Research on Spatial Correlation Evolution of Marine Ecological Efficiency Based on Social Network and Spatial Correlation Matrix Model

Abstract

Most developed countries in the world are maritime powers. This article constructs an ecological efficiency evaluation system based on the characteristics of the ocean itself while taking into account the relationship between land and sea. Based on social network analysis, the relationship between China’s marine ecological efficiency is regarded as a social network system, and the roles and positions played by 11 coastal cities in the network are analyzed from a relational perspective. Next, based on the unexpected super efficiency model to measure the ocean efficiency value in China’s coastal areas, we explore its spatiotemporal evolution characteristics, and measure the ocean ecological efficiency while incorporating ecological environmental pollution as an unexpected output into the evaluation system. Then, the spatial incidence matrix of marine ecological efficiency is calculated through the modified gravity model, and the characteristics of network structure are described with the help of the social network method. In addition, ArcGIS software is used to visualize the spatial evolution process. Finally, QAP regression is used to explore the key factors affecting the spatial correlation network of marine ecological efficiency in China. The results show the following: (1) In terms of time, the marine eco-efficiency of most provinces is not high, and the difference between provinces is obvious, but on the whole, it shows a fluctuating upward trend. (2) From the perspective of space, the overall displacement of the center of marine eco-efficiency in China is large in the north–south direction and small in the east–west direction, and the center of marine eco-efficiency is always concentrated near the Yangtze River Delta. (3) On the whole, the spatial correlation network of China’s marine eco-efficiency is becoming more and more complex. The number of correlation relationships and network density is increasing, and the network framework is gradually maturing. The spatial adjacency matrix, the difference in economic development level, and the difference in population distribution level can significantly promote the formation and development of the spatial correlation network of marine eco-efficiency. However, the differences in the level of opening to the outside world and in the structure of marine industries restrain their spatial networks. The difference between marine science and technology levels is not significant.

1. Introduction

The 21st century is the century of the marine environment. As the strategic space of Chinese economic development, the ocean is becoming increasingly important [1]. With the increasing contradiction between land and population and the decrease of certain terrestrial mineral resources, people focus more on the marine environment, and the marine economy plays an important role in social development. The marine ecological problems are also becoming more and more serious. The marine power strategy is an inevitable choice for social development and continued national prosperity, and marine ecological civilization is a vital part of the marine power strategy. To solve the contradiction between Chinese marine economic development and ecological environment better, we need a set of scientific and systematic theory systems to evaluate the marine ecological environment. This also can provide a reference for future marine development and ecological environment protection. Therefore, based on the features of the ocean itself, and taking into account the correlation between land and sea, this paper uses data envelopment analysis to measure and extend the study of eco-efficiency from land to sea, and constructs a network of spatially related relationships to explore its network characteristics and influencing factors. This will provide a scientific and theoretical basis for research on the sustainable development of the marine economy and marine eco-efficiency. China is a country with large marine resources, and coastal areas are an important part of “the Belt and Road”. However, in the early stages of development, the development model was mainly “extensive”, wasting a significant amount of marine resources. The marine environment also paid a high price for this. Nowadays, China’s marine industry has entered a period of rapid development, but the structural contradictions in the marine industry are still prominent, and pollution problems still exist. Therefore, under the assumption that environmental benefits are taken into account, measuring marine ecological efficiency scientifically, building an ocean spatial correlation network, and conducting research and analysis on the structure’s evolution process all have practical implications for further raising the level of the marine economy in coastal areas and achieving the marine economy’s sustainable development.

