The co-evolution and driving mechanism analysis of ecosystem services value and tourism economic resilience from 286 cities in China

Abstract

Ecosystem service value (ESV) constitutes the ecological value foundation for enhancing tourism economic resilience (TER), serving as the central catalytic mechanism that drives the synergistic evolution of system disturbance resistance, recovery efficiency, and transformative adaptability. This study conducts a comprehensive evaluation of the interrelation between ESV and TER while exploring the key influencing factors to optimize human-environment interactions within the tourism system. The Coupling Coordination Degree (CCD) model is utilized to examine the association between ESs and TER across China. Additionally, spatial evolution patterns and key driving factors are explored using Exploratory Spatial Data Analysis (ESDA) and Geographically Weighted Regression (GWR). The key findings are as follows: (1) ESV displayed an inverted “N” trend, declining from southeast to northwest. (2) The spatial distribution of TER showed a persistent pattern of “high in the south, low in the north,” indicating a sustained stage of coordination. (3) The CCD of TER revealed positive spatial clustering, characterized by a distinct “two poles” pattern. (4) Economic and tourism-related factors were the dominant forces enhancing the CCD of TER, while social factors imposed constraints. These insights can support regional tourism economic planning and ecosystem management strategies.

Introduction

Nature constitutes the fundamental material basis for human survival and development, while the harmonious coexistence between humanity and nature represents the theoretical cornerstone of Marxist ecological civilization¹. Tourism, by nature, depends heavily on both natural resources and the environment, establishing a complex and interdependent relationship between tourism economies and ecosystems. Despite this connection, the rapid expansion of tourism often leads to disruptions, such as land degradation, biodiversity loss, and pollution of water resources^2,3. Such conflicts are visible in popular destinations like Zhangjiajie and Danaanu Island in Fiji^4,5. To mitigate these conflicts, the 20th CPC National Congress highlighted the need for ecological priority and sustainable development, underscoring the importance of examining the relationship between ESV and TER to foster coordinated growth. Tourism, as a strategic pillar industry of the national economy, serves as a critical conduit for realizing the ecological service value of ‘lucid waters and lush mountains.’ Within the policy framework of ecological civilization construction, it becomes an imperative academic endeavor to both theoretically and practically: (1) scientifically quantify the coupling coordination dynamics between ESV and TER, and (2) disentangle the drivers influencing their synergistic development.

Amidst global climate change and escalating resource constraints, research on ESV has emerged as a pivotal interdisciplinary frontier spanning environmental science, ecological economics, and sustainable agriculture⁶. ESV encompasses the tangible and intangible benefits humans derive from ecosystem structures, processes, and functions⁷. Secondary ESV classifications comprise 11 functional categories: food production, raw material supply, water provision, gas regulation, climate moderation, environmental purification, hydrological regulation, soil conservation, nutrient cycling, biodiversity maintenance, and aesthetic landscapes. Seminal work by Costanza et al.⁸ established the first global ESV assessment framework, quantifying 17 ESV through economic valuation models. Xie et al.⁹ advanced this methodology by developing China-specific equivalent factor calculations aligned with national land use patterns. Contemporary ESV estimation predominantly employs two methodological paradigms: (1) unit area value pricing and (2) equivalent factor-based valuation^10,11. Spatial analyses have been conducted across administrative units, grid scales, and natural geographic divisions^12,13. Key research foci encompass ESV quantification, predictive modeling, driving force analysis, and ecological zoning¹⁴. Empirical studies reveal significant spatial dependencies: Lai and Zoppi¹⁵ demonstrated an inverse ESV-urbanization relationship in Cagliari, Italy, while Tang et al.¹⁶ identified substantial ESV variations across land quality classes, highlighting ecological restoration’s multi-scalar value-creation mechanisms. Collectively, ESV scholarship exhibits multidimensional nature, with valuation techniques and driver analysis constituting critical scientific inquiries for addressing ecological preservation under global environmental change.

Resilience theory was first applied to economics by Reggiani et al.¹⁷, which eventually extended to encompass concepts like “economic resilience”¹⁸ and its evaluation¹⁹. In tourism, resilience—commonly referred to as TER—denotes the capacity of a tourism economic system to recover, stabilize, and adapt after external disruptions²⁰. Current research on TER predominantly focuses on four thematic domains: (1) measurement and evaluation frameworks, (2) analysis of influencing factors, (3) spatiotemporal evolution patterns, and (4) interlinkages with high-quality tourism development²¹. For instance, Zhang et al.²² operationalized resilience metrics through four dimensions—stability, diversification, innovation, and vitality. Similarly, Wang et al.²³ developed an assessment system incorporating resistance capacity and recovery efficiency. Methodologically, qualitative approaches including case studies, interviews, and questionnaire surveys dominate exploratory investigations²⁴, while quantitative paradigms employ econometric modeling and geographical detector techniques²⁵. Chinese scholarship predominantly operates at provincial, urban agglomeration, and major economic belt scales²⁶. While existing scholarship has provided crucial methodological frameworks, it exhibits a notable paucity in addressing regional heterogeneity and dynamic evolutionary mechanisms of tourism economic resilience at China’s prefectural-level city scale.

The interrelationship between tourism and ecosystems has been a central theme in tourism geography and ecology since the 1980s. Research initially centered around tourism’s impact on ecological capacity, carrying capacity, sustainability, and ecological safety^27,28,29. With increasing ecological pressures, there is a growing recognition of the need for balanced development between tourism and ecosystems³⁰. The interplay between tourism development and ESV exhibits dualistic dynamics. On the one hand, tourism activities exert substantial disturbances on natural ecosystems, while tourism-driven economic growth enhances destination resilience through resistance, recovery, and adaptive renewal mechanisms, constituting a critical pathway for ESV enhancement³¹. Specifically, Tourism industrialization accelerates tertiary sector dominance, catalyzing occupational shifts from agriculture to service industries, thereby facilitating farmland ecosystem restoration and soil carbon sequestration capacities. Resource-dependent destinations leverage stringent environmental regulations to harmonize tourism-human-land systems, optimizing ESV provisioning efficiency. Tourism revenue streams fund ecosystem conservation and restoration initiatives. Conversely, high-quality ecosystems serve as both attractors and stabilizers for tourism sustainability^32,33. Key synergies manifest through: Provisioning services (food, raw materials, water) underpinning tourism economic foundations; Regulating services (gas/climate/hydrological regulation) enhancing destination climate competitiveness; Cultural services (wetland/forest landscapes) functioning as core attractions and place-making assets; Supporting services (nutrient/soil maintenance) safeguarding ecological carrying capacity thresholds. However, exceeding ecological thresholds triggers nonlinear feedbacks: ecosystem degradation diminishes tourism suitability and economic resilience, while escalating ESV constraining effects emerge through diminished resource regeneration rates and amplified environmental remediation costs. The framework of the relationship between the ESV and TER is shown in Fig. 1.

