Research

INTRODUCTION

The harmonious development of humans and nature has been a core principle and ultimate goal of the development of human society for a long time. However, loss and destruction of the ecological environment due to human activities and constraints of nature on the development of human economic society have become important challenges to achieving this goal. With this background, the concept of a coupled human and natural system, which focuses on the interaction between humans and nature, has been established. Scholars have carried out many studies on this topic from different perspectives and on the basis of different mechanisms (Gunderson et al. 2002, Dietz et al. 2003, Liu et al. 2007). With the enhancement of human activities, natural and social systems have inter-infiltrated, and a pure, natural ecosystem has been replaced by a natural-socio-economic coupling social-ecological system. Therefore, the concept of a social-ecological system (SES) was proposed. SES emphasizes the interdependence and interaction between human society and the natural ecological environment on which it depends (Walker and Salt 2006), which is also known as the “complex human-land system” or “combined system between human and nature.” It refers to the coupling system formed through the interaction between humans and the environment, with characteristics of complexity, nonlinearity, uncertainty, multi-level nesting, etc. SES focuses on the interaction between humans and environmental systems as well as the dynamics and feedback formed on multiple correlation scales, providing a new perspective for humans to solve social and ecological problems.

Proposed by Ostrom (2009), SES has been gradually adopted as a prevalent framework for analyzing sustainability and has been gradually deepened (Guimarães et al. 2013, Lockwood et al. 2014, Chaffin and Gunderson 2016, Virapongse et al. 2016, Mao et al. 2021, Bogert et al. 2022). There are also studies evaluating the quality of SES. Among numerous issues involved in SES, evaluation of stability and vulnerability is the core issue and has been a research hotspot for the sustainable development of SES. The World Bank and the United Nations have conducted many SES vulnerability and adaptability assessments (He et al. 2021), and many scholars have also conducted relevant research. The empirical research about SES was generally carried out on two spatial scales through place-based and system-scale modeling (Naylor et al. 2019). In research on the first scale, there has been not only in-depth research on relatively small scales, such as provincial or smaller administrative regions in China (Naylor et al. 2019, Chien and Saito 2021, He et al. 2021, Jiang et al. 2021, Mafi-Gholami et al. 2021), but also research on a large scale (the country level; He et al. 2018) or an evaluation of SES vulnerability of cities or areas in a country with the same characteristics, such as coal mining cities and marine-protected areas (Tai et al. 2020, Noble et al. 2021). Regarding the vulnerability assessment on the system scale, important research objects have included industries such as fishery and tourism (Thiault et al. 2017, Silva et al. 2019, Lazzari et al. 2021, Kasperski et al. 2021) and biomass (e.g., “biochar”; Müller et al. 2019), the Moorea Coral Reef (Thiault et al. 2018), and even mountains (Gardner and Dekens 2007, Clark et al. 2019). Apart from the main research directions mentioned above, some scholars have also noticed a correlation of the regional system of SES and have researched the vulnerability of the economic-environmental system and the evolution of the relationship between humans and the environment (Liu et al. 2020, Perry et al. 2010). When recognizing the dynamic coordination relationship between the economy and environment, studies tend to apply a Lotka-Volterra model for the empirical test (Pao et al. 2015, Xu et al. 2019, Xing et al. 2021, Guo et al. 2022, Yuan et al. 2022).

Existing studies have conducted vulnerability assessments of the social-ecosystem from different perspectives and have used different methods. However, there is still space for additional research, which is prominently manifested in the following two aspects. First, existing studies mostly analyzed the overall vulnerability of SES from the perspective of the exposure, sensitivity, and adaptability of the overall system but ignored effects of different subsystems within a system and their interactions in the overall system. Second, in existing studies, a multi-index weighting method was used to conduct the evaluation, focusing more on a perspective of absolute and variation quantities and paying less attention to the vulnerability evaluation from the perspective of quality.

Based on the research foundation mentioned above, in this study we aim to conduct an in-depth analysis of the vulnerability of SES. Using a place-based scale proposed by Naylor et al. (2019), a new interdisciplinary, efficiency-based SES coupling coordination degree and an overall system vulnerability evaluation concept are proposed. The proposed vulnerability evaluation concept takes into account not only the economic needs of SES but also its ecological basis. By using 290 cities at the prefecture level or above in mainland China as research objects, an empirical analysis is conducted to determine the vulnerability of SES in China from 2015 to 2019 from an interdisciplinary perspective of ecology and management science for the first time.