2. Literature Review

In the 1990s, the definition of eco-efficiency was first proposed by German economists Sturn and Schaltegger. Eco-efficiency was defined as the ratio of economic growth to ecological load, and was widely used to measure the relationship between the ecological environment and economic benefits [2]. Then, with the continuous further study of eco-efficiency, institutes and many scholars made complementary explanations for the definition and connotation of eco-efficiency from different perspectives. In 1996, the World Business Council For Sustainable Development (WBCSD) defined eco-efficiency as the leadership of improving economic and environmental performance [3]. In the year 1997, Meier thought of eco-efficiency as the correlation between the ecosystem benefits and deficits, where the benefits are economic increase and ecological improvements, and deficits are economic inputs and ecological destruction [4]. In 2001, Sterm and Muller defined the eco-efficiency as the ratio of environmental load to economic growth [5]. According to Lu Zhongwu, to establish a circular economy, eco-efficiency and economic growth are closely related [6]. Although domestic and foreign experts differ on eco-efficiency, the central concept of eco-efficiency is to obtain the greatest possible economic benefits with the least possible resources and environmental pollution, which has become a consensus among scholars in various countries. In the year 1978, operations researchers Rhodes, Charnes, and Cooper first proposed the DEA-CCR model, which was used to evaluate the relative effectiveness between the same sectors [7], then the model was first used by Sarkis to measure eco-efficiency in 2001 [8]. Asmart et al. combined the economic performance and environmental performance, and analyzed 169 cotton cropping systems individually through life cycle evaluation [9]. Zheng et al. used a non-expectation super-efficiency DEA analysis on the inter-provincial regional eco-efficiency in China in the context of new urbanization [10]. Wu and Lei used a three-stage DEA model to measure agricultural land eco-efficiency in several prefecture-level cities in Henan Province and explored its influencing factors [11]. Huang et al. measured the regional industrial eco-efficiency in China based on Pareto’s improved two-stage DEA cross-efficiency model with a circular economy perspective [12]. Cong and Han constructed an ecological indicator system and conducted DEA-Malmquist analysis on Chinese high-energy-consuming industries to identify eco-efficiency gaps and provide countermeasures for industrial development [13]. Wang et al. used the super epsilon-based measure (super-EBM) to measure the green innovation efficiency of 26 cities in the Yangtze River Delta region of China from 2003 to 2018; they used the Moran index, centrality analysis, and a block model to investigate its spatial characteristics and empirically analyze its influencing factors. Their findings expanded the research on traditional innovation efficiency and provided theoretical guidance for formulating regional green innovation coordinated development policy [14]. Gao Xiaotong et al. used the super epsilon-based measure (EBM) model that considers undesirable output to measure the GDE of China from 2000 to 2018 by constructing the evaluation index system of GDE. They used social network analysis (SNA) and a geographical detector to analyze the characteristics of the spatial correlation network characteristics and influencing factors. It was of great significance for realizing coordinated and sustainable development in China [15]. Yan Tao et al. used the super-efficient SBM model to study the urban efficiency and spatial and temporal evolution patterns of 285 prefecture-level cities in China [16]. Yang Yunfei and Qu Guifei measured Chinese marine eco-efficiency by using the DEA method and combined it with the Tobit model to explore the factors of marine eco-efficiency. Finally, they found that the structure of the marine industry and the investment of marine science and technology had significant effects on marine eco-efficiency [17]. Wang Yuan et al. used the VAR model to analyze the dynamic response mechanism of marine science and technology innovation and the high-quality development of the marine economy, and found that marine science and technology innovation plays a significant role in promoting the growth of the marine economy [18]. Hu Shengde and Li Zhonghua constructed the performance index evaluation system of marine ecological environment management based on DPSIR (Drive Pressure State Impact Response) and the TOPSIS model with entropy weight, and the influencing factors of governance performance were discussed [19]. Yinyin Wang, Conghua Xue, Huimin Li, and Kaizhi Yu, based on the SBM model, measured the marine economic efficiency of 11 coastal provinces between 2006 and 2017 in China. The Bayesian model averaging (BMA) method was used to analyze the impact factors of efficiency based on the indicator system constructed on the relationship between the Belt and Road and marine economic efficiency [20]. Guihai Yu, Deyan He, Wenlong Lin, and coworkers constructed a Chinese spatial economic network and explored the structure of the network from three aspects: the characteristics of the whole network, features of individual provinces in the network, and block model analysis based on the modified gravity model. The network analysis method was applied to analyze Chinese economic development over the past 20 years [21].

Most of the existing literature has paid much attention to marine economic outputs but has not fully considered the environmental losses. Therefore, this article adds environmental benefits as non-desired outputs into the indicator evaluation system, in order to be able to measure and evaluate the current status of marine eco-efficiency in China in an accurate manner. In the study of marine eco-efficiency impact factors, the existing literature mostly considers the formation mechanism and the traditional spatial econometric model only explores the spillover and benefit effects of marine eco-efficiency between neighboring provinces from the perspective of geographical adjacency. These articles ignore the existence of correlation relationships between non-neighboring provinces, and cannot explore the overall spatial correlation network structure of marine eco-efficiency. Furthermore, marine eco-efficiency is influenced by multiple factors, so the spatial network system gradually becomes complex and diversified, and it is difficult to measure accurately using traditional measurement methods. Social network analysis can avoid the influence of the multicollinearity of indicators, and study the structure of spatial network and influence factors from a respective correlation. Therefore, this article constructs a spatial correlation matrix with the modified gravity model, uses the social analysis method to explore the structure of the spatial correlation network of Chinese marine eco-efficiency, and uses the QAP method to analyze impact factors in order to provide a scientific reference for the sustainable development of the Chinese marine economy.

3. Marine Eco-Efficiency Measures and Spatial and Temporal Evolution

This chapter mainly measures the marine ecological efficiency of 11 coastal provinces in China during the research period by constructing an evaluation index system for marine ecological efficiency and using Matlab2020b software and the unexpected super-efficiency EBM model. ArcGIS software 10.2 is used to draw a spatial layout evolution map of marine ecological efficiency based on the marine ecological efficiency values every 4 years to analyze the spatial and temporal evolution process of 11 coastal provinces in China.

3.1. Research Subjects and Data Sources

The paper takes 2006–2018 as the study period and 11 coastal provinces (excluding Hong Kong SAR, Macao SAR, and Taiwan) in Tianjin, Hebei, Liaoning (Bohai Sea Rim), Shanghai, Jiangsu, Zhejiang (Yangtze River Delta region), Fujian, Shandong, Guangdong, Guangxi, and Hainan (Pan-Pearl River Delta region) as the decision-making units. The data on the relevant indicators are obtained from the China Statistical Yearbook, China Energy Statistical Yearbook, China Marine Economic Statistical Yearbook, China Environmental Statistical Yearbook, and the statistical bulletin of marine economy of coastal provinces from 2006 to 2019. The geographic spatial data are obtained from the “National Basic Geographic Information Center” and visualized with ArcGIS.

3.2. Marine Eco-Efficiency Input–Output Index System

In the paper, when measuring marine eco-efficiency, we adhere to the principle of “sea and land integration” and take into account the relationship between marine and land areas. We constructed a marine eco-efficiency evaluation system by learning from the German environmental economic account [22]. The specific indicators are shown in

Table 1. Study area marine eco-efficiency input–output index system.