Fig. 1

A framework for the relationship between the ESV and TER.

Contemporary research on ecological-environmental and tourism-economic interdependencies exhibits three salient characteristics: First, methodological evolution has shifted from qualitative descriptions to quantitative analysis, with CCD models becoming the predominant analytical framework. Second, spatial investigations predominantly operate at micro-to-meso scales, exemplified by case studies in Zhejiang Province, China³⁴ and Yangtze River Basin urban clusters³⁵. Third, conceptual scopes have expanded from dual-system coordination to polycentric coupling mechanisms involving multiple subsystems. These advancements have laid a crucial theoretical foundation for the present study. However, there remains a shortage of comprehensive national-scale analyses on the coordination and resilience of tourism development^36,37. To address this gap, understanding the interconnectedness between ESV and TER at the municipal level is essential for guiding sustainable tourism development in China.

This study constructs an evaluation index system to analyze the coupling and coordination between ESV and TER at the municipal level in China, using data from 2012 to 2022. It investigates the spatial and temporal evolution of coordination through the CCD model and identifies key drivers through geographically weighted regression analysis. This approach offers a new perspective by quantifying TER at the inter-municipal level, elucidating the spatial and temporal dynamics of ESV and TER, and contributing to empirical research on sustainable tourism and ecosystem coordination (Fig. 2).

Fig. 2

Research flow chart.

Materials and methods

Study area

China offers a unique context for examining the interaction between ESV and tourism due to its diverse natural and cultural landscapes. As the largest developing nation globally, China is home to 15 World Natural Heritage Sites (e.g., Jiuzhaigou, Sichuan Giant Panda Habitat, Tianshan Mountain in Xinjiang), 40 World Cultural Heritage Sites (e.g., the Great Wall, Mogao Grottoes in Dunhuang), and 224 national scenic areas³⁸. By the end of 2022, tourism revenue had reached 2.04 trillion yuan, with 2.53 billion domestic tourist visits, solidifying tourism’s significant role in China’s economy.

In line with the 14th Five-Year Plan, China aims to further develop over 200 cities within 19 urban agglomerations, such as Beijing-Tianjin-Hebei and the Yangtze River Delta³⁹. The nation’s geographic and ecological diversity includes forests, grasslands, wetlands, deserts, and alpine tundra, along with a complex topography featuring plateaus, hills, and mountainous regions like the Tibetan Plateau, Sichuan Basin, and Yunnan-Guizhou Plateau⁴⁰. This study therefore focuses on the city level, using data from 286 prefecture-level cities while excluding those with significant data deficiencies (Fig. 3).

Fig. 3

Study area. Figure was generated using ARCMAP 10.2, without any modifications to the boundaries of the base map.

Data source

The data used in this study include land use, socio-economic indicators, and tourism economic statistics. Land use data were gathered for six different time periods: 2012, 2014, 2016, 2018, 2020, and 2022. These datasets were sourced from the China Land Use Monitoring Database, which has a spatial resolution of 30 m by 30 m. Socio-economic information, such as GDP, agricultural output, and food prices, was collected from the China Economic and Social Big Data Statistical Platform (CESBDSP). Tourism economic data were mainly derived from the China Statistical Yearbook and annual statistical reports published by various cities on economic and social development. Land use categories—including forestland, grassland, cropland, watersheds, construction land, and unutilized land—were delineated using ArcGIS software for spatial analysis.

Methodology

Estimating ecosystem service value (ESV)

To estimate the ESV in China, relevant models were adapted based on existing frameworks by Costanza et al.⁸, Xie et al.⁹, and Zhao et al.⁴¹. The study used wheat, corn, and rice as representative crops to calculate the ecosystem service values. Data on per-unit yield, cultivated area, and average price from 2012 to 2022 for these crops were used to derive an economic valuation of ESV, calculated at 1085.94 yuan per hectare. This valuation was determined using Eq. (1), and the equivalent value per unit area is presented in Table 1. Subsequently, ESV estimates were derived using Eq. (2). Construction land was excluded due to its lack of ecosystem service function.

Table 1 ESV per unit area by land use type (yuan/hm²).

$${E_C}=1/7\sum\limits_{{i=1}}^{n} {\left( {{P_i} \times {q_i} \times M} \right)}$$

(1)

Where E_c: Economic value of ESV per hectare (yuan/hm2); P_i: Average national grain price of crop i (yuan/kg); qi: Average yields per unit area for crop i (kg/hm²); M: Acreage under cultivation; n: Number of crop types.

$$ESV=\sum\limits_{{k=1}}^{m} {\left( {{A_K} \times V{C_k}} \right)}$$

(2)

Where ESV: Ecosystem service value; Ak: Area of land type k; VCk: Equivalent value coefficient for land type k.

Developing an index system for tourism economic resilience (TER)

The index system for assessing TER was developed following prior studies by Yang and Qin⁴² and Yang et al.⁴³. Four dimensions—defensive capability, recuperative capacity, reconstructive capacity, and update capability—were defined, each representing specific aspects of resilience (Table 2). Defensive Capability assesses resilience to external shocks and is measured by tourism resource assets, economic scale, and local economic base. Recuperative capacity measures the system’s ability to adapt and recover from disruptions, focusing on pollution control levels, green infrastructure, and tourism facilities. Reconstructive capacity examines the labor force and economic contributions to tourism.

Table 2 Index system of TER.

Coupling coordination degree (CCD) model

The degree of coupling coordination (CCD) between TER and ESV was calculated using Eq. (3):

$$D={\left( {{U_1} \times {U_2}} \right)^{0.25}}$$

(3)

Where D is the degree of coupling coordination (ranging from 0 to 1); U₁ and U₂ are Standardized indices of TER and ESV. The classification of coupling coordination levels is shown in Table 3.

Table 3 Classification of coupling coordination levels.