The main innovations and research contributions of this study can be categorized into three elements. (1) The vulnerability of SES is analyzed from the perspective of subsystems and their interactions. Starting from the two important dimensions of SES (economic society and ecological environment), we consider not only the development levels of the economic and social subsystem and the ecological environment subsystem but also the interactions between subsystems in economic production and environmental governance processes. (2) An interdisciplinary method of ecology and management is conducted for SES evaluation. Specifically, in this study, the development quality of the subsystems is measured with a DEA (data envelopment analysis) model, and the symbiotic relationship between the two subsystems is analyzed using a Lotka-Volterra model for niche overlap of subjects, providing a new way of evaluating and analyzing the vulnerability of SES. (3) The vulnerability of SES in multiple cities is investigated, thereby laying a foundation to further explore the correlation of the regional system and the spatial impact of SES vulnerability.

The structure of the remaining sections of this paper is as follows. Section two is the methodology and model building, including SES synergy and competition model, SES subsystem measurement model, and the descriptive statistical analysis of the data used. Section three is the empirical analysis, involving the level of development of the SES subsystem, the interaction between subsystems, and SES vulnerability. Section four concludes the study and proposes further research prospects.

METHODS

In the present study, a multi-dimensional, analytical model for SES co-evolution analysis is built. Given that SES mainly measures the interaction between humans and the environmental system, the economic society and ecological environment are two important dimensions. Activities of the economic and social subsystem are aimed at pursuing economic growth, and the ecological environment subsystem is used to measure services that an ecosystem can provide for human activities. According to the literature review above, most of the existing studies ignored the impact of various subsystems within a system and their interactions in the overall system. Therefore, in this study, SES is divided into an economic and social subsystem and an ecological environment subsystem. Regarding relevant literature and the specific technical path, the internal synergy of the SES subsystem is first analyzed by using a Lotka-Volterra model, and the economic and social subsystem and the ecological environmental subsystem are considered as two aspects of the overall ecosystem of SES for discussion and investigation. Then the development level of various subsystems of SES in China is measured, the coupled and coordinated development degrees are identified, and the stability and vulnerability of SES are evaluated.

Synergy and competition in the subsystems

Existing literature proves that the research and application of a Lotka-Volterra model in the sustainable and green development of the socio-economy and ecological environment are reasonable and feasible. Therefore, this study also applies a Lotka-Volterra model to investigate synergy and competition in social and ecological subsystems. For a subject in the ecological system, the niche evolution process that considers only the subject can be expressed as (Lotka 1925, Volterra 1926):

(1)

where h denotes the niche width of O; v denotes the growth rate of the niche; thus, vh is the natural growth capacity of the niche of O, and vh*h/H is the inhibitory effect resulting from the growth of the niche. Under these two forces, the niche of O eventually approximates the maximum niche width H that can be achieved by natural growth.

With a representing green economy and b representing green environment, A and B denote the quality levels of a and b, respectively; v_A and v_B are the corresponding horizontal growth rates, respectively. Therefore, the synergy and competition relationships between the social and ecological subsystems are as shown in Equation 2.

(2)

where θ_A denotes the influence coefficient of the ecological subsystem on the social subsystem, that is, space H_A occupied by each individual. B is equivalent to the space occupied by θ_A individuals of A; when θ_A<0, it indicates that the ecological subsystem has a promoting effect on the social subsystem; when θ_A>0, it indicates that the ecological subsystem has an inhibitory effect on the development of the social subsystem. θ_B denotes the influence coefficient of the social subsystem on the ecological subsystem; its meaning is similar to that of θ_A.

Furthermore, the symbiotic coordination degree C between the social and ecological subsystems can be obtained on the basis of θ_A and θ_B (Eq. 2) as follows:

(3)

The ecological niche growth of a single subject is an ideal condition. However, in real life, niche overlaps of different subjects in the ecosystem are inevitable because of finite resources and the similarity of the living environment, which affects each other’s niche growth, and for this reason a Lotka-Volterra model was built. As two important subsystems of SES, the social subsystem is aimed at pursuing economic growth, consuming resources in production, and producing economic benefits, but has a negative impact on the environment, whereas the ecological subsystem is aimed at improving the ecological environment through environmental governance and reducing resource utilization in economic production activities and the amount of pollution emissions. At the macro level, administrative units (which are cities in this study) are subjects not only of economic production but also of environmental governance, and the overall resources that can be used in economic production and environmental governance are limited. In an ideal case of a long-term economic and social development process, the social and ecological subsystems can form a dynamic cycle and achieve positive interaction (Fig. 1). Therefore, social and ecological subsystems are regarded as two important aspects of SES, and a niche overlap may occur in their niche growth, which confirms the Lotka-Volterra model.

There are three interaction relations among species: positive, neutral, and negative. The relations between the economic and social subsystem and the ecological environment subsystem and their degree of symbiotic coordination are analyzed by using a Lotka-Volterra model to determine their coordinated development in the SES of each city. Stability of the overall system needs to be achieved through mutualism of the two subsystems. Two subsystems in a commensalism state should develop toward mutualism, and cities in an amensalism state should also develop into a commensalism state and then eventually achieve mutualism. In a competitive state, the economic and social subsystem and the ecological environment subsystem restrict each other. Once in this state, no matter how each subsystem develops and how its level improves, the relationship between them will always be unstable, eventually resulting in an extremely vulnerable state of the SES (Pelling and High 2005).