Target Layer	Guideline Layer	Indicator Layer	Element Layer	Indicator Description
Investment	Resource Consumption	Labor Output	Number of People Involved in Maritime Employment	Reflect the Number of People Engaged in Marine Production
		Capital Output	Marine Fixed Capital Stock	Reflect the State of Development Infrastructure
		Energy Output	Total Energy Consumption in Coastal area	Reflect the Regional Energy Consumption Situation
		Land Consumption	Mariculture area	Reflect Input from Regional Marine Land Resources
Output	Expected Output	Total Marine Economic Development	Gross Marine Product	Reflect the Output of Marine Economy
	Unexpected Output	Environmental Pollution	Chemical oxygen demand and ammonia nitrogen emissions from industrial wastewater discharges in coastal area Emissions of sulfur dioxide and soot from industrial waste gas emissions in coastal area	Reflect Ecological Pollution

.

3.3. Spatial and Temporal Evolution and Variation Analysis of Marine Eco-Efficiency

The SBM model with unexpected outputs was first proposed by Tone [23]. This model takes into account the weighting issue of various input and output indicators and is unaffected by the units of input and output indicators. However, the model struggles to handle challenging radial and non-radial scenarios. The SBM model’s drawbacks in this area can be made up for by the EBM model. The EBM model can assess changes in input–output under the assumption of constant output, accurately evaluating efficiency. It is compatible with both radial and non-radial relaxation factors. Therefore, this article uses Matlab2020b software and an unexpected output super efficiency EBM model to measure marine ecological efficiency, and the results are shown in

Table 2. The Marine eco-efficiency values of Chinese coastal area during 2006–2018.

	2006	2008	2010	2012	2014	2016	2018	Mean Value
Tianjian	0.783	0.807	1.003	0.875	1.007	1.000	1.000	0.925
Hebei	0.516	0.544	0.379	0.512	0.580	0.424	0.444	0.486
Liaoning	0.413	0.397	0.382	0.398	0.378	0.284	0.239	0.356
Shanghai	1.584	1.533	1.475	1.031	1.009	1.173	1.152	1.279
Jiangsu	0.310	0.420	0.597	0.755	0.801	0.722	0.674	0.611
Zhejiang	0.333	0.391	0.479	0.539	0.512	0.487	0.447	0.455
Fujian	0.577	0.660	0.693	0.705	0.743	1.000	1.001	0.768
Shandong	0.451	0.517	0.536	0.566	0.645	0.602	0.572	0.555
Guangdong	0.537	0.606	0.681	0.728	0.751	0.688	0.642	0.662
Guangxi	0.176	0.169	0.171	0.216	0.260	0.247	0.242	0.211
Hainan	0.692	0.759	0.750	0.773	0.718	0.694	0.670	0.722
Average Value	0.579	0.619	0.650	0.645	0.673	0.665	0.644	0.579

Note: Due to page space limitations, only the ocean eco-efficiency values for even years are presented.

.

According to Table 2, the eco-efficiency of the 11 coastal provinces and cities in China varies more significantly and shows an overall disorderly state.

According to Table 2 and Figure 1, during the study period (2006–2018), the overall marine eco-efficiency values in Chinese coastal areas were not high. They had average efficiency values in the range of 0.5–0.7, and showed fluctuating changes locally and a fluctuating upward trend overall. This fluctuating change can be divided into the following three stages: 2006–2009 showed a fluctuating increase, 2009–2013 showed a fluctuating decrease, and in 2013–2018 Chinese marine eco-efficiency began a fluctuating increase. In the long term, there is still much room for improving the economic development, ecological protection, and resource utilization of Chinese coastal areas.

3.4. Trends in the Temporal Evolution of Marine Eco-Efficiency

Taking the average marine eco-efficiency of each coastal area as the standard and referring to the relevant literature [24], the marine eco-efficiency values are divided into four categories: high ecologically efficient areas [0.800, +∞), medium-high ecologically efficient areas [0.600, 0.800), medium-low ecologically efficient areas [0.400, 0.600), and low ecologically efficient areas [0.000, 0.400).

The evolution of the spatial layout is plotted every 4 years based on the Chinese marine eco-efficiency values measured in Table 2 for the years 2006–2018. This is shown in Figure 2.

As a whole, there were inter-provincial differences in Chinese marine eco-efficiency from 2006 to 2018, and there was an increasing trend in Chinese marine eco-efficiency over time. Shanghai remained at high eco-efficiency throughout the study period, while Guangxi was always at low eco-efficiency.