Kernel density Estimation

Kernel Density Estimation (KDE) was applied to objectively analyze the spatial distribution characteristics of CCD:

$$f\left( x \right)=\frac{1}{{nh}}\sum\limits_{{i=1}}^{n} {K\left( {\frac{{x - {x_i}}}{h}} \right)}$$

(4)

Where K(·) represents the kernel function, $n$ is the number of observations, and $h$ is the bandwidth estimated using Gaussian KDE.

Spatial autocorrelation analysis

Spatial Autocorrelation Analysis was conducted to explore regional spatial correlation, employing the following local autocorrelation formula:

$$Moran\text{'}s\, I=\frac{n}{S}\sum\limits_{{i=1}}^{n} {\sum\limits_{{j=1}}^{n} {{W_{ij}}\left( {{Y_i} - \overline {Y} } \right)\left( {{Y_j} - \overline {Y} } \right)} }$$

(5)

Where Y_i, Y_j: Values of the regions i and j; $\overline {Y}$: Mean value; W_ij: Spatial weight matrix; S: Variance of the data.

Geographically weighted regression (GWR)

The GWR model was used to analyze the spatial heterogeneity and changes in ESV and TER across China over time:

$${y_i}={\beta _0}\left( {{u_i},{v_i}} \right)+\sum\limits_{{i=1}}^{k} {{\beta _k}\left( {{u_i},{v_i}} \right){x_{ik}}+{\varepsilon _i}}$$

(6)

Where ${y_i}$: Dependent variable for location i; $\left( {{u_i},{v_i}} \right)$: Coordinates for sample point i; ${\beta _k}\left( {{u_i},{v_i}} \right)$: Regression coefficient for variable k: at point i; ${\varepsilon _i}$: Error term.

Results

Evaluation of ESV and TER

Assessment of ESV

Between 2012 and 2022, the ESV in China demonstrated an overall declining trend with fluctuations (Table 4). The total ESV decreased from 11,113.72 billion yuan in 2012 to 10,800.99 billion yuan in 2022, representing a net decline of approximately 48.72 billion yuan, or 0.45%. Initially, the ESV experienced a sharp drop from 11,113.72 billion yuan in 2012 to 10,848.31 billion yuan in 2014—a reduction of 2.39%. Afterward, it showed a slight recovery, reaching 10,856.92 billion yuan in 2018 (a growth rate of 0.08%), although it remained below its 2012 level. The value subsequently continued to decline, decreasing by another 55.92 billion yuan from 2018 to 2022.

Table 4 ESV in China (Unit: CNY 100 Million) and percent changes.

Throughout this period, the composition of ESV remained relatively stable in its structure. The Regulating Service Value (RSV) constituted the largest share of total ESV at 67.77%, followed by the Support Service Value (SSV) at 21.97%. The Provision Service Value (PSV) and Cultural Service Value (CSV) made up smaller portions, accounting for 5.81% and 4.45%, respectively. These findings suggest that the primary function of China’s natural ecosystems is to regulate environmental quality and maintain ecological stability. All four categories of ESV—PSV, RSV, SSV, and CSV—experienced varying degrees of decline over the decade, with PSV showing the most pronounced reduction at 3.77%. The trend in each category followed a similar pattern: a sharp decline, a partial recovery, and a subsequent reduction, mirroring the overall change in total ESV.

The spatial distribution of ESV across China between 2012 and 2022 revealed a gradient of higher values in the east, decreasing toward the northwest (Fig. 4). This spatial pattern is largely due to the greater ecological value associated with forest areas, which are predominantly located in the northeast and southwest regions. Forested and mountainous regions, such as the Greater and Lesser Xing’an Mountains, play a significant role as ecological barriers, thereby contributing to higher ESV. Additionally, changes in human activities, such as urbanization and economic development, have led to shifts in ecosystem functions, resulting in discernible spatial disparities.

Fig. 4

Spatial Distribution of ESV in China (2012–2022). Figure was generated using ArcMap 10.2, without any modifications to the boundaries of the base map.

Throughout the decade, the general spatial pattern of ESV remained mostly unchanged, with only some shifts in the extent of the highest and lowest value areas. In 2012, areas with high ESV were primarily located in regions like Hulunbuir in Inner Mongolia and Chongqing Municipality. However, by 2018, the highest-value areas had contracted, and regions like Chongqing transitioned from high to lower values. Conversely, low-value areas, such as Yancheng, Lijiang, and Anshun, expanded during this period. Between 2018 and 2022, both high-value and low-value areas expanded, with Chongqing returning to high-value status, while Guiyang and Huai’an dropped to the lowest category.

Assessment of TER

Between 2012 and 2022, the level of TER across Chinese cities exhibited phases of fluctuation, followed by an overall increase. The TER index grew from 0.248 in 2012 to 0.339 in 2022, reflecting a cumulative increase of 36.69%. This improvement highlights a gradual strengthening in the resilience of China’s tourism sector, indicating the effectiveness of tourism development initiatives during the decade. The temporal evolution of TER can be divided into three distinct phases: a period of rapid growth from 2012 to 2018, marked by an average annual increase of 7.18%; a decline from 2018 to 2020 largely attributable to the COVID-19 pandemic; and a steady recovery from 2020 to 2022, with an annual growth rate of 6.1%.

Using natural breaks, TER levels were classified into five categories over the study period. The proportion of cities at the highest TER level initially expanded, peaking at 87.76% in 2018, before later contracting. Cities with higher TER levels also experienced modest changes, expanding and then shrinking, with an overall change of 6.64%. Meanwhile, cities in the medium and lower TER levels experienced alternating phases of expansion and contraction, reaching their lowest proportion in 2018. In contrast, the lowest TER category demonstrated considerable variability, showing an initial decrease, followed by a substantial increase of 66.78% between 2018 and 2020, and then a subsequent decline (Table 5).

Table 5 Percentage of cities by TER level in China (2012–2022).

Spatially, TER across China revealed significant shifts between 2012 and 2022, with higher values concentrated in the southeast and lower values in the northwest (Fig. 5). From 2012 to 2018, low-level TER zones contracted significantly, as regions in the southeast, parts of the Yellow River basin, and the northeast saw increases in TER to higher levels. However, between 2018 and 2020, the COVID-19 pandemic led to a notable decline in TER, with cities in the lowest and lower levels accounting for over 94% of the total. Between 2020 and 2022, the lowest-level TER zones began contracting again, as cities like Heihe, Qiqihar, Jixi, and Zhangzhou moved up to the lower-level category. In parallel, medium-level zones expanded, with cities such as Jiamusi, Zhanjiang, Zhangjiakou, and Chengde advancing from lower to medium levels, while some medium-level cities—such as Wuwei, Baoshan, Meizhou, and Huizhou—progressed to the highest category.