Level of the subsystems

As a complex and open system, the SES dissipative structure generates internal and external energy and exchanges information because of the difference in potential energy. Thus, it can be regarded as an “input-output” system (Cannon and Müller-Mahn 2010). Furthermore, the social and ecological subsystems also appear as two interrelated input-output processes, and balanced development of the social and ecological subsystems needs to be considered for each decision-making unit (DMU). The social subsystem is mainly reflected at the production stage, in which economic growth is taken as the development goal, and the main body of production consumes production factors to derive economic output. The ecological subsystem is mainly at the governance stage. At this stage, improvement of environmental quality is the development goal, and subjects engage in environmental governance and undertake investment to improve environmental quality. Here, the DEA method is used to measure the development level of the green economy in each city. Because the traditional model cannot further distinguish DMUs with an efficiency value of 1, in the current study, a cross-efficiency model is employed to measure the efficiency of the social and ecological subsystems of each city and improve the accuracy of measuring development levels of a city’s green economy and green environment. Furthermore, by referring to the existing literature (Guo et al. 2019, Liu et al. 2020, Shuai and Fan 2020, Liu and Dong 2021), an input-output indicator system for the social and ecological subsystems is constructed in this study (Fig. 2).

The meaning and the data sources of the indicators are as follows:

• Labor: the number of people engaged in a production process in a tested year, measured by the number of employed persons (EP). Labor data are derived from China Population and Employment Statistics Yearbook and China Stock Market and Accounting Research Database.


• Capital: the amount of capital invested in a production process, measured by the total amount of capital stock (CS). Due to the long-term impact of fixed assets, the data are converted from the total amount of investment in fixed assets using the perpetual inventory method. The data for the total amount of investment in fixed assets are derived from China Stock Market and Accounting Research Database and provincial statistical yearbooks.


• Resource: the resources consumed in a production process are represented by (a) the total amount of electricity consumption (EC) and (b) the total amount of water consumption (WC). The data source is the same as that of capital.


• Economic output: the output obtained by economic production activities using the above-mentioned inputs, measured by the gross domestic product (GDP). The data are from the China City Statistical Yearbook.


• Investment in pollution control: the investment in the treatment of environmental pollution is measured by the total amount of investment in pollution control (IPC). Because the data are only disclosed at the provincial level, the data of each city are allocated according to the proportion of each city’s GDP in the province’s GDP. The provincial investment data are from the China Statistical Yearbook on Environment.


• Investment in infrastructure construction: the investment in environmental infrastructure for the improvement of environmental quality is measured by the total amount of investment in infrastructure construction (IIC). The data processing and source are the same as those of investment in pollution control.


• Air quality: the degree of air cleanliness or pollution is measured by the air quality index (AQI). The AQI data are calculated on the basis of the rules in the Ambient Air Quality Standard of China, taking into account five main pollution sources (round ozone, particulate pollution, carbon monoxide, sulfur dioxide, and nitrogen dioxide). The data are taken from http://www.air-level.com.


• Water quality: water quality is measured by city water quality index (CWQI). The data are calculated on the basis of the rules in Environmental Quality Standards for Surface Water of China.

Level of the social system

The efficiency of the social subsystem is reflected in the production process. Assuming that in the production stage of the green economy, there are n DMUs (i.e., all 324 cities in this study), each

(4)

consumes types of production factors, such as labor, capital, and resources, to produce products and ultimately obtain types of economic outputs, the input-output matrices will be

(5)

and

(6)

respectively.

In particular, for DMU_d, the efficiency in the production stage is EP_d, and the weights ω_id^x and ω_pd^y of the input-output factors x_id and y_pd follow the principle of efficiency maximization. Therefore, the self-evaluation model for production efficiency of the green economy of DMU_d can be derived as follows:

(7)

Under the self-evaluation model, each DMU will select the weight that is most favorable to its efficiency; therefore, because of different weights, the efficiency of each DMU is measured by the lack of comparability. Furthermore, a cross-efficiency model introduces a mutual evaluation mechanism to measure the efficiency of each. The cross-efficiency EP_do of DMU_o relative to DMU_d can be solved using the following formula:

(8)

where ω_id^x* and ω_pd^y* are the optimal weights obtained based on the planning formula of EP_do. The cross-efficiency of DMU_o is the efficiency after all other evaluations of DMUs, and therefore, in terms of DMU_o, the ultimate cross-efficiency EP_o is as follows:

(9)

EP_o is the development level of the social subsystem of DMU_o.

Level of the ecological system

The efficiency of the ecological subsystem is embodied in the governance process. Each DMU can invest the economic output obtained through economic production in the environmental governance stage. The investments in environmental governance can be divided into two categories: one is investment in environmental pollution control, and the other is investment in environmental infrastructure construction. The investment in environmental governance improves environmental quality, which is prominently reflected in air quality, the quality of surface water, park green space, etc. In terms of DMU_j, its input-output matrices in the environmental governance stage are

(10)

and

(11)

respectively.