According to the formula, we calculated the coordinates of the center of gravity of Chinese marine eco-efficiency from 2006 to 2018 and plotted the migration trajectory of the center of gravity and the distance moved by the center of gravity. As shown in Figure 3 and Figure 4, the center of gravity of Chinese marine eco-efficiency maintained a displacement in the range of 117.186–117.574° E, 30.766–31.544° N from 2006 to 2018, with a total displacement of 49.67 km, including 46.37 km to the west and 17.79 km to the south. During the study period, the center of gravity of marine ecosystem efficiency in China showed certain phase characteristics. From 2006 to 2009, the center of gravity of efficiency showed an overall shift to the southwest, but in 2007, the center of gravity of efficiency shifted to the southeast, and continued to shift to the southwest after 2008. From 2009 to 2014, the center of gravity of marine ecosystem efficiency showed an overall shift to the northwest, among which the center of gravity in 2010 and 2011 showed a brief shift to the west, and continued to shift to the northwest after 2012. From 2009 to 2014, the center of gravity of marine eco-efficiency was shifted to the northwest, with a brief shift to the west in 2011, and continued to shift to the northwest after 2012. From 2014 to 2018, the center of gravity of marine eco-efficiency showed a continuous shift to the south until 2017, when it shifted to the north and returned to the vicinity of the center of gravity from 2016. The center of gravity of Chinese marine eco-efficiency moved from 117.574° E, 31.544° N in 2006 to 117.295° E, 31.140° N in 2018, with an overall larger displacement in the north–south direction and a smaller displacement in the east–west direction.

The displacement trajectory of the center of gravity of Chinese marine eco-efficiency is mainly divided into the migration of the center of gravity to the southwest from 2006 to 2009, the migration of the center of gravity to the northwest from 2009 to 2014, and the migration of the center of gravity to the southeast from 2014 to 2018. However, the center of gravity of Chinese marine eco-efficiency always concentrates on moving near the Yangtze River Delta region. ① The Yangtze River Delta region is located in the middle of Chinese coastline, with Shanghai as the core. The leader of the development of the marine economy in the country, Shanghai relies on the economic base and technological innovation of marine intelligent manufacturing, and the rise of modern services in the ocean to promote the synergistic development of the regional marine economy. Thus, the Yangtze River Delta marine eco-efficiency has a strong pull-up effect. ② From 2006 to 2009, the center of gravity of marine eco-efficiency showed a southwestward migration. The reason for this is that marine eco-efficiency in the Pan-Pearl River Delta region has a spatial pulling effect on the whole country. ③ From 2009 to 2014, the center of gravity of marine eco-efficiency showed a northwestward migration. In this period, the northern Bohai Sea region began to develop marine transportation, coastal tourism, and other industries. In the early days, the management of pollution problems on the bases of the old industries was beginning to bear fruit, and the ecological environment had improved. At the same time, the marine eco-efficiency of Tianjin also has a benign radiating effect on the surrounding areas. ④ The center of gravity of marine eco-efficiency from 2014 to 2018 showed a southeastward migration. With ocean power proposed and the concept of sustainable development becoming popular, the Yangtze River Delta region as well as the Pan-Pearl River Delta region began to shift from the pursuit of a high-speed marine economy to a high-quality marine economy.

4. Structural Characteristics of the Marine Eco-Efficiency Spatial Association Network

A modified gravity model is constructed to obtain the correlation matrix of marine eco-efficiency, and then a social network analysis is applied to study and analyze the network characteristics of the spatial correlation matrix of marine eco-efficiency in China from two perspectives: overall network and individual network.

4.1. Construction of Spatial Correlations of Marine Eco-Efficiency

The construction of a spatial association matrix is the basis for applying social network analysis. The gravity model is a classic model for determining spatial association relationships, which can accurately portray the network structure dynamics and build a comprehensive index evaluation system with various influencing factors such as environment, economy, and geography. For this, we refer to the research method of Zhao Lin, Liu Jia, et al. [25,26] to determine Chinese spatial correlations of marine eco-efficiency by constructing a modified gravity model. The modified gravitational model equation is as follows:

$$Y_{ij}=K_{ij}\times \frac{C_i\times C_j}{D_{ij}/{(g_i-g_j)}^2},K_{ij}=\frac{C_i}{C_i+C_j}$$

(1)

In the formula, i and j denote the provinces and cities along the coast of China; Y_ij denotes the linkage strength of marine eco-efficiency between provinces; K_ij represents the contribution rate in the linkage of marine eco-efficiency between provinces; C_i and C_j represent the marine eco-efficiency value of provinces; D_ij is the spherical distance between provincial capital cities of provinces; and g_i − g_j is the economic distance between provinces. According to the above formula, we use MATLAB software to calculate the inter-provincial linkage intensity matrix of marine eco-efficiency. The mean value of each row of the matrix is used as the critical value. If Y is greater than the critical value, the value is 1, which means that there is a relationship between the marine eco-efficiency of the 2 provinces; otherwise, the value is 0, which means that there is no relationship between the marine eco-efficiency of the 2 provinces. The spatial correlation matrix S of marine eco-efficiency in China is obtained by the above conversion.

4.2. Marine Ecological Spatial Linkage Network Analysis

The meaning of social network analysis is to quantify the correlations and accurately reflect the links of marine eco-efficiency in each province [27]. Among them, the overall spatial network characteristics are measured by network density, network efficiency, and network hierarchy metrics to reveal the overall correlation characteristics of the spatial network [26]. Individual spatial network characteristics are measured by point-out, point-in, point-centrality, near-centrality, and intermediate centrality metrics to reveal the position of each provincial node in the spatial network [28]. Therefore, we use UNCINET and ArcGIS software to analyze both the overall network and the individual network to explore the connections between the nodes of spatial networks.

4.2.1. Overall Network Feature Measurement Method

(1)

Network Density

Network density reflects the sparseness of the connections among nodes in the spatial network, and a larger value indicates more connected members in the network; that is, there are frequent interactions and close relationships among the marine eco-efficiency of such Chinese provinces. The expression of network density [29,30] is as follows:

$$D=\frac{A}{m\times (m-1)}$$

(2)

In the formula, m denotes the number of nodes in the network, and A represents the number of association relationships that actually exist in the spatial network.

