhou, although it generates the same gross industrial output. Its discharge of sulfur dioxide,waste solids and waste water,which are much higher than those of other two cities,has brought a heavy burden on the environment in the Pearl River Delta. There is no doubt that Dongguan has caused negative impacts at the second level greater than that at the first level. To give full play to the benign leading and driving role of the central city belt in industrial development of the entire area,the eco-efficiency of the IS in Dongguan should be improved.

Fig.8 Urban sources of impacts of the industrial system on the external characteristics in the Pearl River Delta. The y-axis represents the ratio of the nine cities in relative level of each impact with respect to the urban agglomeration,and the x-axis represents the contribution rate of the nine cities in the absolute quantity of each impact on the urban agglomeration.

3.5 Data uncertainty

Although the data of the various cities in this study are all from their statistical yearbooks, some problems inevitably exist such as missing data and bad data quality. For example, the energy consumption data of each industrial sector in Zhongshan are lacking. However,according to the data published in 2015,the total industrial energy consumption of cities,except for Zhongshan in the Pearl River Delta,accounts for 96.2%of that of the urban agglomeration,and the eight cities are thus used to represent the urban agglomerations for the analysis of resource impact. Moreover, the types of industrial products published in yearbooks of different cities are very different but it is unknown if the unpublished products are due to the lack of corresponding production activities.Therefore,this study selects products for which most of the 9 cities have published output data as indicators.

4 Conclusions

At the methodological level,this work reconstructs the IS according to the material flow process from resources to products,which includes the four phases RE,MP,PM,and MR.On this basis,a framework and an evaluation index system are established to evaluate the structure, the impacts, and the relations of a specific IS to its externalities, which are universal and can thus provide a comprehensive method of characteristics identification for ISs in different regions. Then, the possibility of achieving sustainable development in an entire area by structural adjustment and urban industrial management can be discussed.

In application,this study quantitatively identified the basic characteristics of ISs in Shenzhen,the Guangzhou-Shenzhen-Dongguan urban belt, and the urban agglomeration, three levels based on data for nine cities in the Pearl River Delta in 2015. The main contributions are as follows.

First, we calculated structural coefficients of IS components at three levels, as well as their contribution rates to each external impact, and determined that the industrial development of the Pearl River Delta shows significant centralization and mainly focuses on CEM,EAM and AMM in the PM phase. With the convergence to the core city Shenzhen, the advantages of CEM in economic growth and corresponding product output become increasingly obvious. Additionally, the ISs at the three levels also have similar external characteristics compared with China as a whole,for example,higher total profit,larger output of electronic units,lower energy consumption and lower pollution discharge. However,at the same time,the differences of the relative level of each impact at the three levels reflect the imbalance of development in the Pearl River Delta.

Then, we calculated the energy efficiency of the industrial phases and sectors at the three levels and determined that the energy efficiency of ISs at the three levels are all higher than that of China. However,the driving effects of phases in increasing energy efficiency of the entire IS are very different because of their scale and energy consumption. The driving effect of the MR phase is negligible,whereas that of the PM phase,especially of its dominant sector CEM,is strong at each level. Moreover, the number of sectors that can play important roles in improving the energy efficiency of the IS via structural adjustment increase with the enlargement of level,which shows that the irrationality of industrial structure in marginal cities is greater than that in core cities.

Finally, we calculated the external characteristic indexes of nine cities in the Pearl River Delta, then compared them with those of the urban agglomeration. We determined that Huizhou,Zhaoqing and Jiangmen are the main sources of negative impacts of the IS in the urban agglomeration,whereas Dongguan is the main source of negative impacts of IS in the central city belt.

