Research

INTRODUCTION

Protected areas are an effective means of managing resources and conserving biodiversity (Juffe-Bignoli et al. 2014, Huang et al. 2019). Initially, protected areas were considered as “nature for itself” zones, aiming to minimize the impact of human activities (Mace 2014). With the advancement of ecological protection efforts, today’s protected areas are no longer viewed as purely ecological islands (Cumming et al. 2015). It is increasingly recognized that a protected area is a social-ecological system (Strickland-Munro et al. 2010), characterized by complexity, uncertainty, self-organization, and diversity (Kroner et al. 2019). It emphasizes the interaction and symbiosis between humans and nature (Liu et al. 2007). Actually, the SESs thinking is increasingly reflected in international environmental policies, such as the United Nations Sustainable Development Goals, which emphasize the importance of achieving multiple goals that span ecological and social well-being. Consequently, there is a growing focus on maintaining the resilience and adaptability of protected areas, and the long-term sustainability of populations, communities, and ecosystems has become the focus of research (Palomo et al. 2014, Cumming and Allen 2017, Rodríguez-Rodríguez et al. 2021, Gatiso et al. 2022). China’s nature reserves are still at the exploratory stage in coordinating ecological conservation and socioeconomic development, because of top-down management and the lack of local managers’ professional knowledge such as systematic thinking and resilience theory (Zhou et al. 2014, Zhang et al. 2020). Applying SESs theory to protected areas can provide important theoretical and practical significance to alleviating the contradiction between social development and ecological protection through in-depth analysis of interactions within SESs elements (Guerrero and Wilson 2017).

Qilian Mountain is an important ecological security barrier in China, and is responsible for maintaining the Qinghai-Tibet Plateau’s ecological balance. It was designated as a national nature reserve in 1988, when China’s nature reserves increased dramatically in number and area (Huang et al. 2019). However, its management at that time focused more on water conservation forestry rather than strict protection. This led to various human activities such as mining, unauthorized construction of hydropower facilities, and polluting from local factories, all of which were prohibited by national policy but permitted by the local government. In 2017, the General Office of the Central Committee of the Communist Party of China and the General Office of the State Council gave a briefing on the destruction of ecological environment in the reserve. Subsequently, numerous ecological protection projects were carried out in the reserve. At the same time, in order to improve the governance system of protected areas, the reserve was assigned as a pilot area of the China National Park System Pilot Project (Qilian Mountain National Park, QMNP). Unlike traditional nature reserves, national parks add the recreation function under the strictest ecological protection, which considers the significance of social subsystems in the system’s sustainable development (Yan and Ding 2020). Coordinating the relationship between ecological protection and economic development has emerged as an imperative task for national parks.

The social-ecological systems framework (SESF) was developed by Ostrom as a diagnostic tool (Ostrom 2009). It contains action situations in which human-nature interactions lead to certain outcomes of sustainability. These interactions and outcomes (I-O) are directly influenced by the properties of four subsystems: actors, governance system, resource system, and resource units (McGinnis and Ostrom 2014). The SESF integrates data from various natural and social science, recognizing the linkages and feedbacks between humans and natural systems. Based on Ostrom’s framework, this paper first constructs the social-ecological action situations of QMNP and makes a qualitative analysis. Then the comprehensive evaluation index system of the QMNP is constructed by referring to the index system of Ostrom, and the interaction among subsystems is quantitatively analyzed using coupling coordination degree model (CCDM). Qualitative and quantitative analysis of action situations can reflect the operation mechanism and existing problems of the system, which has practical significance for the management of QMNP. Furthermore, operationalizing the SESF in the national park could provide systematic governance thinking and analytical tools for other protected areas.

THEORY SETTINGS AND STUDY AREA

Theory settings

The SESF was proposed by Ostrom in 2009 as a tool for analyzing the SESs sustainability based on the common-pool resources (CPRs) and institutional analysis and development framework (IAD; Basurto and Ostrom 2009). The framework consists of four core subsystems: resource system (RS), resource units (RU), governance system (GS), and actors (A), as well as the interactions and outcomes (I-O) produced by the four subsystems in the settings of social, economic, political, and related ecosystems. At the same time, the subsystems are also feedback from the interactions-outcomes. These interactions connect ecosystems and social systems as a whole, constituting the “action situations” in natural resource governance.

The purpose of the SESF is to identify the interaction among biophysical, institutional, and social, which may influence groups in developing institutional arrangements for natural resources (Schlüter et al. 2019). McGinnis and Ostrom modified the framework in 2014, and updated and refined the secondary and tertiary indexes under the four subsystems (McGinnis and Ostrom 2014). Other researchers improved the variables and applied the framework to various types of environmental governance, such as marine fishery (Aswani et al. 2013, Basurto et al. 2013, Barnett and Eakin 2015), forest (Fleischman et al. 2010), protected areas (Dumyahn and Pijanowski 2011, Govigli et al. 2021), wildlife (Dressel et al. 2018, Ferreira-Rodriguez et al. 2019), water and irrigation (Cox 2014, Flynn and Davidson 2016), and community self-organization (Delgado-Serrano and Ramos 2015), etc. For example, Epstein et al. 2013) proposed adding ecological rules as a seventh core subsystem, which will allow scholars to combine the knowledge of natural science. Guevara et al. (2016) emphasized the importance of the external social, economic, and political environment, and created new variables to better explain how external variables affect SESs and their management. Cole et al. (2019) combined IAD and SESF to propose a new framework, namely the Combined IAD-SES (CIS) Framework, which applies variable-oriented and process-oriented approaches to the research of environmental sustainability. It is also used as an analysis/classification tool to identify the influence variables (Schmitt-Harsh and Mincey 2020).

Action situations have become an important unit of analysis in SESs (Partelow 2018). The identification of action situations is mainly through qualitative analysis methods such as document analysis, field site visits, interviews, and focus group discussion (Kasymov et al. 2023, Kimmich et al. 2023). Quantitative analysis of action situations is rarely practiced because of the complexity and uncertainty of systems, and limitations related to data and methods. For example, Vogt et al. (2015) applied SESF in Yellowwood Lake Watershed to analyze forest dynamics in Indiana, USA, and improved the ecological indicators of SESF to explicitly incorporate ecological principles into the action situation. Although the purpose of the article is to emphasize the theoretical development of SESF, it may be challenging for the study area to implement specific operations when there is a lack of quantitative analysis of I-O.