(2)

Network Efficiency

Network efficiency reflects the efficiency of connections among nodes in a spatial network; that is, the number of redundant lines present in the network. If the network efficiency is higher, then it indicates that there are fewer redundant lines among the provinces of Chinese oceans, a less self-centered association among the nodes, and a lack of effective collaboration. A lower network efficiency indicates a rich network of spillover channels, obvious multiple superimposed spillover effects among provinces, and a more stable spatial network of marine eco-efficiency. The expression of network efficiency [30] is as follows:

$$E=1-\frac{l}{\max (l)}$$

(3)

In the formula, l indicates the number of redundant lines, and max(l) represents the maximum number of symmetrical possible pairs of points in the overall network.

(3)

Hierarchy of Network

The hierarchical degree of the network measures the degree of asymmetric accessibility among provinces in the spatial network and reflects the control capability of each node in the network in which it is located. A higher value of the rank degree of the network indicates that a few provinces in the Chinese marine eco-efficiency spatial network are the dominant players in the network, and most provinces are in the marginal positions in the network. On the contrary, a lower value of the network hierarchy degree indicates that the rigid hierarchical structure of the spatial network is broken, and the provinces interact with each other frequently and have close relationships. The expression of hierarchy in the network [30] is as follows:

$$H=1-\frac{k}{\max (k)}$$

(4)

In the formula, H indicates the hierarchy of the network, which takes values in the range [0, 1]; k indicates a symmetric reachable point logarithm; and max(k) indicates the maximum possible point of symmetry up to the logarithm.

4.2.2. Individual Network Feature Measurement Methods

(1)

Degree Centrality

Degree centrality measures the number of links to a point in the spatial network; that is, if a node has more direct contact with other nodes, then the node is closer to the center. The larger the value of degree centrality, the more contact the node province has with other nodes and the more “energy” it has in the network. The expression of degree centrality is as follows:

$$D_e=\frac{n}{N-1}$$

(5)

In the formula, n denotes the number of directly connected provinces; N indicates the most possible number of directly connected provinces.

(2)

Proximity Centrality

The proximity centrality depicts the sum of the shortest paths between this node and other nodes. For the marine eco-efficiency spatial association network, the higher the proximity centrality value, the closer the province is to other provinces, and the more associations are likely to exist. The expression of proximity centrality is as follows:

$$C_i={\displaystyle \sum _{i=1}^n{d}_{ij}}$$

(6)

In the formula, d_ij denotes the shortest shortcut distance between the capital cities of provinces i and j.

(3)

Intermediate Centrality

Intermediate centrality measures the degree of dominance of nodes in the network over elements such as information resources. A larger value indicates that the node, if occupying more positions, has a greater dominant role over the other elements. It is in a dominant position in the network, and other nodes need to pass through it to establish connections. The expression of intermediate centrality is as follows:

$$C{B}_a=\frac{2{\displaystyle {\sum }_b^n{\displaystyle {\sum }_k^n{B}_{bk}(i)}}}{N^2-3N+2},B_{bk}(a)=\frac{g_{bk}(a)}{g_{bk}}$$

(7)

In the formula, B_bk(a) denotes the degree of action of a on b, k linkage; and g_bk indicates the number of shortcuts between b and k.

4.3. The Evolution of the Spatial-Related Network Structure of Marine Eco-Efficiency

4.3.1. The Overall Network Characteristics Analysis

Based on the modified gravity model to construct the spatial correlation matrix of marine eco-efficiency, the cross-sectional data of 2006, 2010, 2014, and 2018 are selected at 4-year intervals to draw the spatial correlation network map of marine eco-efficiency (Figure 5). Then, the Ucinet6.0 software is used to calculate the four index values of network relationship number, network density, network rank degree, and network efficiency to reveal the overall network evolutionary characteristics of marine eco-efficiency in China.

As shown in Figure 5, from 2006 to 2018, the overall network of marine ecological spatial associations in China showed an increasingly complex trend. The non-contiguous provinces break through the traditional spatial location and produce cross-regional linkage effects, gradually forming the spatial structure characteristics of pole–nuclear diffusion, and the Yangtze River Delta region is more spatially connected than the Yellow Bohai Sea region and the Pan-Pearl River Delta region. After 2018, the spatially linked network of marine eco-efficiency in China has significantly increased the number of nodes connected, there are more spillover channels in the network, and the overall more rigid hierarchical structure has been broken, to some extent eliminating the dominant position of a few provinces in the network. The cost of transmission of marine eco-efficiency among provinces is reduced, and the network structure trends to a steady state. The continuous improvement of spatial network structure can not only promote the effective improvement of the eco-efficiency of each coastal area but also promote the development of inter-provincial ecological networks to realize the sustainable development of the marine economy and the ecological society of green water and green mountains.

4.3.2. The Individual Network Characteristics Analysis

The Ucinet6.0 software is used to calculate five index values of point-in degree, point-out degree, degree centrality, intermediate centrality, and near centrality for each province and city in the coastal region of China. In order to reveal the centrality characteristics of the associated individuals in the spatial association network in each province and city during the study period, the network centrality analysis of marine eco-efficiency spatial association is shown in

Table 3. Evolution of Carbon Emission Network Points in and out of China’s Tourism Industry and Degree Centrality, 2006–2018.