Similar to the economic production stage, in terms of DMU_d, efficiency at the environmental governance stage is EG_dd, and the weights ω_rd^z and ω_ld^u of the input-output factors z_rd and u_ld follow the principle of maximum efficiency. Thus, the self-evaluation model for environmental governance efficiency of DMU_d can be obtained as follows:

(12)

Furthermore, the cross-efficiency EG_do of DMU_o for the mutual evaluation mechanism relative to DMU_d is introduced as follows:

(13)

where ω_rd^z* and ω_ld^u* denote the optimal weights. The ultimate cross-efficiency of DMU_o is as follows:

(14)

EG_o is just the development level of the ecological subsystem of DMU_o.

Model solving and relationship identification

Calculation of the symbiotic coordination

Next, we solve the model. The general form of model x can be expressed as follows:

(15)

Because both levels of the social and ecological subsystems are discrete data, the data are discretized by mapping the relationship between gray derivatives and even logarithms in the gray theory (Wu and Wang 2011), and dA/dt and dB/dt form mapping relations with even logarithms (A_t+1, A_t) and (B_t+1, B_t), respectively. The background values for time t are (A_t+1, A_t)/2 and (B_t+1, B_t)/2, respectively, so the formula x can be discretized as follows:

(16)

The original data are substituted for regression, so all parameters can be obtained as follows:

(17)

(18)

(19)

(20)

(21)

and

(22)

Thus, the degree of symbiotic coordination

(23)

between the social and ecological subsystems can be obtained. Based on the definition formula of C,

(24)

Identification of the coordination relationship

According to C, θ_A, and θ_B, the coordinated development status of the social and ecological subsystems can be determined from the perspective of biological interaction between species (Table 1; Fig. 3).

Descriptive statistics of indicators

Changes in the input-output indicator data in the economic production process were analyzed first. Among them, capital investment in economic production continued to increase, and the trend of annual average investment in fixed assets in each city increased in an approximately linear manner, from 180.29 billion yuan in 2015 to 242.35 billion yuan in 2019. Average annual growth rate also increased continuously, from 12.96 billion yuan to 16.78 billion yuan. As an indispensable energy resource in the production process, the average power consumption of 290 sampled cities also increased yearly; from 2015 to 2019 it increased by over two times, from 10.22 billion kilowatt-hours (kWh) to 22.95 billion kWh. In contrast, the level of intensive use of water resources in each city, measured by the average level, increased slightly. Compared with that of previous years, average water consumption decreased in 2016 and 2019, with overall water consumption slightly decreasing from 1.77 to 1.72 billion cubic meters. Analysis results of the kernel density function also support this trend (Fig. 4). From 2015 to 2019, the kernel density peaks of input factors other than consumption of water sources showed an obvious trend of continuously shifting to the right. In addition, gaps between input factors in each city narrowed and the kernel density curve changed from narrow to wide. As the goal of the economic production stage, economic output of each city also grew linearly, from 247.49 billion yuan to 327.60 billion yuan, with the growth rate being faster than the growth rate of fixed asset investment.

In the environmental governance stage, investment in environmental pollution control and environmental infrastructure increased slightly, but gaps between cities increased; average values of the two increased from 28.33 billion yuan and 16.34 billion yuan to 34.34 billion yuan and 21.18 billion yuan, respectively, but minimum values fluctuated. The minimum investment in environmental pollution control increased from 2.14 billion yuan to 2.37 billion yuan in 2016 and then decreased to 1.21 billion yuan in 2019 in a fluctuating manner. Similarly, investment in environmental infrastructure increased from 0.64 billion yuan to 1.19 billion yuan in 2017, which was also the highest level of this indicator in five years. Moreover, the minimum investment in environmental pollution control continuously decreased. By 2019, its level had fallen back to the 2015 level (0.85 billion yuan). In terms of output, Figure 5 shows that air quality and surface water quality of each city improved significantly, and green space per capita also increased.

RESULTS

In this section, the efficiency of economic production and environmental governance in each city was analyzed first to understand the development and absolute level of economic and ecological subsystems of each city. Then, based on the results, an in-depth analysis was conducted on the symbiotic coordination of the economic and ecological subsystems to measure the coupling and coordination degree of SES in each city.