The literature review reveals that the SESF has been used as a foundational analytical tool for complex systems (Ernst et al. 2013). Although the framework takes both social and ecological processes into account, it does not explicitly specify which action scenarios are crucial. There may be multiple action situations in a given region, but different actors have a different focus. For example, in our study, the primary objective of Qilian Mountains as a national park is to protect the authenticity and integrity of its natural ecosystem, so the park managers prioritize ecological protection. However, residents living within the national park put their own survival and development needs first. Researchers are required to identify the focal action scenarios themselves, necessitating high professional knowledge. As previously mentioned, although many researchers have developed an index system of the SESF for different research targets and areas, there remain challenges and limitations in measuring the degree of interaction and outcome of action situations.

Study area

The Qilian Mountain is located at the border of Qinghai and Gansu provinces in northwest China (Fig. 1). There are unique eco-geographical features and various kinds of ecosystems, including forest, grassland, desert, farmland, wetland, glacier, and snowy mountains (cryosphere). It is a priority area for biodiversity conservation in China, providing an important migration corridor and habitat environment for wildlife, such as snow leopard (Panthera uncia), brown bear (Ursus arctos), white-lipped deer (Cervus albirostris), blue sheep (Pseudois nayaur), wild yak (Bos mutus), and Black-necked Crane (Grus nigricollis).

In 2017, in order to improve the management and sustainable development of the region, the nature reserve and its surrounding areas were integrated into a national park (Fig. 1b). The Qilian Mountain National Park Administration was established relying on the Xi’an Forest Resources Supervision Commissioner Office of National Forestry and Grassland Administration in 2018. QMNP implemented a hierarchical management organization and adopted a vertical management system of “National Forestry and Grassland Administration - Qilian Mountain National Park Administration - Gansu/Qinghai Provincial Administration - Management Branch - Protection station” (The governance system can be found in Fig. 2.) QMNP is divided into two areas: Gansu province and Qinghai province.

According to the 2020 data, 33,174 residents consisting of 30 ethnic groups live in Gansu part, mostly, Han, Tibetan, Mongolian, Yugur, Kazak, Hui, and Tu. The residential areas are mainly concentrated in the townships of Kangle, Huangcheng, Haxi, Zhuaxixiulong, Tiantang, and Tanshanling (Fig. 1c). Arable farming and animal husbandry are the primary livelihood for local residents, mainly planting oat grass, wheat, potatoes, and Chinese herbal medicine, as well as feeding yaks and Tibetan sheep. In 2022, the number of livestock was approximately 10 million. The average annual income for a resident was RMB 7100 (US$1000). Additionally, ecotourism has been developed in some places of the park. For example, Tiantang received 149,900 tourists in 2022 with about RMB 70 million (US$10 million) in tourism income. Furthermore, some scientific research monitoring activities are to protect the ecological environment of Qilian Mountain. The monitoring data cover wildlife data, meteorological data, and local resource data, etc.

FRAMEWORK AND QUALITATIVE ANALYSIS OF ACTION SITUATIONS

Action situations are the focus of the SESF, reflecting the integrity and complexity of SESs (Stevenson and Tissot 2014). The QMNP is an adaptive system that interacts with and is nested within ecological and social subsystems. For the purpose of systematically analyzing the interaction between subsystems of the region, we constructed the social-ecological action situations of QMNP (Fig. 2) based on Ostrom’s framework and Schlüter et al.’s (2019) proposed social-ecological action situations. These action situations were identified through our three field visits to Qilian Mountain in 2018 and 2020, as well as face-to-face interviews with local protection station staff and residents. They were summarized into three categories in this paper (Fig. 2, Table 1): Ecological-Ecological Action Situations (E-Es), Social-Social Action Situations (S-Ss), and Social-Ecological Action Situations (S-Es). Among them, E-Es are mainly manifested as predation and competition among biological entities in the study area, as well as species-habitat, plant-soil, and others. S-Ss are mainly represented by the formulation of rules, management, implementation, supervision, and coordination, information sharing, trading, cooperation, and competition, etc. S-Es mainly show protection, conflict, and utilization between natural resources and social systems.

Qilian Mountains is a complex adaptive system with numerous uncertainties and unpredictability (Guo et al. 2023). We listed the types of ecological-ecological, social-social, and social-ecological action situations (Table 1). However, it should be noted that not all situations are covered in the table. The purpose is to provide a paradigm for similar research.

Ecological-Ecological Action Situations (E-Es)

The E-Es of QMNP mainly occur within or between the resource system and resource units, and are defined by the participating ecological entities with their properties and biophysical rules that control the interactions (Schlüter et al. 2019). Ecological entities can be any part of an ecosystem, including individual organisms like a tree, biotic population such as snow leopard, white-lipped deer, blue sheep, or more aggregate units like water, soil, or vegetation. These entities have attributes such as spatial distribution, growth rates, self-organizing ability, and information transmission.

Resource units are part of the resource system. The interactions and feedbacks between resource units constitute a complete resource system. The self-organization of the resource system maintains stability of resource units. The ecosystem of Qilian Mountains is composed of multiple biological communities, species, and abiotic environmental elements, forming a complex ecological network. This complexity is reflected in the diversity of interactions between resource units, including predation, competition, mutualism, and ecological, geological, or biophysical processes, such as the survival and reproduction of populations under suitable ecological conditions (Table 1). The diversity makes the system have a high capacity to resist disturbance and self-regulate. Even if some resource units suffer damage or loss, others can compensate by adjusting their own state or function to avoid the collapse of the system (Biggs et al. 2012). These self-organizing processes interweave and influence each other to build resilience and adaptability of the Qilian Mountain ecosystem. This enables the system to maintain stability in the face of internal disturbances and external environmental changes. The outcomes will have short- or long-term impacts on biological entities or SESs, such as changes in ecosystem species richness or soil nutrient content, or even related human livelihood.

Social-Social Action Situations (S-Ss)

The S-Ss of QMNP occur mainly within or between the governance system and actors, and are defined by the participating human actors, their abilities, interaction rules, and institutions that control them. Human actors can be individuals, groups, or organizations involved in the QMNP, such as farmers and herdsmen, administrative departments, researchers, communities, tourists, etc.