Area	Degree of Point Entry/Degree of Point out				Degree of Center
	2006	2010	2014	2018	2006	2010	2014	2018
Tianjian	2/3	2/5	2/7	2/3	30.000	50.000	70.000	30.000
Hebei	2/1	1/1	1/5	1/4	20.000	10.000	50.000	40.000
Liaoning	2/0	2/0	3/0	3/0	20.000	20.000	30.000	30.000
Shanghai	3/9	3/8	5/7	5/8	90.000	80.000	90.000	90.000
Jiangsu	1/1	1/1	4/4	4/7	10.000	10.000	70.000	80.000
Zhejiang	1/1	1/1	1/2	2/2	10.000	10.000	20.000	20.000
Fujian	1/1	1/1	3/1	3/3	10.000	10.000	30.000	40.000
Shandong	2/1	2/1	2/3	4/3	20.000	20.000	40.000	40.000
Guangdong	3/2	3/2	4/2	3/2	30.000	30.000	40.000	30.000
Guangxi	2/1	3/1	4/1	4/1	20.000	30.000	40.000	40.000
Hainan	2/1	3/1	4/1	4/2	20.000	30.000	40.000	40.000
Average Value	1.9/1.9	2/2	3/3	3.2/3.2	25.455	27.273	47.273	43.636

and

Table 4. Evolution of China’s tourism carbon emission network towards and intermediate centrality, 2006–2018.

Area	Proximity Centrality				Intermediate Centrality
	2006	2010	2014	2018	2006	2010	2014	2018
Tianjian	15.022	18.576	41.270	35.886	4.444	15.556	5.000	9.630
Hebei	15.357	16.993	38.828	42.120	2.222	0.000	1.296	3.889
Liaoning	15.657	15.657	15.909	26.285	0.000	0.000	0.000	0.000
Shanghai	51.705	44.712	22.304	56.980	30.000	30.000	20.370	43.148
Jiangsu	31.098	28.825	21.244	51.844	0.000	0.000	3.889	11.296
Zhejiang	31.098	28.825	20.040	38.516	0.000	0.000	0.000	0.000
Fujian	31.098	28.825	20.304	42.857	0.000	0.000	0.741	23.889
Shandong	15.357	18.188	34.968	45.541	2.222	8.889	1.296	19.444
Guangdong	14.990	23.413	25.556	33.225	2.222	2.222	2.222	10.556
Guangxi	14.754	24.726	25.495	30.859	0.000	1.667	0.370	0.741
Hainan	14.754	24.726	25.495	37.750	0.000	1.667	0.370	18.519
Average Value	22.808	24.860	26.492	40.169	3.737	5.455	3.232	12.828

.

According to Table 3 the following results are shown: (1) The relationship between spillover and benefit. The point-out degrees of Shanghai and Tianjin provinces were always greater than the point-in degrees during the study cycle, indicating that these regions are mostly spilling over to other regions in the marine eco-efficiency linkage network and have significant spillover effects on other surrounding regions. The point-out degrees of provinces and cities such as Liaoning, Guangxi, and Hainan are always smaller than the point-in degrees, which indicates that the beneficial relationship of these regions in the association network is always larger than the spillover relationship, forming a beneficial effect on other surrounding regions. (2) Degree Centrality. Overall, the centrality of each node of the spatially linked network of marine eco-efficiency in China from 2006 to 2018 shows an upward trend, and the degree of clustering of network linkages decreases. In 2018, the degree centrality of Shanghai and Jiangsu is obviously higher than the average value. The marine industry has a reasonable structure, a high utilization rate of resources, and a first-mover advantage in environmental pollution management, so it is in the center of the spatial association network and more likely to have an association relationship with other nodes. By analyzing the changes in the number centrality from 2006 to 2018, it can be seen that the degree centrality of Zhejiang, Fujian, and Hainan has increased significantly, and these provinces receive the eco-efficiency spillover from Shanghai–Jiangsu in the Yangtze River Delta region, increase interaction and exchange with other provinces, and gradually improve their position in the spatial association network.

According to Table 4, the following results are indicated: (1) Proximity Centrality. The proximity centrality of each province shows a fluctuating upward trend during the period 2006–2018. This indicates that the channels of interaction between nodes in the spatially linked network for marine eco-efficiency have increased and the links are becoming stronger. Among them, the provinces of Shanghai, Jiangsu, Fujian, and Zhejiang have ranked ahead, close to the middle degree at a high level, so they are in a dominant position in the spatial association network, and can play a better role in radiation between regions. (2) Intermediate Centrality. The intermediate centrality of Shanghai and Tianjin is always higher than the average value, and they have the ability to dominate and control the elements in the spatial network of marine eco-efficiency and can promote other provinces to increase their connection. Shanghai and Tianjin can play the role of bridge between the north and the south by taking advantage of their geographical locations. It should be noted that the intermediate center of Guangdong has not reached the average value in a few years, but still can play the role of “bridge”, because its surrounding provinces, such as Guangxi and Hainan, have a poor marine economic base and are relatively remote. They have long been in a dominant position in the spatial correlation network by receiving spillover from Guangdong and other places.