Comparison of efficiencies

The efficiency level of the economic and ecological subsystems was used to measure their development level. Figures 6 and 7 show average efficiency values of the two subsystems. From 2015 to 2019, the average efficiency values of the economic production process and environmental governance process of 290 cities were 0.5335 and 0.1974, respectively. The average efficiency level of the economic production process was significantly higher than that of the environmental governance process. Thus, from the perspective of input and output, the average level of the economic subsystem of each city was higher than that of its ecological subsystem. Compared with the input-output efficiency in the environmental governance process, resource input and utilization efficiency in the economic production process were higher. Although the absolute efficiency of the economic and ecological subsystems differed, the differences between the two subsystems began to decrease. From 2015 to 2019, the standard deviation of the efficiency of the economic subsystem increased from 0.1743 to 0.1907, whereas that of the ecological subsystem decreased from 0.2033 to 0.1911. Starting from the internal gaps, the gaps between the environmental and economic subsystems continued to decrease.

It can be seen from the spatial distribution of cities that cities with higher economic efficiency are mostly located in the eastern coastal areas. In terms of economic production efficiency, Beijing, Shanghai, Shenzhen, etc., have always been leading, but the environmental governance efficiency of these cities is relatively low. This phenomenon does not directly indicate that ecological subsystems of these cities are at an absolute disadvantage but reflects that under current circumstances, environmental governance in these economically developed cities has not yet achieved the expected effects. Compared with the investment in environmental governance in these cities, that of the ecological subsystem should be commensurate with better development quality.

The average efficiency of the economic and ecological subsystems was comprehensively considered. From 2015 to 2019, average efficiency first increased, then decreased and increased again, from 0.3836 in 2015 to 0.3922 in 2016, and then dropped drastically to 0.3144 in 2017. After that, average efficiency of SES began to enter a stage of rapid growth. Moreover, the internal gap in the efficiency of SES was smaller than that within the economic and ecological subsystems, and the gap between systems was smaller than that within a system.

Coupling of the economic and social subsystems

The modes of the economic and ecological subsystems of each city in China were dominated by mutualism and amensalism. Among the 290 cities, the number of cities with mutualism, commensalism, amensalism, and competition were 102, 3, 108, and 77, respectively. The proportion of mutualism and amensalism cities was 35.17% and 37.24%, respectively (Table 2). Among all cities, only three were in the commensalism state in terms of economic development and environmental governance. One of them demonstrated that environmental governance promoted economic development, and the other two showed that economic development promoted environmental governance, but environmental governance did not improve production efficiency. Among the 108 cities in the amensalism state, the economic subsystems of SES in more than 60% of the cities played a dominant role, that is, progress at the economic level was conducive to the improvement process of environmental governance. However, for these cities, if their pursuit of environmental quality progresses, it will restrict rapid development of the economy. For the remaining 42 of these 108 cities, their economic development process will restrict improvement of environmental quality governance, but in turn, improvement of environmental quality provides good external conditions for economic development, thereby enhancing economic development (Fig. 8).

Regarding cities with different economic development levels, economically developed cities showed significant polarization, and the SES of economically underdeveloped cities showed higher vulnerability. Among cities with a top ten ranking in GDP in 2019, only the economic and ecological subsystems of Tianjin were in a mutualism state, and its SES had a relatively strong stability. In addition, the economic and ecological subsystems of Shenzhen, which ranked third in the economy, were in a commensalism state, and the improvement of its economic production efficiency was conducive to the improvement of its environmental governance efficiency. Among the other eight cities, four were in amensalism and competition states, and most of them were in a state of amensalism, indicating that improvements in environmental governance will restrict rapid development of the economy. In addition, among the 58 cities ranked in the top 20% in GDP, the economic and ecological subsystems of 19 cities were in a mutualism state, and their SES was relatively stable. However, the economic and ecological subsystems in 23 cities were in a competition state and the SES was relatively vulnerable, and the cumulative proportion of cities of these two symbiotic types accounted for more than 70% of the total. The performance of economically underdeveloped cities was relatively more consistent, and the social-ecological systems of these cities were more vulnerable. Among the 29 cities ranked in the bottom 10% in economic development in 2019, only four cities achieved mutualism between the economic and ecological subsystems, and the other cities were in an amensalism or competition state; they realized benign interaction between economic development and environmental governance processes but not in a way conducive to the stability of SES.

Figure 9 shows the development levels of the economic and ecological subsystems of 290 sampled cities and their symbiosis types. The figure shows that for cities in a mutualism state, the efficiency of economic production and environmental governance was relatively high, whereas for cities in an amensalism state, either economic production or environmental governance was effective. This feature was not obvious among cities in a mutualism state. The distribution of the economic and environmental efficiency of cities in an amensalism state was similar to that of cities in a mutualism state, but the figure also shows that efficiency of the economic and ecological subsystems of cities in a mutualism state was significantly lower than that of cities in an amensalism state. Regarding cities in a competition state, there were cities with extremely high economic production efficiency but extremely low environmental governance efficiency, as well as those with extremely high environmental governance efficiency but extremely low economic production efficiency. However, most cities in a competition state had low economic efficiency, which also became an internal reason for the relatively vulnerable subsystem of the SES.