Interactions within the governance system are reflected in the formulation, supervision, and implementation of policies and rules. For example, the Qilian Mountain National Park Administration is mainly responsible for revising and implementing policies, standards, and plans, with a focus on ecological protection and restoration. Local governments mainly perform social management, coordinate society and economy, supervise the market, and cooperate with the National Park Administration in ecological protection as needed. The communication and cooperation between management agencies and local governments is the S-Ss. Outcomes of these interactions, such as formulated rules, will have impacts on the local ecological environment and residents’ production and life. Interactions among actors are manifested in resource competition, information sharing, social cooperation, trading and so on. For instance, farmers exchange Chinese herbs planting skills, herdsmen rent grassland to others, some residents take part in agritainment to provide tourist services, etc. These interactions are influenced by the external market, economic, and political environment. Interaction between the governance system and actors is usually reflected in how the managers guide or restrict the actors’ activities through formulating policies or rules. Interactions-outcomes are supported and limited by rules, interests, and goals of actors.

Social-Ecological Action Situations (S-Es)

The S-Es of QMNP mainly occur between the resource system/resource units and the governance system/actors, and are defined by human actors and ecological entities, their abilities, and the social and biophysical rules and regimes that control their interactions.

Interactions between the resource system and governance system are reflected in the comprehensive management of the national park by the administration according to the ecological environment of QMNP, social needs, and upper policies. Decision making will determine the development direction of the resource system, such as the transformation of Qilian Mountains from a nature reserve to a national park. Development trends of the resource system will affect how managers adjust or formulate new policies and rules. Interactions between the resource units and governance system are manifested in the ecological monitoring and natural resource management. The governance system uses environmental monitoring technologies, such as remote sensing technologies, camera traps, sensor networks and GIS, timely observation of dynamic changes of resources, signs of ecological damage, or restoration process.

Through feedback of the monitoring data, managers can constantly adjust the management measures to ensure the health status of resource units. Management is attached to ecological entities, and changes in resource units prompt adjustments in the governance system to maintain the stability of the resource system. Interactions between the resource units and actors are mainly manifested in the utilization or dependence of natural resources. Possible outcomes include the collapse or increase in sustainability of natural resources, changes in land utilization, social welfare, or cultural practices. Additionally, conflicts between wildlife and residents in the national parks are also frequent in the S-Es. Interactions between the resource system and actors show that the resource system provides a living environment or natural resources for actors, and actors utilize the resource units from the resource system through activities such as cultural tourism.

Outcomes of S-Es in QMNP can be material or immaterial (Schlüter et al. 2019). For example, herdsmen utilize grassland resources for economic benefits, tourists derive spiritual or cultural pleasure from visiting the natural park. The outcomes may have positive or negative impacts on SESs. Actors can actively respond to the negative impacts by adjusting the governance system or formulating more diversified policies. For example, in the early stage, grassland degradation was serious because of the overgrazing in QMNP. After a grazing ban period, grassland vegetation recovered in a short time. However, an extended grazing ban would result in fire hazards and further grassland degradation. Therefore, effective management of grassland resources involves implementing a seasonal grazing and fodder-livestock balance system. The economic income of herdsmen would fluctuate with changes in grazing policies.

QUANTITATIVE ANALYSIS OF ACTION SITUATIONS

Typically, one or more elements of the SES interact in social, ecological, or social-ecological processes, with their outcomes depending on the properties of these elements and the processes that govern their interaction (Vogt et al. 2015). The SESF identifies broad characteristics of resource system and resource units, governance system and actors, and lists a large number of second- and third-layer variables (McGinnis and Ostrom 2014). Researchers can choose variables based on their study objectives and specific problems (Binder et al. 2013, Nagendra and Ostrom 2014). In order to ensure practical significance in our research results, we considered 14 relevant variables from Ostrom’s framework based on our knowledge of the research area, and made a quantitative analysis of the four dimensions and action situations.

Typical townships selection

Natural resources and social characteristics are different among regions in QMNP. Therefore, we selected nine representative towns as the specific research objects (Table 2). The selection process is as follows: (1) use GIS to identify the townships with an administrative area greater than 50% within QMNP; (2) determine whether there are settlements in the township through Remote Sensing Image, eliminate them if not; (3) select representative and typical townships from the remaining towns considering characteristics such as ethnic groups, livelihood, location, industries, settlements distribution density, rural residents’ income, etc.; (4) take into account data availability.

Index system

The main goal of the national park is to achieve sustainability and coordination between ecological protection and the community economy. We focused on the action situations, namely the interactions and feedbacks between ecosystems (resource system and resource units) and social systems (governance system and actors). In selecting specific SES variables for the study, we considered residents as the main actors, and selected 14 variables related to “residents-natural resources” (Table 3).

In the index system, RS3-Size of resource system was identified as an important variable by Ostrom (Ostrom 2009). There are significant differences in the size of the townships within the national park (Table 4). The larger the ratio, the greater the impact of national park management, the more limited use of natural resources. Under this indicator, we added Normalized Difference Vegetation Index (NDVI) as a tertiary variable. The higher the NDVI value indicates greater vegetation coverage and a larger resource scale in the national park. The road density was selected to characterize the RS4-Human-constructed facilities, which is closely related to human activity in the national park.

RU5-Number of units is primarily used to measure the abundance of natural resources in the study area. Land use intensity directly reflects the impact of human activities on natural resources and becomes an important driving force for global environmental change (Cao et al. 2019). The land use related to residents mainly includes forest land, grassland, and cultivated land. Therefore, RU5-Number of units was represented by the per capita grassland/ forest land/ cultivated land area.

The actors dimension included five secondary indicators that are specifically important for the study area. Among them, and because detailed data on the number of residents living within the national park were not available, we used the density of rural residential land (A11) to represent the number of relevant actors (A1) in this study. A61-Neighborhood relationship was represented by the scope of interpersonal communication, harmony, and trust of neighborhood relations; A71-Knowledge of SES was represented by the residents’ view of the relationship between social systems and ecological systems; A91-Technologies available was represented by livelihood diversity, which referred to the number of livelihood activities. The higher the livelihood diversity, the more options residents have (Li et al. 2022). Residents are the key implementers of policies, and their perception will directly affect policies’ effectiveness. Therefore, GS5-Operation rule was represented by residents’ understanding/ satisfaction/ implementation of relevant policies in the QMNP.