Fig. 5

Spatial Distribution of TER Levels in China (2012–2022). Figure was generated using ArcMap 10.2, without any modifications to the boundaries of the base map.

Time evolution analysis of CCD

Using the CCD model, the CCD of ESV and TER in Chinese cities from 2012 to 2022 was analyzed, dividing the dataset into eastern, central, and western regions for comparison (for regional classifications, please refer to http://www.stats.gov.cn/) (Table 6). Results indicate that the CCD exhibited an overall “N”-shaped growth trajectory, with average CCD values increasing from 0.544 in 2012 to 0.588 in 2022, peaking in 2018 at 0.700. Initially, the coordination level increased from elementary to intermediate but later reverted to elementary coordination. Throughout the study period, the overall CCD remained relatively low, indicating that further improvements in inter-system coordination are necessary.

Table 6 Regional CCD for ESV and TER (2012–2022).

From a regional perspective, discrepancies existed in CCD across different areas: the western region demonstrated above-average CCD, owing to active integration into the “Belt and Road” initiative and strong ecological protection, which benefited the tourism industry. The central region’s CCD was close to the national average, whereas the eastern region remained below the average level. At the municipal level, differences in economic conditions, political backing, and resource availability resulted in varied CCD levels. Hulunbuir had the highest CCD, which increased from 0.778 in 2012 to 0.922 in 2018, before dropping to 0.752 in 2022, transitioning from intermediate to advanced coordination. The diversity of Hulunbuir’s natural ecosystems, coupled with ongoing tourism development, has positively impacted CCD. Chongqing Municipality followed a similar trend, while several cities in the eastern and western regions, such as Liaoyuan, Yuncheng, Hebi, and Chaoyang, had lower but stable CCDs, averaging below 0.57 with minimal fluctuations.

A regional perspective reveals clear disparities in CCD. The western region consistently demonstrated a CCD above the national average, benefiting from strong ecological preservation measures and integration into the “Belt and Road” initiative. This actively promoted tourism, contributing to improve CCD. The central region’s CCD remained close to the national average, whereas the eastern region fell slightly below average, indicating room for improvement in achieving balanced growth between ESV and TER.

At the municipal level, differences in CCD across cities could be attributed to factors like economic development, policy support, and resource availability. Cities such as Hulunbuir exhibited the highest CCD, rising from 0.778 in 2012 to 0.922 in 2018 before declining to 0.752 in 2022. This trend reflects a progression from intermediate to advanced coordination, largely driven by Hulunbuir’s rich natural ecosystems and focused tourism efforts. Chongqing showed a similar pattern, albeit with lower values. In contrast, cities in both the eastern and western regions, including Liaoyuan, Yuncheng, Hebi, and Chaoyang, maintained lower but relatively stable CCDs, averaging below 0.57 with minimal fluctuations throughout the study period.

On a national scale, several observations were made regarding CCD evolution: firstly, during the COVID-19 pandemic, the gravity center of the density distribution curve shifted to the right and subsequently back to the left, indicating initial declines followed by recovery. Secondly, the right tail of the curve shortened over time, indicating a narrowing gap between the CCD levels in Chinese cities and an increase in the number of cities with intermediate coordination. Lastly, a distinct double-peak pattern appeared in 2014, which gradually evolved into a single peak, suggesting improvements in CCD polarization. Analysis of the three regions showed that the CCD of each region generally followed a single-peak pattern (Fig. 6).

Fig. 6

Kernel density map of CCD for the nation and the three regions.

Moreover, the distribution centers of the CCD exhibit a “right-shift-left-shift-right-shift” trajectory. In the west, the distribution curve of nuclear density changes from “high peak-low wave width” to “low peak-high wave width”, with a large difference in the whole. In 2014, the heights of the density distribution curves in the eastern and central parts of the country China both changed significantly, showing a phased fluctuation trend, with the height of the main peak falling rapidly from 2012 to 2014 and rising from 2014 to 2022, with the stability to be strengthened. The central portion has the most pronounced right trailing feature, then the western portion, which continues to weaken, while the eastern portion has a less pronounced right trailing feature. Over time, the curves in the eastern and central regions have evolved in a “single-peak-bimodal-single-peak” pattern, suggesting that the problem of intraregional variability in the regions mentioned above has been alleviated as a whole.

Spatial evolution analysis of CCD

To explore the spatial dynamics and heterogeneity of the CCD between ESV and TER across China, six distinct time points (2012, 2014, 2016, 2018, 2020, and 2022) were analyzed (Fig. 7). Findings revealed a spatial distribution pattern characterized by “high values along the periphery, low values in the center, elevated values in the south, and reduced values in the north,” reflecting distinct core-periphery features. From 2012 to 2022, the CCD exhibited an overall rising trend, with areas of high CCD gradually expanding outward from the Pearl, Yangtze, and Songhua River regions, while lower CCD areas within central China showed signs of contraction.

Fig. 7

Spatial differentiation characteristics of CCD. Figure was generated using ArcMap 10.2, without any modifications to the boundaries of the base map.

In 2012, the CCD landscape across China was primarily classified into three categories: 93.7% of cities (268 cities) were in the elementary coordination stage, 3.15% (9 cities) were in borderline disorder, and another 3.15% were in intermediate coordination. The spatial pattern showed a gradient reduction from the southern and northern parts of China toward the central regions. This was particularly evident in the middle and lower Yellow River basins, including cities such as Rizhao, Linyi, Anyang, Hebi, and Zhoukou. Although tourism activities had elevated regional economic standings, unsustainable practices had negatively impacted ESV, resulting in low CCD in these regions.

Between 2012 and 2018, regions characterized by elementary coordination or borderline disorder shrank, while intermediate coordination areas expanded. This shift represented a slight positive spatial evolution, with cities like Tianjin and Shijiazhuang advancing to intermediate coordination. Cities including Rizhao and Linyi also moved from borderline to intermediate coordination. However, the period between 2018 and 2022 witnessed the adverse effects of the COVID-19 pandemic, which significantly lowered TER nationwide and, by extension, the coupling coordination between tourism and ecosystems. As a result, regions exhibiting intermediate and excellent coordination regressed, and areas with elementary coordination or borderline disorder expanded. For instance, Hulunbuir and Chongqing fell from excellent to intermediate coordination, while Heihe, Wuwei, and other cities transitioned from intermediate to elementary coordination. Cities such as Jiaxing fell into the borderline disorder category.