The stability of the SES of a region was further analyzed by combining its economic production and environmental governance capability. Economic production and environmental governance efficiencies were averaged and then classified into three categories (high, medium, and low). The aim of averaging them was that for a region, whether the subsystem of an SES was in a state of mutualism, commensalism, amensalism, or competition, even if it could not achieve stable SES development it should at least have a leading advantage in the development level of a subsystem to reduce the ecosystem’s vulnerability and achieve stable development. Results of analysis indicated that cities in a mutualism state had the largest number at high and medium levels, followed by cities in an amensalism state. However, except for those cities in a commensalism state, low-level cities still occupied the largest proportion of various symbiosis types (high, medium, and low levels), and there was room for improvement in the efficiency of economic production and environmental governance (Fig. 10, Table 3).

CONCLUSIONS

In this study, using 290 cities at the prefecture level and above in China as research objects, a vulnerability assessment of SES was conducted from a place-based perspective. After measuring development levels of economic production activities and ecological environment in human society, vulnerability of SES in each city was evaluated by investigating the interaction between the economic and social subsystem and ecological environment subsystem and their dynamic feedback.

The main conclusions of this study are as follows. From the perspective of input and output, the average level of the economic subsystem of each city is higher than that of the ecological subsystem. Compared with the input-output efficiency in the environmental governance process, the input and utilization efficiency of resources in the economic production process is higher. From the perspective of spatial distribution, cities with higher economic efficiency are mostly located in eastern coastal areas, but the performance of ecological subsystems of economically developed cities has not yet reached the expected effects. In terms of the interaction between subsystems and the degree of coupling and coordination, modes of the economic and ecological subsystems of each city are dominated by mutualism and amensalism. The vulnerability of SES in more than half of the cities is still relatively high, and there is still room for improvement in the degree of coordination between economic production activities and the natural environment. The vulnerability of SES in economically developed cities shows significant polarization. The two subsystems in most of such cities are in a state of mutualism or amensalism, whereas the SES of economically underdeveloped cities is even more vulnerable.

DISCUSSION

In this study, a new combination of ecology and management models was used to assess vulnerability of the SES of each city. The model combining DEA and Lotka-Volterra proposed in this paper may serve as a reference for identifying the degree of coordination of various essential dimensions in green development. In addition, research results indicate that some cities’ higher green development levels may result from short-term economic development eroding the ecological environment. However, if this situation is maintained for a long time, its unhealthy economic development mode will seriously damage the ecological environment and ultimately endanger the green development of this region. Therefore, to maintain long-term and healthy green development, it is necessary to deeply analyze the degree of coordination between different fields of green development.

This study’s research design and method still have some limitations. First, the green development system is divided only into green economy and green ecology. However, many other factors, such as innovation and policy, may also affect the green development process. Second, the analysis of the linkage influence relationship between subsystems is relatively lacking. Interaction between cities has become increasingly frequent and close in recent years. Therefore, this study needs to concentrate more on the linkage mechanism and mutual influence of the green development system and its subsystems among cities. Third, analysis of the driving factors affecting coordination of green development SES is relatively insufficient. Analysis at the city level also makes it possible to further explore the correlation of the regional system of SES vulnerability. Further research can be conducted from the following three perspectives: (1) conducting a multi-dimensional assessment of SES vulnerability by combining place-based, system-based, and exposure-sensitivity-adaptability perspectives; (2) exploring the regional system correlation and spatial impact of SES vulnerability; and (3) deeply analyzing the key factors that affect the vulnerability of SES.

LITERATURE CITED

Bogert, J. M., J. Ellers, S. Lewandowsky, M. M. Balgopal, and J. A. Harvey. 2022. Reviewing the relationship between neoliberal societies and nature: implications of the industrialized dominant social paradigm for a sustainable future. Ecology and Society 27(2):7. https://doi.org/10.5751/ES-13134-270207

Cannon, T., and M. Müller-Mahn. 2010. Vulnerability, resilience and development discourses in context of climate change. Natural Hazards 55:621-635. https://doi.org/10.1007/s11069-010-9499-4

Chaffin, B. C., and L. H. Gunderson. 2016. Emergence, institutionalization and renewal: rhythms of adaptive governance in complex social-ecological systems. Journal of Environmental Management 165:81-87. https://doi.org/10.1016/j.jenvman.2015.09.003

Chien, H., and O. Saito. 2021. Evaluating social-ecological fit in urban stream management: the role of governing institutions in sustainable urban ecosystem service provision. Ecosystem Services 49:101285. https://doi.org/10.1016/j.ecoser.2021.101285

Clark, V. R., J. d. D. Vidal, I. M. Grundy, T. Fakarayi, S. L. Childes, N. P. Barker, and H. P. Linder. 2019. Bridging the divide between intuitive social-ecological value and sustainability in the Manica Highlands of southern Africa (Zimbabwe-Mozambique). Ecosystem Services 39:100999. https://doi.org/10.1016/j.ecoser.2019.100999