Data collection

The NDVI was derived from Geospatial Data Cloud (https://www.gscloud.cn). The road data were obtained from the National Catalogue Service for Geographic Information (https://www.webmap.cn). The land use data were obtained from GlobeLand30 (https://www.webmap.cn/commres.do?method=globeIndex). The rural residential land data were provided by National Tibetan Plateau Data Center (http://data.tpdc.ac.cn). Table 4 provides the primary data of these indicators.

Indicator data for the governance system (GS51, GS52, GS53) and actor dimension (A11, A21, A61, A71, A91) were obtained from the residents’ questionnaire survey in October 2020 (The questionnaire can be found in Appendix 1). Questionnaire indicators were assigned by a five-point Likert Scale. Considering the vast territory of the QMNP and residents living in scattered distribution, questionnaires were randomly distributed by government workers and protection station managers to local residents. We randomly entered the farmers’ homes for in-depth interviews to verify the reliability of the questionnaire data. A total of 513 questionnaires were sent out. The questionnaires with incomplete or untrue information and the same answers were deleted, and 487 valid questionnaires were obtained. Cronbach’s α was 0.749 (> 0.7), indicating that the reliability of the questionnaire data is high and can be used for further analysis. Among the 487 questionnaires, we used 425 questionnaires from the nine typical townships for this study, excluding those that received only a small number of responses. For example, there were only two questionnaires received in Dahonggou, so they were eliminated from our study. Details on the number of collected questionnaires can be found in Table 2.

Method

Comprehensive evaluation index (CEI) of subsystem

The CEI was calculated as the weighted sum of standardized values for each subsystem index, reflecting the development level of the subsystem. The weight of indicators was calculated by entropy method, chosen for its objectivity. Detailed steps are as follows:

First, standard processing of data. All 14 variables were normalized to a scale of 0-1 to facilitate analysis and comparison. The data were analyzed using IBM SPSS Statistics, and conformed to a normal distribution for further analysis.

Next, specific gravity (f_tj), entropy value (e_j), and information utility value (d_j) were calculated. Specific gravity was calculated so that the probability corresponding to each index summed to 1. Entropy value can judge the dispersion degree of an index. A smaller entropy value indicated greater dispersion and thus greater influence on comprehensive evaluation. The information utility value is defined as inversely proportional to the entropy value, and is often used to measure the importance or contribution of an indicator. The formulas are:

(1)

(2)

(3)

Finally, the weight of the jth index (w_ti) was calculated and the results are shown in Table 5. The weight of each indicator reflects its relative importance within the overall evaluation system. Each indicator is assigned a weight value ranging from 0 to 1, with the sum of weights for each indicator layer totaling 1. The calculation formula is as follows:

(4)

According to the above formulas, the CEI of four subsystems was calculated by the linear weighted method. The formula is as follows:

(5)

After calculating the CEI of subsystems, it was necessary to classify them in order to better compare the township development levels across different dimensions. The CEI values range from 0 to 1 and were divided into four grades using quartiles (Leslie et al. 2015): extremely low (0–0.25), low (0.25–0.5), high (0.5–0.75), and extremely high (0.75–1). The higher the index, the greater the development level and degree of the township in this dimension.

Coupling coordination degree model (CCDM)

We attempted to use the CCDM for a quantitative analysis of action situations. The CCDM is used to assess the degree of interaction and coordination development among two or three subsystems (Wang et al. 2022). Formulas are as follows:

(6)

(7)

(8)

In these formulas, C represents the coupling degree, CEI_RS, CEI_RU, CEI_GS, and CEI_A represent the CEI of resource system, resource units, governance system, and actors, respectively. α, β, γ, and δ are adjustment coefficients.

National parks should be protected first and utilized second. We set the adjustment coefficients for ecosystems as α = β = 0.26, and social systems as γ = δ = 0.24 (Yang et al. 2020, Pan et al. 2022). When analyzing the coupling relationship between two subsystems, the adjustment coefficient (α or β) of ecosystems is 0.6, and for social systems (γ or δ) is 0.4 (Table 3). T represents comprehensive coordination index among subsystems. D represents the coupling coordination degree. C and D are expressed in values ranging from 0 to 1 where values closer to 1 indicate extremely related subsystems. D is divided into 10 levels (Ariken et al. 2020, Weng et al. 2021, Gao et al. 2022; Table 6). In addition to analyzing D value, it is also necessary to compare differences in CEI. If the difference is small, it can be regarded as synergistic development (Wang et al. 2015).

Result analysis

Comprehensive evaluation index (CEI)

In terms of subsystems, the resource system and resource units exhibited a higher CEI than the governance system and actors (Table 5). This indicates that the ecological protection effect of QMNP has become increasingly apparent in recent years, with the development level of ecosystems surpassing that of social systems. In 2017, the Chinese government proposed a national park pilot in the Qilian Mountains, elevating ecological environment protection to unprecedented heights. The implementation of relevant policies has led to gradual improvements in the area’s ecological environment. However, economic and social development experienced a short period of stagnation and inadaptability. On the spatial scale, regions demonstrating strong performance in one subsystem do not necessarily excel in other three subsystems, indicating spatial heterogeneity within QMNP regarding sustainable utilization and management of natural resources (Fig. 3).

In Qifeng, the index of actors and the governance system was at an extremely low level (< 0.25), while the other two dimensions were at a low level. Despite occupying the largest area among the nine townships within the national park, Qifeng has the lowest NDVI value (0.26). The land use type is mainly grassland and bare land, resulting in low biodiversity and a fragile ecological environment. The local residents are mainly herding and mostly Tibetan, with a limited perception of policies. They mainly reside on the periphery of the national park, and the rural residential land is scattered and small. Therefore, it is recommended that Qifeng strengthen the governance of natural resources to maintain ecosystem stability. At the same time, policies should prioritize residents’ well-being. For example, create public posts engaged in ecological conservation, and improve the living environment of residents.