The empirical results obtained in this work show that Shenzhen, the central city,plays a demonstrable role in guiding the development of IS in the Pearl River Delta;however,the fact that marginal cities cannot keep up with the speed of development of the urban agglomeration makes the characteristics of the third level inferior to those of the first level. Therefore,to improve the industrial development of IS in the Pearl River Delta and reduce its negative impacts,the government should aim to: (1)Implement policies to promote the guiding role of core cities in industrial development and focus on strictly controlling the development of sectors with high pollution and high energy consumption in Dongguan, such as PAM and EHP, to further enhance the positive driving effect of the central city belt to marginal cities. (2) Establish a platform for inter-city industrial cooperation and promote the rational allocation of resources and innovative technology sharing as follows: (a) Share the production technology of high-tech sectors to promote the diffusion of pillar sectors and the formation of scale effect and competition ability in marginal cities. (b)Share the technology of cleaner production and pollutant end treatment,especially with Huizhou,Zhaoqing and Jiangmen,to promote the development of green industry. (3)Optimize the industrial structure,and improve the transformation and upgrading of the IS.Adjust the industrial structure of the IS in the Pearl River Delta by increasing the support for sectors with high eco-efficiency, such as related sectors in the MR phase and TOM,to realize the sustainable development of the IS.

However, due to data limitations, this study fails to analyze the contribution of various sectors to environmental impacts, and the environmental efficiency is not considered when discussing eco-efficiency. Moreover,the dynamic changes of industrial characteristics are not considered. Therefore, the influence of urban interactions on the characteristics of the IS at different levels has not been analyzed in this article but can be discussed in detail in future studies.

Acknowledgments

This work has been supported by the National Key Research and Development Program of China(No.2016YFC-0502802).

APPENDIX

Table A1 The codes and compositions of industrial phases.

Aims and Scope

This peer-reviewed journal publishes original research results in the field of environmental and ecological science, with focus on sustainable management of natural, human-dominated, and man-made ecosystems through appropriate resource use, quality assessment and recovery. The aim of the journal is to provide a place for a deep exchange of new ideas and concepts in the fields of environmental accounting, human and systems ecology, environmental management and environmental policy making. Manuscripts on environmental accounting and managements are solicited, including: mathematical modeling; computational and management techniques in environmental, ecological, energy and information science; environmental technology and engineering; human managed ecosystems such as agricultural, urban, coastal, riparian, and wetland ecosystems. Topics of interest include but not limited to:

· Environmental and ecological economics, environmental accounting (Life Cycle Assessment, emergy, footprints, etc), environmental impact assessment, ecosystem services and natural capital assessment, energy and resource use, social factors and impacts, circular economy.

· Modeling in environmental conservation and restoration, eco-hydrology and water resources management, ecological processes and patterns, climate change effects, environmental engineering and technology.

· Planning and management in human dimension-institutions and patterns for socio-economic systems, industrial ecology, ecological informatics, landscape design, and urban metabolism and planning.

· Environmental policy, legislation, and innovations with environmental and strategic impact assessment, project appraisal and auditing, and environmental protection.

No length limitations for contributions are set, but only concisely written manuscripts are considered for publication. Brief papers can be published as Technical Notes. Review papers and discussions of previously published papers are also welcome.

Peer Review

Journal of Environmental Accounting and Management adopts the traditional blind-review policy in which the reviewers’ names are concealed for submission author(s). Each submitted manuscript is reviewed by at least two expert referees in the related research field.

Manuscript Submission Guidelines

Author(s) should submit manuscript(s) via the website: https://www.editorialmanager.com/JEAM/Default.aspx. For any questions, author(s) can directly contact one of Editors. Manuscript preparation should follow the sample papers and journal manuscript templates (word file or Latex files), which can be downloaded from the website: https://lhscientificpublishing.com/Journals/JEAMDefault.aspx

Proofs and Electronic Offprints

Authors will receive a PDF proof of their articles and, on publication, electronic offprints of the article.

Open Access

After manuscript acceptance, author(s) has the right to request publication of the manuscript by L&H Scientific publishing with open access. If published as an open access, the author(s) has a responsibility to paying a publication fee of US$60 per page. The copyright is still held by the Publisher. Without any request for open access publication, all accepted manuscripts are assumed to be published traditionally.