Dietz, T., E. Ostrom, and P. C. Stern. 2003. The struggle to govern the commons. Science 302(5652):1907-1912. https://doi.org/10.1126/science.1091015

Gardner, J. S., and J. Dekens. 2007. Mountain hazards and the resilience of social-ecological systems: lessons learned in India and Canada. Natural Hazards 41:317-336. https://doi.org/10.1007/s11069-006-9038-5

Guimarães, M. H., J. Ballé-Béganton, D. Bailly, A. Newton, T. Boski, and T. Dentinho. 2013. Transdisciplinary conceptual modeling of a social-ecological system—A case study application in Terceira Island, Azores. Ecosystem Services 3:e22-e31. https://doi.org/10.1016/j.ecoser.2012.12.007

Gunderson, L. H., and C. S. Holling, editors. 2002. Panarchy: understanding transformations in human and natural systems. Island, Washington, D.C., USA.

Guo, S.-D., H. Li, R. Zhao, and X. Zhou. 2019. Industrial environmental efficiency assessment for China’s western regions by using a SBM-based DEA. Environmental Science and Pollution Research 26:27542-27550. https://doi.org/10.1007/s11356-019-06062-5

Guo, X. R., L. Wu, and M. Wang. 2022. Application of Grey Lotka-Volterra model in water-economy-industry-technology innovation system in Beijing-Tianjin-Hebei region. International Journal of Environmental Research and Public Health 19(15):8969. https://doi.org/10.3390/ijerph19158969

He, L., J. Shen, and Y. Zhang. 2018. Ecological vulnerability assessment for ecological conservation and environmental management. Journal of Environmental Management 206:1115-1125. https://doi.org/10.1016/j.jenvman.2017.11.059

He, Y., C. Zhou, and T. Ahmed. 2021. Vulnerability assessment of rural social-ecological system to climate change: a case study of Yunnan Province, China. International Journal of Climate Change Strategies and Management 13(2):162-180. https://doi.org/10.1108/IJCCSM-08-2020-0094

Jiang, Y., B. Shi, G. Su, Y. Lu, Q. Li, J. Meng, Y. Ding, S. Song, and L. Dai. 2021. Spatiotemporal analysis of ecological vulnerability in the Tibet Autonomous Region based on a pressure-state-response-management framework. Ecological Indicators 130:108054. https://doi.org/10.1016/j.ecolind.2021.108054

Kasperski, S., G. S. DePiper, A. C. Haynie, S. Blake, L. L. Colburn, A. Freitag, M. Jepson, M. Karnauskas, K. M. Leong, D. Lipton, M. Masi, C. Speir, H. Townsend, and M. Weijerman. 2021. Assessing the state of coupled social-ecological modeling in support of ecosystem based fisheries management in the United States. Frontiers in Marine Science 8. https://doi.org/10.3389/fmars.2021.631400

Lazzari, N., M. A. Becerro, J. A. Sanabria-Fernandez, and B. Martí-López. 2021. Assessing social-ecological vulnerability of coastal systems to fishing and tourism. Science of the Total Environment 784:147078. https://doi.org/10.1016/j.scitotenv.2021.147078

Liu, J., T. Dietz, S. R. Carpenter, M. Alberti, C. Folke, E. Moran, A. N. Pell, P. Deadman, T. Kratz, J. Lubchenco, et al. 2007. Complexity of coupled human and natural systems. Science 317(5844):1513-1516. https://doi.org/10.1126/science.1144004

Liu, K., G.-l. Yang, and D.-g. Yang. 2020. Investigating industrial water-use efficiency in mainland China: an improved SBM-DEA model. Journal of Environmental Management 270:110859. https://doi.org/10.1016/j.jenvman.2020.110859

Liu, Y., and F. Dong. 2021. How technological innovation impacts urban green economy efficiency in emerging economies: a case study of 278 Chinese cities. Resources, Conservation and Recycling 169:105534. https://doi.org/10.1016/j.resconrec.2021.105534

Lockwood, M., M. Mitchell, S. A. Moore, and S. Clement. 2014. Biodiversity governance and social-ecological system dynamics: transformation in the Australian Alps. Ecology and Society 19(2):13. http://dx.doi.org/10.5751/ES-06393-190213

Lotka, A. J. 1925. Elements of physical biology. Williams and Wilkins, Baltimore, Maryland, USA.