The indexes of Kangle and Mati were consistent in resource system, governance system, and actors dimensions. The resource system was at a high level for both townships, while the other two dimensions were at a low or extremely low level. The resource unit of Mati was taller than Kangle. Both townships are rich in natural resources, with livelihoods primarily centered around traditional stockbreeding that heavily relies on natural resources. Influenced by national culture and traditional concepts, the adaptability of residents is weak. The two townships possess unique geographical locations, natural landscapes, and culture. It is possible to suggest that these two townships have great development potential if they undergo industrial transformation such as developing eco-tourism.

In Huangcheng, the index of the resource system was low (0.492), while the other three dimensions were at a relatively high level. In recent years, Huangcheng has been undergoing a transformation of agriculture and stockbreeding. Based on its resource advantage, Huangcheng takes the Gansu alpine fine-wool sheep as the lead, and makes the standardization, industrialization, and scaled traditional animal husbandry. It also achieved a dynamic balance of grass and livestock by guiding the herdsmen to purchase forage, artificial grass planting, borrow grazing from other places (straw supplemental feeding). In 2020, approximately 13,000 livestock borrowed grazing from neighboring counties. Additionally, Huangcheng actively adjusted the planting structure by cultivating plateau summer vegetables, carrots, angelica, and other characteristic agricultural products to improve land output efficiency. These efforts have improved economic income of residents and enhanced their happiness, so they have a higher perception of relevant policies.

The CEI of the resource system in Danma is the lowest among the nine townships, while the road density was the highest. It indicates that the natural resources of Danma are limited, and human activities have a high interference with the ecological environment. The township has a predominantly Han population, with herding and farming as primary sources of income, resulting in lower overall income levels. In addition, recent efforts by the national park to carry out ecological restoration have restricted residents’ productive activities. It is essential for QMNP to not only focus on ecological restoration but also adjust industrial structures to enhance local social and economic development.

In Qilian township, both the index of resource system and actors were at a low level (0.483 and 0.341, respectively), while the resource unit and governance system were at a high level. The CEI of the governance system was actually the highest among all nine townships. Similar to Danma, the residents of Qilian township are mainly Han nationality with a better understanding of national park policies and greater openness to new things. However, because of environmental remediation efforts in recent years, social development has lagged behind, leading to low performance of actors. It is recommended that Qilian township should focus on developing characteristic industries such as planting characteristic agricultural products, and expanding their scale to enhance the sustainable development of the social system.

In Maozang, the resource system index was at a high level (0.692), while the other three dimensions were all at low or extremely low levels. Maozang is the only pure animal husbandry township in Tianzhu County, with residents relying heavily on natural resources for generations. In recent years, because of strict protection within the national park, the species and number of wild animals in Maozang have increased. Furthermore, the proximity of Maozang settlements to the core area of the natural park has heightened conflicts between wildlife and residents. With its unique geographical location and natural landscape, there is great potential for ecotourism development in Maozang without causing damage to the ecological environment.

The indexes of four subsystems in Tiantang and Haxi were at high level or above. Despite having a very high distribution density of settlements, they cause less damage to the ecological environment compared to other townships (the NDVI of Tiantang and Haxi are the highest and second highest among the nine townships: 0.67 and 0.58, respectively). The reason is that both two towns develop characteristic agricultural products (such as planting Chenopodium quinoa, Chinese herbs, rape, etc.), while also transforming their industries and improving livelihoods diversity through tourism development. As a result, Tiantang and Haxi scored higher in indicators A71-Knowledge of SES and A91-Livelihood diversity as well as had higher knowledge, satisfaction, and implementation of policies.

Coupling coordination degree (D)

The coupling coordination type of QMNP was primary coordination (D = 0.688), indicating the need for enhanced coordination among the four subsystems. This result is consistent with the findings of Gao and Li (2022) and Wang et al. (2023). The next step was analyzing the interaction between the four subsystems. Based on the constructed index system, only RS-RU (E-Es), RU-A (S-Es), and GS-A (S-Ss) were selected for quantitative analysis in this paper, with the following results.

RS-RU: The D of Danma and Qifeng were at the near coordination stage (D < 0.6), and others showed moderate coordination or higher (Fig. 4). The difference between the resource system and resource units in Danma was -0.217, suggesting that the resource system is lagging. With the established index framework, reducing grazing intensity would be beneficial for restoring vegetation and improving NDVI value in this area. Although there is synchronization in development of the resource system and resource units in Qifeng, the coupling coordination is at the stage of near coordination. Because it is an ecologically vulnerable area (Du et al. 2022), increased investment in environmental protection measures focusing on reducing human activities’ disturbance to ecosystem is needed here. For other townships showing good coordination, sustainability can be improved based on CEI differences as shown in Table 7.

RU-A: Qifeng and Maozang were at the stage of near coordination, and others were at primary coordination or above. The difference between resource units and actors of Qifeng was 0.116, indicating that the township needs to strengthen social and economic construction, such as centralized resettlement of scattered residents, strengthening ecological protection publicity, and providing sustainable livelihood skills training for residents. The resource units and actors in Maozang were synchronized but at the near coordination stage. It indicates that Maozang needs to synchronously improve the resource units and actors. That is, it needs to transform inefficient traditional industries into green and efficient industries, such as carrying out ecological animal husbandry demonstrations and ecotourism, which will reduce residents’ dependence on natural resources, and increase livelihood diversity and economic income.

GS-A: Maozang and Qifeng were at the near disorders stage (D were 0.492 and 0.476, respectively), Mati was at the stage of near coordination, while others were at primary coordination or above. Difference between the governance system and actors in Maozang was -0.148, indicating that the governance system is lagging. Actors of Qilian and Danma are also lagging. Tiantang was at high coordination but with a relatively lagging governance system compared with actors. The results show that Mati and Qifeng need to simultaneously strengthen the social economy and management. Maozang and Tiantang should work on improving residents’ perception of policies through some methods, for example, involving residents in policy formulation, strengthening publicity of relevant policies like national parks or nature reserves.