During the recovery phase from 2020 to 2022, a resurgence of intermediate coordination was observed, with 70 cities—including Heihe, Pu’er, and Wuwei—progressing from elementary to intermediate levels. Concurrently, areas marked by borderline disorder contracted, with cities like Huanggang and Jiaxing improving from borderline to elementary coordination. These shifts highlight a gradual balancing of CCD, despite the challenges posed by external disruptions like the pandemic.

Spatial correlation analysis of CCD

To gain a deeper understanding of the spatial dependencies in the CCD of ESV and TER, spatial correlation analyses, including both global and local Moran’s I, were performed and visualized through Local Indicators of Spatial Association (LISA) clustering (Fig. 7).

The global Moran’s I results, presented in Table 7, ranged from 0.234 to 0.691, indicating a significant positive spatial autocorrelation in CCD across Chinese cities (p < 0.01). Specifically, the Moran’s I value increased from 0.331 in 2012 to a peak of 0.691 in 2014, suggesting enhanced spatial clustering and an overall rise in CCD values, which contributed to increased spatial differentiation. However, between 2014 and 2022, the Moran’s I showed a wave-like decline to 0.251, signaling weakened clustering and reduced spatial disparities. This implies that the CCD among the 286 cities evolved towards more spatial uniformity and balance over time.

Table 7 Global Moran’s I results for CCD (2012–2022).

Local Moran’s I analysis for the years 2012, 2014, 2016, 2018, 2020, and 2022 yielded LISA clustering maps (Fig. 8) which further illustrated localized spatial variations. In 2012, “high-high” clusters were observed in 17 cities, including Ordos, Tongliao, and Lincang, indicating areas where both CCD and neighboring cities’ CCD were high. “High-low” clusters were predominantly concentrated in cities like Chongqing and Zibo, while “low-high” clusters were identified only in Yingtan. Cities exhibiting “low-low” clustering, indicating low CCD values surrounded by similarly low areas, included Weihai, Yantai, and Weifang.

Fig. 8

LISA clustering diagram of CCD. Figure was generated using ArcMap 10.2, without any modifications to the boundaries of the base map.

By 2014, the number of “high-high” clusters had increased to 31 cities, extending to regions such as Yuxi and Pu’er. Conversely, the “low-low” clusters expanded to 56 cities, encompassing areas like Weihai, Yantai, and Dalian. However, by 2016, the number of “high-high” clusters had decreased to 16 cities, with notable reductions in areas like Lincang and Chifeng. Meanwhile, the “low-low” clusters contracted to 12 cities, including Dalian and Anshan. Similar patterns of change persisted in subsequent years, with fluctuations in the numbers of “high-high” and “low-low” clusters, largely reflecting regional economic and environmental shifts.

By 2022, the “high-high” clusters remained largely stable, while “low-low” clusters expanded to include Zhoukou, suggesting lingering spatial disparities despite progress in other regions. The analysis of spatial clustering revealed a consistent trend of “high-high” clusters predominantly in central and western areas, while “low-low” clusters were more frequent in the eastern region. This pattern indicates that, while some progress has been made in narrowing the spatial gap and promoting balanced development, notable differences among regions remain.

Driving mechanisms behind CCD

To understand the key drivers that influence the CCD between ESV and TER, this study draws upon prior research⁴⁴ and available data to assess four dimensions: natural, economic, social, and tourism factors. (1) Natural factors influence the CCD by regulating the dynamic matching between ecosystem service supply and tourism economic demand. Mean annual precipitation variation restructures water resource allocation patterns, constraining the spatiotemporal adaptability between ecological regulation services and tourism facility carrying capacities. Mean annual temperature fluctuations alter vegetation phenological rhythms, disrupting the co-evolutionary pathway between landscape aesthetic services and seasonal tourism demand. NDVI dynamics reflect vegetation coverage threshold effects, with its degradation directly weakening the positive feedback mechanism between ecological barrier functions and tourism economic resilience. (2) Economic factors exert dual regulatory effects through scale-structure mechanisms on coupling coordination. GDP growth drives tourism infrastructure expansion, enhancing short-term economic resilience while generating spatial crowding-out effects on ecological regulation services, thereby reducing coupling efficiency. Economic recession conversely stimulates ecological resource capitalization, reconstructing resilience cultivation pathways through service value realization. (3) Social factors mediate coupling coordination via spatial restructuring and threshold effects. Road network density enhancement improves tourism flow accessibility, yet infrastructure sprawl induces ecological patch fragmentation, undermining regulatory service stability. Population density intensification accelerates tourism consumption demands, triggering resilience attenuation feedback when exceeding community ecological carrying thresholds. (4) Tourism factors bilaterally regulate coupling coordination through agglomeration-structure mechanisms. A-grade attraction proliferation creates brand siphon effects that temporarily boost tourism economic resistance, while overexploitation induces ecological stress responses that compromise cultural service elasticity. Star-rated hotel expansion optimizes reception capacity thresholds but competes for spatial carriers of ecological regulation services through land-use rivalry.

To perform a comparative analysis, two key years, 2012 and 2020, were chosen. The driving factors served as explanatory variables, and CCD was the dependent variable. The analysis was conducted using Geographically Weighted Regression (GWR), allowing an understanding of spatial variations across different regions. Given the significant spatial positive correlation in the CCD between ESV and TER, the application of traditional OLS models for regression analysis may overlook spatial dependencies. Consequently, we sequentially implemented OLS regression followed by GWR model validation. The regression results indicated that the mean variance inflation factor (VIF = 4.8) for all explanatory variables in both 2012 and 2022 remained below 7.5, confirming the absence of multicollinearity in the model. Based on the adjusted R-squared (R² values derived from the GWR model for these two representative years, the explanatory power reached 50.92% and 55.77%, respectively. These results represent improvements of 3.49% and 9.17% in goodness-of-fit compared to the OLS model, accompanied by lower Akaike Information Criterion corrected (AICc) values. These findings collectively demonstrate the superior fitting performance of the GWR model in capturing spatially heterogeneous relationships (Table 8).

Table 8 Overview of selected factors and their characteristics.