4.4. The Impact Factors of Marine Eco-Efficiency

During the study of the measurement and spatio-temporal evolution of marine eco-efficiency in China, significant differences were observed among regions. Therefore, we need to further study the impact factors of eco-efficiency in order to identify the key factors that promote or inhibit Chinese marine eco-efficiency, so as to provide scientific and effective theoretical references for the high-quality and sustainable development of Chinese marine economy in the future. Learning from previous research [31,32,33,34], and considering the characteristics of marine eco-efficiency comprehensively, suitable influencing factors were screened to construct the social network and regression econometric models. The specific indicators and analysis results are shown in

Table 5. Description of indicators of spatial correlations of marine eco-efficiency impact factors.

Variable Type	Variable Name	Variable Symbols
Spatial aAjacency	Spatial Adjacency Matrix	D
Economic Development Level	Gross Domestic Product	GDP
External Opening Level	Total Import and Export	OPEN
Population Distribution Level	Population Density	PD
Marine Industry Structure	Ratio of Marine Tertiary to Secondary Production	MIS
Marine Science and Technology Level	Marine Researchers	MT

and

Table 6. Results of QAP regression analysis of factors influencing spatial correlations of marine eco-efficiency.

Variable Name	Non-Standardized Regression Coefficient	Standardization Regression coefficient	Significance Probability Value	Probability 1	Probability 2
Intercept term	−56.169	0.000	-	-	-
D	509.033	0.289	0.002	0.002	0.998
GDP	0.007	0.243	0.032	0.032	0.968
OPEN	−0.063	−0.237	0.014	0.986	0.014
PD	0.306	0.530	0.010	0.010	0.990
MIS	−129.039	−0.100	0.096	0.904	0.096
MT	−0.050	−0.103	0.139	0.861	0.139

.

• Spatial Adjacency Matrix (D)

The spatial adjacency relationship passed the 1% significance level test, and the regression coefficient was positive. This indicates that geographic adjacency plays a significant role in the formation of the marine eco-efficiency spatial association network and has a positive contribution.

2.

Economic Development Level (GDP)

The economic development level passed the 5% significance level test, and the regression coefficient was positive, indicating that the greater the gap between economic development levels among provinces, the stronger their marine eco-efficiency linkages will be. Therefore, the disparity of economic development levels between regions has a significant effect on the spatial correlation network of marine eco-efficiency.

3.

External Opening Level (OPEN)

The level of external openness passed the 5% significance test, and the regression coefficient was negative, indicating that the difference in the level of external openness has a negative inhibitory effect on the formation of spatially linked networks of marine eco-efficiency.

4.

Population Distribution Level (PD)

The population distribution levels passed the 1% significance test, and the regression coefficients were positive, indicating that the differences in population distribution levels positively contribute to the marine eco-efficiency spatial association network.

5.

Marine Industrial Structure (MIS)

The marine industry structure passed the 10% significance test, and the regression coefficient was negative; that is, the smaller the difference in marine structure, the stronger the spatial correlation and spillover effect of marine eco-efficiency in coastal provinces can be.

6.

Marine Science and Technology Level (MT)

The level of marine science and technology failed the significance level test, and the regression coefficient was negative. This is either because of the late training of marine science and technology personnel or the lag in science and technology literacy among science and technology personnel, so its effect on the spatial correlation network of marine eco-efficiency was weak.

5. Research Conclusions and Recommendations

5.1. Research Conclusions

Against the backdrop of the strategy of becoming a maritime power and achieving sustainable development, the healthy and sustainable development of China’s marine economy is particularly important. At the same time, with the continuous development of the marine economy, the activities of the marine economy are gradually shifting towards land, and the cost of treating the pollutants generated will also be greatly added to the marine economy. Therefore, when considering marine economic issues, it is necessary to fully consider “land–sea coordination”. This article constructs investment indicators from the perspectives of capital investment, human investment, energy investment, and land consumption, and incorporates environmental pollution as an unexpected output into the indicator system. The unexpected super efficiency EBM model is used to measure the marine ecological efficiency values of 11 coastal provinces and cities in China from 2006 to 2018, and the overall trend of marine ecological efficiency in China during the research period is described. At the same time, ArcGIS software is used to study and analyze the spatiotemporal evolution pattern of marine ecological efficiency. While elaborating on marine ecological efficiency from the time dimension, the spatial dimension is introduced to provide a more comprehensive explanation of marine ecological efficiency. In order to further study and analyze the spatial correlation network of marine ecological efficiency and its influencing factors, firstly, the spatial incidence matrix is constructed. This is created through the modified Gravity model, and its network structure is depicted at the overall and individual levels, respectively. Secondly, from the perspective of relationships, the indicator difference matrix is constructed to empirically analyze the specific influencing factors of marine ecological efficiency using QAP regression analysis, which can avoid the influence of multicollinearity. The main conclusions are as follows:

• There are significant differences in marine ecological efficiency between regions.

In terms of time, the level of marine eco-efficiency in most provinces was not high and varied significantly between provinces during the study period. However, on the whole, it showed a fluctuating upward trend. In terms of space, the displacement trajectory of the center of gravity of China’s marine eco-efficiency was mainly divided into the migration of the center of gravity to the southwest from 2006 to 2009, the migration of the center of gravity to the northwest from 2009 to 2014, and the migration of the center of gravity to the southeast from 2014 to 2018, but the center of gravity of China’s marine eco-efficiency was always concentrated near the Yangtze River Delta region. This phenomenon is mainly because the marine eco-efficiency of the Yangtze River Delta region has a spatial pull effect on the whole country and can have a radiation-driven effect on the surrounding areas. Since the areas of marine economic development are mainly concentrated in the Yangtze River Delta region, the center of gravity of China’s marine eco-efficiency has always been concentrated near the Yangtze River Delta region.