CONCLUSION AND DISCUSSION

The SESF clearly emphasizes the importance of action situations. In the process of development, protected areas have been disturbed by various factors, and the causes of regime shift are complex. Any single attribute cannot be understood entirely in isolation, and it is necessary to analyze the interaction between variables to reveal the causal relationship within SES. The SESF approach helps us to identify and examine microscopic variables in social and ecological aspects. Our analysis highlights the significance of identifying interactions between subsystems for promoting sustainable development in protected areas. Notably, our examination of QMNP reveals that the governance system plays a crucial role in regulating interactions between the resource system and actors. The behavioral decisions made by actors are identified as the main driver of resource system development.

Applying the SESF framework to QMNP holds great importance for effective park management. Our findings indicate that although efforts are being made to protect the environment, there is also a need to enhance the social subsystem in order to bridge the gap between the ecological and social systems, for example, taking advantage of local resources to develop eco-friendly agriculture, tourism, and other green industries to enhance residents’ livelihood diversity. Existing research scale is mainly focused on county scale, national park/ nature reserve scale, or a specific geographic region like eastern Qilian Mountains. Our study reveals spatial differences in the sustainability of subsystems at the township scale within QMNP. Therefore, township-scale differences in economic development and habitat quality should be considered when developing policies.

It is necessary to regard protected areas as SESs. Operationalizing the SESF in protected areas offers several advantages: (1) The SESF is an effective tool for identifying relevant variables and interactions in protected areas, despite certain limitations in the analysis method. Researchers can systematically analyze complex systems through its thinking. (2) It facilitates data integration across different disciplines, especially integrating multiple sets of qualitative and quantitative data into one data set. This allows for simultaneous analyzing of one or more SESs. (3) The framework can be used as a management tool for protected areas. Managers can analyze the past, present, or even future development of protected areas and choose one of the four subsystems for in-depth study.

It is a serious challenge to study all situations within a region because the action situations are complex, uncertain, and unpredictable. However, researchers can select specific action situations according to the purpose and content of their study. Moreover, unlike other areas, protected areas do not exist in isolation but are influenced by larger or smaller natural environments and social development settings. Therefore, studying interactions across scales is indispensable for understanding protected area sustainability, which will be the focus of our next study.

LITERATURE CITED

Ariken, M., F. Zhang, K. Liu, C. L. Fang, and H. T. Kung. 2020. Coupling coordination analysis of urbanization and eco-environment in Yanqi Basin based on multi-source remote sensing data. Ecological Indicators 114:106331. https://doi.org/10.1016/j.ecolind.2020.106331

Aswani, S., G. G. Gurney, S. Mulville, J. Matera, and M. Gurven. 2013. Insights from experimental economics on local cooperation in a small-scale fishery management system. Global Environmental Change 23(6):1402-1409. https://doi.org/10.1016/j.gloenvcha.2013.08.003

Barnett, A. J., and H. C. Eakin. 2015. “We and us, not I and me”: justice, social capital, and household vulnerability in a Nova Scotia fishery. Applied Geography 59:107-116. https://doi.org/10.1016/j.apgeog.2014.11.005

Basurto, X., S. Gelcich, and E. Ostrom. 2013. The social-ecological system framework as a knowledge classificatory system for benthic small-scale fisheries. Global Environmental Change 23(6):1366-1380. https://doi.org/10.1016/j.gloenvcha.2013.08.001

Basurto, X., and E. Ostrom. 2009. Beyond the tragedy of the commons. Economia Delle Fonti Di Energia e Dell’ambiente 35-60. https://doi.org/10.3280/EFE2009-001004

Biggs, R., M. Schlüter, D. Biggs, E. L. Bohensky, S. BurnSilver, G. Cundill, V. Dakos, T. M. Daw, L. S. Evans, K. Kotschy, A. M. Leitch, C. Meek, A. Quinlan, C. Raudsepp-Hearne, M. D. Robards, M. L. Schoon, L. Schultz, and P. C. West. 2012. Toward principles for enhancing the resilience of ecosystem services. Annual Review of Environment and Resources 37:421-448. https://doi.org/10.1146/annurev-environ-051211-123836

Binder, C. R., J. Hinkel, P. W. G. Bots, and C. Pahl-Wostl. 2013. Comparison of frameworks for analyzing social-ecological systems. Ecology and Society 18(4):26. https://doi.org/10.5751/ES-05551-180426

Cao, W., L. Huang, T. Xiao, and D. Wu. 2019. Effects of human activities on the ecosystems of China’s National Nature Reserves. Acta Ecologica Sinica 39:1338-1350.

Cole, D. H., G. Epstein, and M. D. McGinnis. 2019. Combining the IAD and SES frameworks. International Journal of the Commons 13(1):244-275. https://doi.org/10.18352/ijc.864

Cox, M. 2014. Applying a social-ecological system framework to the study of the Taos Valley Irrigation System. Human Ecology 42(2):311-324. https://doi.org/10.1007/s10745-014-9651-y

Cumming, G. S., and C. R. Allen. 2017. Protected areas as social-ecological systems: perspectives from resilience and social-ecological systems theory. Ecological Applications 27(6):1709-1717. https://doi.org/10.1002/eap.1584

Cumming, G. S., C. R. Allen, N. C. Ban, D. Biggs, H. C. Biggs, D. H. M. Cumming, A. De Vos, G. Epstein, M. Etienne, K. Maciejewski, R. Mathevet, C. Moore, M. Nenadovic, and M. Schoon. 2015. Understanding protected area resilience: a multi-scale, social-ecological approach. Ecological Applications 25(2):299-319. https://doi.org/10.1890/13-2113.1

Delgado-Serrano, M. D., and P. A. Ramos. 2015. Making Ostrom’s framework applicable to characterise social ecological systems at the local level. International Journal of the Commons 9(2):808-830. https://doi.org/10.18352/ijc.567

Dressel, S., G. Ericsson, and C. Sandstrom. 2018. Mapping social-ecological systems to understand the challenges underlying wildlife management. Environmental Science and Policy 84:105-112. https://doi.org/10.1016/j.envsci.2018.03.007

Du, Q. Q., Y. F. Sun, Q. Y. Guan, N. H. Pan, Q. Z. Wang, Y. R. Ma, H. C. Li, and L. S. Liang. 2022. Vulnerability of grassland ecosystems to climate change in the Qilian Mountains, northwest China. Journal of Hydrology 612:128305. https://doi.org/10.1016/j.jhydrol.2022.128305