The GWR analysis revealed that the drivers of CCD varied spatially across China, showing distinct patterns for each dimension. The findings highlighted that economic and tourism factors exerted the most substantial positive effects on CCD, whereas social factors had a predominantly negative influence. The natural factors demonstrated mixed effects, contributing positively and negatively to CCD in different regions. The driving factors, ordered by the magnitude of their influence, are as follows: economic > tourism > social > natural (Fig. 9). The contributions of the four dimensions to CCD are outlined in greater detail below:

Fig. 9

Drivers of spatial differentiation in CCD between ESV and TER.

(1) Natural factors

The influence of natural factors on the CCD of ESV and TER has noticeable regional variability and temporal dynamics. Annual precipitation has a positive effect on CCD, with regression coefficients ranging from 0.119 to 0.350 in 2012 to 0.076–0.26 in 2022. Over this decadal period, the spatial development pattern transitioned from north-south zonal stratification to east-west zonal stratification, while the magnitude of this climatic effect demonstrated a gradual attenuation trend. This spatial difference is the result of synergies between the evolution of the climate system and human activities. In the semi-arid north, a 100 mm increase in precipitation significantly increases the net primary productivity of vegetation, whereas in the humid south, the marginal ecological benefits are sharply reduced by exceeding the vegetation water demand threshold. In addition, the Northwest relies on eco-tourism projects to maintain high coupling, while the South reduces precipitation sensitivity through maturing urbanisation and widespread irrigation facilities. The positive impact of mean annual precipitation is particularly significant in the area north of the middle and lower reaches of the Yangtze River. At the same time, the low-value zones are mostly clustered in the area south of the middle and lower reaches of the Yangtze River, although their intensity decreases (Fig. 10a1,b1). The negative effect of annual mean temperature on the coupling of the two is a global and continuous weakening, with regression coefficients ranging from − 0.297 to − 0.08 in 2012 to − 0.171 to − 0.02 in 2022, mainly due to the differences in regional climate adaptive capacity and industrial transformation process. In the northeast, as a high negative impact area, permafrost degradation has weakened the ecological regulation function of the black soil area, and the low-temperature-dependent ice and snow tourism economy is more sensitive to warming. In the south, the low negative impact zone extends from southwest to southeast, with the southwest mountainous areas strengthening ecological resilience through ecological compensation mechanisms, and the southeast coast promoting the low-carbon transformation of the tourism industry based on the penetration of the digital economy, which effectively reduces the dependence on the natural climate (Fig. 10a2,b2). The positive influence of NDVI on the coordination of the two couplings shows a gradient attenuation of ‘high in the west and low in the east’, with regression coefficients ranging from 0.039 to 0.34 in 2012 to 0.037–0.24 in 2022, and its spatial differentiation is closely related to the life cycle of tourism sites and the transformation of industrial structure. High-value areas are mainly distributed along the first step of China’s terrain, limited by the singularity of tourism products in the alpine region and the threshold of environmental carrying capacity, the marginal benefits of ecological gains to economic resilience diminish. The formation of low-value zones east of Daxinganling stems from the characteristics of tourism economic transformation in the old industrial base of Northeast China, i.e., the industrialisation of ice and snow tourism weakens the dependence on vegetation. In addition, the geographical gradient difference is further regulated by the supply and demand of the tourism factor market, and the western part relies on the pilot of the national park system (e.g., Sanjiangyuan eco-tourism) to extend the ecological value transformation chain, while the eastern part breaks through the natural constraints through the construction of intelligent tourism system (e.g., digitalisation of the Changbaishan Mountain scenic area), which forms a regional adaptive differentiation in the efficacy of the role of the NDVI (Fig. 10a3,b3).

Fig. 10

Spatial distribution of regression coefficients for the GWR model (2012 vs. 2022). Figure was generated using ArcMap 10.2, without any modifications to the boundaries of the base map.

(2) Economic factors

The economic factor exerts a significant and continuously increasing positive influence on the CCD of ESV and TER. In 2012, GDP per capita significantly positively influenced the coupling and coordination of the two, mainly in the southeastern region of China, with regression coefficients ranging from 0.615 to 0.88. By 2022, the positive impact of GDP per capita continued to increase, with regression coefficients ranging from 0.785 to 0.88, expanding over time in the region south of the Qinling-Huaihe River, showing a gradient pattern of gradual weakening from south to north. The reason for this is that the southeastern part and the area south of Qinling-Huaihe River accelerate the upgrading of eco-tourism consumption preference by relying on economic advantages, and strengthen the ecological service premium capacity through the product supply side innovation such as lodging clustering and intelligent scenic spots; the northern part is subject to the traditional path-locking effect of the resource-based economy, and the tourism development is inclined to the construction of artificial scenic spots, and the ecological compensation is lagging behind, which results in the spatial decay of the transformation of the economic gain to ecological coordination, and confirms that This confirms the stage gap between ‘ecological premium’ and ‘resource realisation’ in the life cycle theory of tourist sites (Fig. 10a4,b4).

(3) Social factors

Social factors adversely influence the capacity for the CCD of ESV and TER. The impact of road network density on the coupling and coordination of the two is shown as a continuous negative impact, with regression coefficients ranging from − 0.186 to − 0.17 in 2012 to − 0.348 to − 0.34 in 2022. In 2012, it is shown as a north-south hierarchical belt distribution pattern, and the high value area of negative impact is concentrated in the northeastern region, reflecting the encryption of the road network in the transition period of heavy industry to aggravate the decline of ecological services, such as the expansion of construction land in Shenyang, and other resource cities with elevated traffic-orientated hardening of land in the development of tourism. By 2022, the negative impact is enhanced globally, and the high value area of negative impact is concentrated east of the Hu Huanyong Line, showing a gradient pattern of gradual attenuation from southeast to northwest, mapping the stage of scale expansion of the tourism economy, with the high-density road network in the southeast accelerating ecological overload by shortening tourists‘ spatial and temporal distances, such as a surge in the annual tourists’ density in the West Lake scenic area; and the low-value area in the northwest is constrained by the ecological red line to maintain the stock of natural capital, forming a gradient decay pattern. This spatial heterogeneity reveals a nonlinear conflict between infrastructure efficiency and ecological thresholds in the development process of tourist destinations (Fig. 10a5,b5). The influence of population density on the coordination of the two couplings also shows a continuous negative influence, which gradually increases with time, with regression coefficients ranging from − 0.216 to − 0.13 in 2012 to − 0.538 to − 0.4 in 2022. The area of high negative influence evolves from southwest to northwest, mainly located west of the Hu Huanyong line. In contrast, the area of low negative influence gradually turns from northeast to southeast, primarily located east of the Huanyong line. This development trend is opposite to the road network density. The reason is that the eastern region, through industrial upgrading and technological inputs, achieves a dynamic balance between population concentration and intensive management of ecotourism, forming a negative eastward shift towards the low value zone. In contrast, the area west of the Hu Huanyong line has a low ecological vulnerability threshold, and the increase in population density will accelerate the expansion of the ecological footprint. This spatial variation reveals that there is a significant geographical threshold for the response of ecotourism system to population pressure, and the contradiction between the rough tourism development mode and ecological sensitivity in the western region deepens with the increase of population density (Fig. 10a6,b6).