Mafi-Gholami, D., S. Pirasteh, J. C. Ellison, and A. Jaafari. 2021. Fuzzy-based vulnerability assessment of coupled social-ecological systems to multiple environmental hazards and climate change. Journal of Environmental Management 299:113573. https://doi.org/10.1016/j.jenvman.2021.113573

Mao, Z., J. Centanni, F. Pommereau, A. Stokes, and C. Gaucherel. 2021. Maintaining biodiversity promotes the multifunctionality of social-ecological systems: holistic modelling of a mountain system. Ecosystem Services 47:101220. https://doi.org/10.1016/j.ecoser.2020.101220

Müller, S., N. Backhaus, P. Nagabovanalli, and S. Abiven. 2019. A social-ecological system evaluation to implement sustainably a biochar system in South India. Agronomy for Sustainable Development 39:43. https://doi.org/10.1007/s13593-019-0586-y

Naylor, L. A., U. Brady, T. Quinn, K. Brown, and J. M. Anderies. 2019. A multiscale analysis of social-ecological system robustness and vulnerability in Cornwall, UK. Regional Environmental Change 19:1835-1848. https://doi.org/10.1007/s10113-019-01530-7

Noble, M. M., D. Harasti, J. Pittock, and B. Doran. 2021. Using GIS fuzzy-set modelling to integrate social-ecological data to support overall resilience in marine protected area spatial planning: a case study. Ocean & Coastal Management 212:105745. https://doi.org/10.1016/j.ocecoaman.2021.105745

Ostrom, E. 2009. A general framework for analyzing sustainability of social ecological systems. Science 325(5939):419-422. https://doi.org/10.1126/science.1172133

Pao, H.-T., H. (A.) Chen, and Y.-Y. Li. 2015. Competitive dynamics of energy, environment, and economy in the U.S. Energy 89:449-460. https://doi.org/10.1016/j.energy.2015.05.113

Pelling, M., and C. High. 2005. Understanding adaptation: what can social capital offer assessments of adaptive capacity? Global Environmental Change 15(4):308-319. https://doi.org/10.1016/j.gloenvcha.2005.02.001

Perry, R. I., R. E. Ommer, M. Barange, and F. Werner. 2010. The challenge of adapting marine social-ecological systems to the additional stress of climate change. Current Opinion in Environmental Sustainability 2(5-6):356-363. https://doi.org/10.1016/j.cosust.2010.10.004

Shuai, S., and Z. Fan. 2020. Modeling the role of environmental regulations in regional green economy efficiency of China: empirical evidence from super efficiency DEA-Tobit model. Journal of Environmental Management 261:110227. https://doi.org/10.1016/j.jenvman.2020.110227

Silva, M. R. O., M. G. Pennino, and P. F. M. Lopes. 2019. Social-ecological trends: managing the vulnerability of coastal fishing communities. Ecology and Society 24(4):4. https://doi.org/10.5751/ES-11185-240404

Tai, X., W. Xiao, and Y. Tang. 2020. A quantitative assessment of vulnerability using social-economic-natural compound ecosystem framework in coal mining cities. Journal of Cleaner Production 258:120969. https://doi.org/10.1016/j.jclepro.2020.120969

Thiault, L., P. Marshall, S. Gelcich, A. Collin, F. Chlous, and J. Claudet. 2017. Mapping social-ecological vulnerability to inform local decision making. Conservation Biology 32(2):447-456. https://doi.org/10.1111/cobi.12989

Thiault, L., P. Marshall, S. Gelcich, A. Collin, F. Chlous, and J. Claudet. 2018. Space and time matter in social-ecological vulnerability assessments. Marine Policy 88:213-221. https://doi.org/10.1016/j.marpol.2017.11.027

Virapongse, A., S. Brooks, E. C. Metcalf, M. Zedalis, J. Gosz, A. Kliskey, and L. Alessa. 2016. A social-ecological systems approach for environmental management. Journal of Environmental Management 178:83-91. https://doi.org/10.1016/j.jenvman.2016.02.028

Volterra, V. 1926. Fluctuations in the abundance of a species considered mathematically. Nature 118:558-560. https://doi.org/10.1038/118558a0

Walker, B., and D. Salt. 2006. Resilience thinking: sustaining ecosystems and people in a changing world. Island, London, UK.

Wu, L., and Y. Wang. 2011. Estimation the parameters of Lotka-Volterra model based on grey direct modelling method and its application. Expert Systems with Applications 38(6):6412-6416. https://doi.org/10.1016/j.eswa.2010.09.013

Yuan, L., R. Y. Li, W. He, X. Wu, Y. Kong, D. M. Degefu, and T. S. Ramsey. 2022. Coordination of the industrial-ecological economy in the Yangtze River economic belt, China. Frontiers In Environmental Science 10:882221. https://doi.org/10.3389/fenvs.2022.882221

Xu, M., C. Chen, and X. Deng. 2019. Systematic analysis of the coordination degree of China’s economy-ecological environment system and its influencing factor. Environmental Science and Pollution Research 26:29722-29735. https://doi.org/10.1007/s11356-019-06119-5

Xing, M., F. Luo, and Y. Fang. 2021. Research on the sustainability promotion mechanisms of industries in China’s resource-based cities—from an ecological perspective. Journal of Cleaner Production 315:128114. https://doi.org/10.1016/j.jclepro.2021.128114