Dumyahn, S. L., and B. C. Pijanowski. 2011. Beyond noise mitigation: managing soundscapes as common-pool resources. Landscape Ecology 26:1311-1326. https://doi.org/10.1007/s10980-011-9637-8

Epstein, G., J. M. Vogt, S. K. Mincey, M. Cox, and B. Fischer. 2013. Missing ecology: integrating ecological perspectives with the social-ecological system framework. International Journal of the Commons 7(2):432-453. https://doi.org/10.18352/ijc.371

Ernst, B., J. Chamorro, P. Manríquez, J. M. L. Orensanz, A. M. Parma, J. Porobic, and C. Román. 2013. Sustainability of the Juan Fernández lobster fishery (Chile) and the perils of generic science-based prescriptions. Global Environmental Change 23(6):1381-1392. https://doi.org/10.1016/j.gloenvcha.2013.08.002

Ferreira-Rodriguez, N., O. Defeo, G. Macho, and I. Pardo. 2019. A social-ecological system framework to assess biological invasions: Corbicula fluminea in Galicia (NW Iberian Peninsula). Biological Invasions 21(2):587-602. https://doi.org/10.1007/s10530-018-1846-5

Fleischman, F. D., K. Boenning, G. A. Garcia-Lopez, S. Mincey, M. Schmitt-Harsh, K. Daedlow, M. C. Lopez, X. Basurto, B. Fischer, and E. Ostrom. 2010. Disturbance, response, and persistence in self-organized forested communities: analysis of robustness and resilience in five communities in Southern Indiana. Ecology and Society 15(4):9. https://doi.org/10.5751/ES-03512-150409

Flynn, C. D., and C. I. Davidson. 2016. Adapting the social-ecological system framework for urban stormwater management: the case of green infrastructure adoption. Ecology and Society 21(4):19. https://doi.org/10.5751/ES-08756-210419

Gao, Y., Q. Feng, and Z. X. Li. 2022. An assessment on the coordinated development of ecology, economy and society in Qilian Mountain National Park in Gansu section. Chinese Journal of Ecology 41:1197-1204.

Gao, Y., and Z. X. Li. 2022. Study of coupling coordination development of ecology and economy of Gansu area of Qilian Mountain National Park in China based on DFSR model. Fresenius Environmental Bulletin 31(7):7019-7030.

Gatiso, T. T., L. Kulik, M. Bachmann, A. Bonn, L. Bösch, D. Eirdosh, A. Freytag, S. Hanisch, M. Heurich, T. Sop, K. Wesche, M. Winter, and H. S. Kuhl. 2022. Effectiveness of protected areas influenced by socio-economic context. Nature Sustainability 5:861-868. https://doi.org/10.1038/s41893-022-00932-6

Govigli, V. M., A. Efthymiou, and K. Stara. 2021. From religion to conservation: unfolding 300 years of collective action in a Greek sacred forest. Forest Policy and Economics 131:14. https://doi.org/10.1016/j.forpol.2021.102575

Guerrero, A. M., and K. A. Wilson. 2017. Using a social-ecological framework to inform the implementation of conservation plans. Conservation Biology 31(2):290-301. https://doi.org/10.1111/cobi.12832

Guevara, L. E. T., A. Schlüter, and M. C. Lopez. 2016. Collective action in a tropical estuarine lagoon: adapting Ostrom’s SES framework to Ciénaga Grande de Santa Marta, Colombia. International Journal of the Commons 10(1):334-362. https://doi.org/10.18352/ijc.623

Guo, L., J. Li, Y. Huang, and J. Zuo. 2023. Exploring uncertainties of ecological management in the Qilian Mountain Nature Reserve, Northwest China. Journal for Nature Conservation 76:126507. https://doi.org/10.1016/j.jnc.2023.126507

Huang, Y., J. Fu, W. Wang, and J. Li. 2019. Development of China’s nature reserves over the past 60 years: an overview. Land Use Policy 80:224-232. https://doi.org/10.1016/j.landusepol.2018.10.020

Juffe-Bignoli, D., N. D. Burgess, H. Bingham, E. M. S. Belle, M. G. de Lima, M. Deguignet, B. Bertzky, A. N. Milam, J. Martinez-Lopez, E. Lewis, A. Eassom, S. Wicander, J. Geldmann, A. van Soesbergen, A. P. Arnell, B. O’Connor, S. Park, Y. N. Shi, F. S. Danks, B. MacSharry, N. Kingston. 2014. Protected planet report 2014. UNEP-WCMC, Cambridge, UK.

Kasymov, U., I. Ring, G. Gonchigsumlaa, N. Dejid, and L. Drees. 2023. Exploring complementarity among interdependent pastoral institutions in Mongolia. Sustainability Science 18(1):115-131. https://doi.org/10.1007/s11625-022-01198-9

Kimmich, C., M. H. Ehlers, E. Kellner, C. Oberlack, A. Thiel, and S. Villamayor-Tomas. 2023. Networks of action situations in social-ecological systems: current approaches and potential futures. Sustainability Science 18:1-10. https://doi.org/10.1007/s11625-022-01278-w

Kroner, R. E. G., S. Y. Qin, C. N. Cook, R. Krithivasan, S. M. Pack, O. D. Bonilla, K. A. Cort-Kansinally, B. Coutinho, M. M. Feng, M. I. M. Garcia, Y. F. He, C. J. Kennedy, C. Lebreton, J. C. Ledezma, T. E. Lovejoy, D. A. Luther, Y. Parmanand, C. A. Ruiz-Agudelo, E. Yerena, V. M. Zambrano, and M. B. Mascia. 2019. The uncertain future of protected lands and waters. Science 364(6443):881-886. https://doi.org/10.1126/science.aau5525

Leslie, H. M., X. Basurto, M. Nenadovic, L. Sievanen, K. C. Cavanaugh, J. J. Cota-Nieto, B. E. Erisman, E. Finkbeiner, G. Hinojosa-Arango, M. Moreno-Baez, S. Nagavarapu, S. M. W. Reddy, A. Sanchez-Rodriguez, K. Siegel, J. J. Ulibarria-Valenzuela, A. H. Weaver, and O. Aburto-Oropeza. 2015. Operationalizing the social-ecological systems framework to assess sustainability. Proceedings of the National Academy of Sciences of the United States of America 112(19):5979-5984. https://doi.org/10.1073/pnas.1414640112