(4) Tourism factors

Tourism has a significant positive effect on the CCD of ESV and TER, and the effect continues to strengthen over time. The regression coefficients of tourism attractiveness ranging from 0.196 to 0.42 in 2012 to 0.313–0.40 in 2022. The developmental trajectory transitioned from a “north-high-south-low” spatial configuration to a tiered “west-high-east-low” gradient pattern over this decade, especially the low value area in the southwestern part of the country jumps up to the high value area, which is unstable, reflecting the dynamic mechanism of the regional resource endowment and policy orientation. Reflecting the dynamic game mechanism between regional resource endowment and policy orientation. Specifically, the early pattern of ‘high in the north and low in the south’ originated from the advantages of cultural heritage clustering and the dividends of policy pilots in northern regions such as Xi’an, while the homogenisation of traditional scenic spots in the south led to the decline of marginal benefits. In the later period, the ‘West high, East low’ stepped leap reflects the accelerated capitalisation of ecological resources in the West, and the tourism attractiveness index of Southwest China, such as Northwest Yunnan, has risen through the innovation of eco-tourism products and the development of national culture IPs. (Fig. 10a7,b7). The north-south divergence of accommodation capacity to the coupling coordination of the two stems from the difference in regional resource endowment and the suitability of the industrial model, with regression coefficients ranging from 0.173 to 0.51 in 2012 to 0.569–0.62 in 2022. Qinling-Huaihe River south of the high value area relying on subtropical ecological resources high service value, through the construction of lakes and other ecologically friendly hotel cluster layout, to achieve ecological capital to the economic resilience of the efficient transformation; while the north is limited by the temperate ecosystem threshold constraints, only grassland ecological restoration cycle of more than 10 years, the crude lodging expansion led to unit bed ecological footprint higher than the south. In addition, the chaining rate of star-rated hotels in the north is significantly lower than that in the south, and the integration of culture and tourism is lagging behind, resulting in a continuous weakening of the coupling coordination. (Fig. 10a8,b8).

Discussion

This study conducted a comprehensive analysis of the temporal and spatial evolution of the CCD between ESV and TER across 286 cities in China from 2012 to 2022. Four key dimensions—natural, economic, social, and tourism—were considered to evaluate their influence on the dynamics of CCD. The principal findings of the study are summarized below.

First, the ESV in China from 2012 to 2022 exhibited an inverted “N” shaped trend, a pattern consistent with the findings of Liu et al.⁴⁵, who specifically studied the southwestern region of China. The initial period (2012–2014) was marked by significant environmental degradation, leading to a decrease in ESV of approximately 265.4 billion yuan. During this time, China implemented the Air Pollution Prevention Action Plan in 2013, targeting reduced coal consumption and improvements in air quality by 2017. Furthermore, China’s commitment to peak carbon emissions in 2014 successfully mitigated ecological decline⁴⁶. On the other hand, the changes in the TER index were more prominent, showing a spatial distribution with higher values in southeastern China and lower values in the northwest, reflecting the disparity in economic development and tourism infrastructure across regions. The southeastern areas, characterized by socio-economic affluence and diverse tourism resources, were more resilient to external disruptions, a conclusion that corroborates Yang et al.⁴⁷. However, the COVID-19 pandemic in 2019 severely disrupted tourism nationwide, leading to an abrupt drop in TER.

Second, the spatial distribution of CCD across China followed a “high surrounding, low central, high in the south, low in the north” pattern, with an observed regional hierarchy of west > central > east. This distribution aligns with findings from Gu et al.⁴⁸, who reported a similar pattern in western Hunan. The rapid growth of China’s tourism economy has often come at the cost of ecosystem degradation due to poor planning and unsustainable practices⁴⁹. Meanwhile, regions that began developing tourism later, such as those in western China, have focused on ecotourism, wellness, and nature-based tourism. These emerging forms of tourism have leveraged favorable natural conditions to boost ESV and foster sustainable development⁵⁰. As a result, regions like Inner Mongolia have capitalized on their grasslands for eco-friendly tourism, while Yunnan has adopted a green approach to tourism leveraging its rich natural resources. In contrast, cities like Shanghai, although highly popular as tourist destinations, have faced challenges in maintaining CCD due to dense populations and significant modifications to natural landscapes, resulting in environmental strain.

Third, the exploration of factors influencing CCD between ESV and TER considered four dimensions: natural, economic, social, and tourism. Economic and tourism-related factors emerged as the primary drivers of improved CCD, highlighting the importance of regional economic health and the contribution of tourism activities to ecological development. These findings are consistent with research by Jin et al.⁵¹, which emphasized the pivotal role of economic stability and tourism in fostering coupled growth. In contrast, social factors—including population density and infrastructure development—exhibited a predominantly negative impact on CCD, often hindering coordinated development. These observations align with conclusions by Wang and Niu⁵². Notably, natural factors had both positive and negative effects, which varied across regions and over time, emphasizing the complex role that environmental conditions play in the coupled development of ecosystems and tourism. These mixed outcomes are consistent with the work of Peng et al.⁵³. The study also delved into the concept of tourism economic resilience, building upon the research by Huang et al.⁵⁴, which underscored the coordinated link between ESV and TER, demonstrating the interconnectedness of natural systems and economic sustainability.

The findings of this study have significant implications for policy and regional development planning. Promoting economic and tourism activities while ensuring ecological sustainability is essential for balanced growth across regions. Policies should encourage sustainable tourism practices that not only enhance local economic resilience but also protect ESV. For regions like the southeast, which already have a well-developed tourism sector, strategies should focus on mitigating the environmental impact of high tourist numbers. Conversely, for developing regions in western China, the focus should be on leveraging natural resources responsibly, with an emphasis on ecotourism to stimulate both economic growth and environmental conservation. Additionally, addressing the negative impacts of social factors such as high population density and infrastructure strain requires well-planned urban development strategies that reduce pressure on local ecosystems while improving residents’ quality of life. Balancing infrastructure growth with green development initiatives will be critical for enhancing CCD in densely populated areas.