2.

The network structure of marine ecological space needs to be optimized.

In terms of the overall network characteristics, from 2006 to 2018, China’s marine ecological spatial association network as a whole showed an increasingly complex trend, and the degree of spatial association in the Yangtze River Delta region was higher than that in the Yellow and Bohai Sea region and the Pan-Pearl River Delta region. The number of associated relationships in the national marine ecological spatial network increased from 21 in 2006 to 35 in 2018, while the network density increased from 0.191 to 0.318. Correspondingly, the frequency of interaction between the relationship nodes of the spatial network increased, and the net framework gradually matured. However, inter-province differences in the spatial linkage strength of marine eco-efficiency still exist, and the network structure needs to be further optimized. From the perspective of individual network characteristics, provinces and cities such as Liaoning, Guangxi, and Hainan have been benefiting more than spilling over in the correlation network, forming a beneficial effect on other surrounding areas. Shanghai and Jiangsu have a reasonable marine industry structure, a high resource utilization rate, and a first-mover advantage in environmental pollution control, so they are at the center of the spatially linked network and are more likely to have linkages with other nodes. Zhejiang, Fujian, and Hainan are gradually rising in the spatially linked network because they receive eco-efficiency spillover from Shanghai and Jiangsu.

3.

Various factors affect marine ecological efficiency.

QAP regression analysis shows that the spatial adjacency matrix, differences in economic development levels, differences in population distribution levels, and the marine eco-efficiency spatial correlation matrix are significantly correlated. Furthermore, the regression coefficient is positive, showing a facilitating effect. The greater the differences in their influencing factors, the stronger the marine eco-efficiency linkage will be. The difference in the level of openness to the outside world and the difference in the structure of the marine industry is also significantly correlated with the correlation network, but the regression coefficient is negative, which shows that the larger the difference is, the more it inhibits the formation and development of the spatial network. The spatial correlation of marine science and technology level differences in marine eco-efficiency is weak, probably due to the late training of marine science and technology personnel and the lag in science and technology literacy of science and technology personnel.

5.2. Recommendations

From the measurement of marine ecological efficiency, the exploration of spatial network structure, and its influencing factors, it is revealed that the overall value of marine ecological efficiency in China is low, and there are obvious differences among provinces. The difference matrix of influencing factors plays a significant role in the formation and development of the network, and the improvement of marine ecological efficiency has a long way to go. Based on the above conclusions, the following suggestions are proposed:

• Take advantage of the location and promote the sustainable development of the marine economy

The spatially linked network of marine eco-efficiency in China is showing an increasingly complex trend, so it is necessary to analyze the actual situation of coastal provinces from a relational perspective, promote the synergistic plan of marine industries, and personnel, and at the same time, effectively enhance the regional collaborative effect of marine eco-efficiency based on the centrality characteristics of each regional network. The Yangtze River Delta region, with Shanghai as its core, is in a dominant position in the spatial network. The development of its marine economy has a radiative spillover effect on other members of the spatial network. Therefore, it should take advantage of its unique geographic advantages, modify the structure of the marine industry, propel the general growth of the marine economy in the spatial network, and promote the marine economy’s sustainable development.

2.

Promote regional collaboration and accelerate the development of green ocean economy

The key impact factors of marine eco-efficiency correlations are considered in an integrated manner, focusing on the “two-legged approach” and giving full play to the market’s decisive role in resource allocation while giving full play to the government’s macro-regulatory role. The government should encourage provinces with lower levels of marine technology to introduce high-tech enterprises and marine laboratories and provide certain policy incentives to improve marine ecological efficiency through technology. At the same time, establishing a reliable system for monitoring the marine environment and legal protection, tightening controls on businesses that pollute excessively, and effectively coordinating the relationship between marine economic development and environmental protection will all help to hasten the creation of a green and sustainable marine ecological environment and advance the growth of a green marine economy. Additionally, it is critical to foster technical collaboration and exchange across the different provinces and cities in the spatial network, foster regional cooperation, and gradually raise marine ecological productivity.

3.

Optimize the industrial structure and promote the formation and development of spatial networks

According to the results of the QAP regression analysis, it can be concluded that there is a significant relationship between marine industrial structure differences through correlation network regression. Therefore, in the spatial association network, cooperation between provinces and cities should be strengthened, regional cooperation such as technological exchange should be strengthened, and market mechanisms should be used to promote the transfer of marine industrial structures. While optimizing the structure of the marine industry, narrowing the differences in marine economic development between each other can strengthen the relationships in the spatial correlation network. In addition, it is necessary to continue to strengthen the efforts of coastal cities to open up to the outside world, drawing on the development experience of countries with strong marine economic development. They can adopt appropriate development policies based on their own development situation, improve the marine ecological efficiency of coastal cities, promote the formation and development of marine spatial networks, and contribute to the sustainable development of the marine economy.

The impact of marine eco-efficiency is a very complex issue, and there may have been some errors in selecting the more critical six influencing factors for the study. In reality, there must be other influencing factors that have an impact on marine eco-efficiency. Although the measurement results of marine eco-efficiency have been analyzed in detail from the perspective of time and space, the differences that exist between efficiency in provinces have not been explored in depth, which is the direction of further in-depth research on this topic in the future.