Li, J., G. Q. Ma, J. H. Feng, L. Y. Guo, and Y. Z. Huang. 2022. Local residents’ social-ecological adaptability of the Qilian Mountain National Park Pilot, Northwestern China. Land 11(5):742. https://doi.org/10.3390/land11050742

Liu, J., T. Dietz, S. R. Carpenter, M. Alberti, C. Folke, E. Moran, A. N. Pell, P. Deadman, T. Kratz, J. Lubchenco, E. Ostrom, Z. Ouyang, W. Provencher, C. L. Redman, S. H. Schneider, and W. W. Taylor. 2007. Complexity of coupled human and natural systems. Science 317(5844):1513-1516. https://doi.org/10.1126/science.1144004

Mace, G. M. 2014. Whose conservation? Science 345(6204):1558-1560. https://doi.org/10.1126/science.1254704

McGinnis, M. D., and E. Ostrom. 2014. Social-ecological system framework: initial changes and continuing challenges. Ecology and Society 19(2):30. https://doi.org/10.5751/ES-06387-190230

Nagendra, H., and E. Ostrom. 2014. Applying the social-ecological system framework to the diagnosis of urban lake commons in Bangalore, India. Ecology and Society 19(2):67. https://doi.org/10.5751/ES-06582-190267

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

Palomo, I., C. Montes, B. Martín-López, J. A. González, M. García-Llorente, P. Alcorlo, and M. R. García Mora. 2014. Incorporating the social-ecological approach in protected areas in the Anthropocene. BioScience 64(3):181-191. https://doi.org/10.1093/biosci/bit033

Pan, J. F., Y. W. Ma, Y. Chen, S. Q. Cai, Y. M. Chen. 2022. Biodiversity assessment of Pudacuo National Park from the perspective of coupled social-ecological system. Journal of Ecology and Rural Environment 38:609-620.

Partelow, S. 2018. A review of the social-ecological systems framework: applications, methods, modifications, and challenges. Ecology and Society 23(4):36. https://doi.org/10.5751/ES-10594-230436

Rodríguez-Rodríguez, D., R. Larrubia, and J. D. Sinoga. 2021. Are protected areas good for the human species? Effects of protected areas on rural depopulation in Spain. Science of the Total Environment 763:144399. https://doi.org/10.1016/j.scitotenv.2020.144399

Schlüter, M., L. J. Haider, S. J. Lade, E. Lindkvist, R. Martin, K. Orach, N. Wijermans, and C. Folke. 2019. Capturing emergent phenomena in social-ecological systems: an analytical framework. Ecology and Society 24(3):11. https://doi.org/10.5751/ES-11012-240311

Schmitt-Harsh, M. L., and S. K. Mincey. 2020. Operationalizing the social-ecological system framework to assess residential forest structure: a case study in Bloomington, Indiana. Ecology and Society 25(2):14. https://doi.org/10.5751/ES-11564-250214

Stevenson, T. C., and B. N. Tissot. 2014. Current trends in the analysis of co-management arrangements in coral reef ecosystems: a social-ecological systems perspective. Current Opinion in Environmental Sustainability 7:134-139. https://doi.org/10.1016/j.cosust.2014.02.002

Strickland-Munro, J. K., H. E. Allison, and S. A. Moore. 2010. Using resilience concepts to investigate the impacts of protected area tourism on communities. Annals of Tourism Research 37(2):499-519. https://doi.org/10.1016/j.annals.2009.11.001

Vogt, J. M., G. B. Epstein, S. K. Mincey, B. C. Fischer, and P. McCord. 2015. Putting the “E” in SES: unpacking the ecology in the Ostrom social-ecological system framework. Ecology and Society 20(1):55. https://doi.org/10.5751/ES-07239-200155

Wang, S. J., C. L. Fang, and Y. Wang. 2015. Quantitative investigation of the interactive coupling relationship between urbanization and eco-environment. Acta Ecologica Sinica 35(07):2244-2254.

Wang, X. F., X. Huang, X. R. Zhang, Y. Yan, C. W. Zhou, J. T. Zhou, and X. M. Feng. 2022. Analysis of the spatio-temporal change of social-ecological system coupling: a case study in the Qinghai-Tibet Plateau. Global Ecology and Conservation 33:e01973. https://doi.org/10.1016/j.gecco.2021.e01973

Wang, Y. B., R. T. Zhao, Y. Li, R. Yao, R. X. Wu, and W. L. Li. 2023. Research on the evolution and the driving forces of land use classification for production, living, and ecological space in China’s Qilian Mountains Nature Reserve from 2000 to 2020. Environmental Science and Pollution Research 30:64949-64970. https://doi.org/10.1007/s11356-023-26857-x

Weng, G. M., Y. B. Tang, Y. Pan, and Y. Q. Mao. 2021. Spatiotemporal evolution and spatial difference of tourism-ecology-urbanization coupling coordination in Beijing-Tianjin-Hebei urban agglomeration. Economic Geography 41(12):196-204.

Yan, K., and Y. Ding. 2020. The exploration of China’s National Park System Pilot Project: taking Qilian Mountain National Park as an example. International Journal of Geoheritage and Parks 8(4):210-214. https://doi.org/10.1016/j.ijgeop.2020.11.008

Yang, Y., X. Li, and Q. Q. Liu. 2020. Coupling and cooperative evaluation of ecological environment and regional economic development: a case study of Hunchun Area in Amur Tiger and Leopard National Park. Journal of Northeast Forestry University 48(04):76-80.

Zhang, Y., X. Xiao, R. Cao, C. Zheng, Y. Guo, W. Gong, and Z. Wei. 2020. How important is community participation to eco-environmental conservation in protected areas? From the perspective of predicting locals’ pro-environmental behaviours. Science of the Total Environment 739:139889. https://doi.org/10.1016/j.scitotenv.2020.139889

Zhou, D., Z. Y. Wang, J. Lassoie, X. P. Wang, and L. X. Sun. 2014. Changing stakeholder relationships in nature reserve management: a case study on Snake Island-Laotie Mountain National Nature Reserve, Liaoning, China. Journal of Environmental Management 146:292-302. https://doi.org/10.1016/j.jenvman.2014.07.018