This study emphasizes the need for future research to further understand the dynamic interactions between ESV and TER at a more granular scale. Detailed case studies focusing on specific regions or cities can help elucidate the local factors that drive or hinder coupled development. Moreover, the integration of new technologies, such as remote sensing and spatial analysis tools, could provide deeper insights into the spatial and temporal variability of CCD, informing more targeted and effective policy interventions.

Conclusions

Conclusions

To foster an ecological civilization, respecting natural environments and embracing sustainable development practices are pivotal for the advancement of tourism. This study investigated the relationship between ESV and TER at the municipal level in China from 2012 to 2022. The primary findings provide insights into how green development strategies can be implemented effectively to promote coordinated growth between tourism and ecosystems. The key conclusions drawn are as follows:

First, ESV in China exhibited an inverted “N” shaped trend during the study period. The ESV decreased from 11.114 trillion yuan in 2012 to 10.801 trillion yuan in 2022, marking an average annual decline of approximately 4.872 billion yuan. RSV and SSV were the most influential components, underscoring the need for improving the stability of ESV structures. Conversely, the TER composite index demonstrated a rising trend, growing from 0.248 in 2012 to 0.339 in 2022, suggesting enhanced resilience and positive developments in China’s tourism sector.

Second, the CCD between ESV and TER fluctuated over time, following an “N” shaped trend that mostly hovered at an elementary coordination level. During the decade from 2012 to 2022, the average CCD increased from 0.544 to 0.588, evolving through different phases—elementary coordination (2012–2014), intermediate coordination (2016–2018), and a reversion to elementary coordination (2020–2022). Spatially, the regional hierarchy for CCD was western > central > eastern, with a distribution pattern that showed “high values surrounding the center, high in the south and low in the north.” The emergence of a polar core suggests a synergistic and positive relationship between ESV and TER in regions with high coordination.

Third, a positive spatial clustering effect was observed in CCD across China, but the intensity of clustering declined gradually. The clustering analysis identified “low-low” and “high-high” clusters, indicative of distinct areas of coordinated or poorly coordinated development. Over time, disparities diminished, showing an evolution toward a more spatially balanced distribution.

Fourth, the spatial heterogeneity of influencing factors on CCD was significant, with distinct variations in impact across different regions. Economic factors exerted the greatest influence on CCD, followed by tourism, social, and natural factors. Economic drivers consistently showed a positive effect on CCD, although their influence gradually decreased from the southern to northern parts of China. Similarly, tourism factors played an essential role in enhancing CCD, with their impact shifting geographically towards the northeast. On the other hand, social factors had a predominantly negative effect, with high negative values centered along the Hu Huanyong line. The influence of natural factors, meanwhile, showed both positive and negative variations, indicating considerable spatial variability.

In conclusion, this study constructs an evaluation index system for the coupling and coordination of ESV and TER at the municipal level in China, based on the relevant data at the municipal level in China from 2012 to 2022, and explores the coupling and coordination relationship between the two and their spatial-temporal evolution characteristics by using the coupling and coordination degree model on the basis of the quantitative measurement of the two. At the same time, a geographically weighted regression model is used to identify the main factors affecting the coupled and coordinated development of ESV and TER, with a view to providing reference for the scientific and rational formulation of green and sustainable development paths for the tourism industry. The innovativeness of this paper is mainly reflected in the following three aspects: firstly, the use of long time data to quantitatively measure China’s inter-municipal tourism economic resilience based on the evolutionary resilience perspective; secondly, the summary of the spatial and temporal evolution law of China’s inter-municipal ESV and TER and the mechanism of influencing it; and thirdly, the introduction of the ESV and TER into the study of the coordinated development of the tourism economy and the ecological environment, which is more scientific, objectively measure the functional services provided by the ecological environment for human society, and promote the empirical research on the coordinated development of the two. This study offers theoretical and practical insights into the synergies and driving mechanisms between ESV and TER. However, there are some limitations: (1) Due to data constraints, the study only used data from 286 Chinese cities, which may not fully represent ESV and TER across all regions of China. Future research is needed to validate these findings. (2) The study focused solely on the municipal level in China, limiting its applicability. Multi-case studies across various scales are required for more robust conclusions. (3) The study primarily explored coordination effects and influencing mechanisms, but further research is needed to understand the underlying empirical mechanisms.

Recommendations

Building on the conclusions presented, several recommendations are proposed to enhance the coordinated development of ESV and TER:

• (1) For eastern regions with high ecosystem service value (e.g., Yangtze River Delta, Pearl River Delta), establish an early-warning mechanism for ecological carrying capacity and a tourism revenue feedback mechanism. Utilize policy instruments such as carbon sink trading and ecological tax rebates to guide high-density tourism zones toward transitioning to a “low ecological loss–high economic resilience” model. In ecologically fragile western regions (e.g., five northwestern provinces), prioritize integrated policies combining ecological restoration and tourism-based poverty alleviation. Leverage the national park concession system to develop alternative tourism formats like nature education and ecological research programs, thereby mitigating conflicts between ecological conservation and economic development.

• (2) Develop a dynamic assessment system for “ESV-TER CCD” (incorporating 20 core indicators such as ecological footprint and tourist overload index) through the Ministry of Culture and Tourism’s digital regulatory platform. Integrate coupling coordination metrics into local government performance evaluations. In typical polarized clusters like Beijing-Tianjin-Hebei and Chengdu-Chongqing, pilot a tourism ecological accounting system. Implement spatial carrying capacity calculations to achieve precise matching between tourist flows and ecological restoration efforts, thereby preventing ecosystem service degradation caused by intensive tourism development.

• (3) Address the “high periphery–low interior” spatial polarization of coupling coordination through three targeted regional interventions: Firstly, implement “tourism innovation enclave” policies in the Pearl River Delta and Yangtze River Delta to drive industrial upgrading in central-western regions via technology spillover effects. Secondly, establish a basin-scale ecological product value realization mechanism in the Yellow River Economic Belt, combining cross-provincial ecological compensation with tourism revenue-sharing agreements. Finally, explore hybrid development models integrating industrial heritage revitalization and ecotourism in northeastern old industrial bases, addressing social factor constraints on institutional innovation.