The Interaction Between Economic Development and Environmental Protection in Yunnan Province over the Past Two Decades

Abstract

Over the past two decades, urbanization and industrialization in Yunnan Province have rapidly advanced, spurring economic growth but also leading to intensified environmental challenges, particularly in urban areas where issues like air pollution and declining vegetation cover have become increasingly prominent. Balancing economic development and environmental protection has thus emerged as a core priority for Yunnan. This study quantifies and analyzes the temporal and spatial changes, coupling coordination degree, and driving factors underlying key indicators of economic development and environmental protection in Yunnan’s urban areas from 2000 to 2019. The results indicate that: (1) Economic growth is evident across all four types of regions, though growth rates have decelerated. The first type of region shows a “U”-shaped trend in FVC, the second type of region a “W” trend, and the third and fourth type of region show a consistent upward trend. (2) Overall coupling between economic development and environmental protection remains low across Yunnan, with a strong correlation between the degree of coupling coordination and urban economic development levels: the first type of region has the highest coupling coordination degree, while the fourth type of region has the lowest coupling coordination degree. (3) Industrial structure is the most influential factor in coupling coordination between economic development and environmental protection across Yunnan (74.9%). Influencing factors differ by region: secondary industry predominates in the first type of region (24.4%), foreign investment leads in the second (24.1%), technological innovation significantly impacts the third (23.9%), and marketization drives the fourth type of region (25.8%).

1. Introduction

With the rapid development of the global economy, issues related to resource overconsumption and environmental degradation have become increasingly pressing [1]. This is particularly true in developing countries, where swift industrialization and urbanization have brought economic prosperity but also accelerated ecological deterioration [2,3,4]. As the world’s largest developing country, China faces the dual pressure of sustaining economic growth while protecting the environment [5]. In recent years, the Chinese government has promoted green and sustainable development strategies to reconcile economic progress with ecological preservation [2,6,7]. In this context, Yunnan Province, as an important province in the southwestern region, has a unique strategic position and challenges. Located on the southwestern border of China, Yunnan is a core area of diverse biological resources, with high forest coverage, abundant water resources, and complex and sensitive ecosystems [8]. Although Yunnan Province is known for its diverse ecosystems and rich biodiversity [9], its fragile ecosystems are particularly vulnerable to damage from climate variability, agricultural expansion, and rapid urbanization [10]. In addition, as an open frontier facing Southeast and South Asia, Yunnan has become an important hub for cross-border trade and regional economic cooperation between China and these regions by virtue of its unique geographic location [11]. However, this geographical advantage has also brought about a remarkable contradiction between resource development and environmental protection. With economic development, the over-exploitation of resources and environmental burdens have intensified, and the carrying capacity of ecosystems has faced serious challenges. Therefore, environmental protection in Yunnan has far-reaching significance for ecological security and sustainable development of the whole country and the world. In Yunnan’s economic structure, agriculture, minerals, and tourism dominate, and the industrial structure is still mainly resource-dependent, with a relatively low degree of industrialization and a relatively lagging stage of economic development. Therefore, in the course of its development, Yunnan not only needs to enhance its economic competitiveness but must also avoid irreversible damage to the ecological environment, a dual challenge that constitutes the uniqueness of the contradiction between Yunnan’s economic development and environmental protection.

With the deepening of the “Western Development” policy, Yunnan has seen significant economic growth over the past two decades [12]. However, this rapid growth has also brought serious ecological challenges, including increased air pollution, declining forest cover, and escalating pressures on water resources [9,13]. These issues not only impact the region’s ecological security but also pose risks to residents’ health and quality of life [2,14,15,16]. Thus, achieving a balance between economic development and environmental protection has become a central challenge that Yunnan Province must address.

Exploring the relationship between economic development and environmental protection, scholars have proposed a variety of research methods, including the environmental Kuznets curve, the ecological footprint model, and the coupling coordination model. The environmental Kuznets curve assumes that there is an inverted U-shaped relationship between environmental pollution and economic development and that pollution will increase in the early stage as the economic level increases but will decrease when the economic development reaches a certain level. This theory provides a possible explanatory path between economic development and environmental protection and has been verified in empirical studies in many countries [2,9,17]. The ecological footprint model, on the other hand, assesses the impact of human activities on the environment by quantifying the resource consumption of an area against its ecological carrying capacity. It can reveal whether resource utilization in a region is at a sustainable level and is widely used in global environmental pressure assessment and national resource policy formulation [18,19]. However, these two approaches mainly focus on the assessment of a single parameter and fail to effectively address the systemic aspects of integrated economic ecosystems [20]. In addition, these methods are usually calculated in monetary terms only, failing to fully consider the value of local ecosystem services or existing natural capital, which may mislead policymakers [21]. More importantly, these methods ignore the multiple interactions between the natural environment and socio-economic systems, resulting in important economic and environmental relationships not being adequately captured. Unlike these approaches, the coupling coordination model is a more scientific approach as a dynamic and integrated analysis tool. The basic principle of the coupling coordination model originates from the coupling theory in physics, which quantifies the degree of coordination among systems by analyzing the interactions among multiple systems [22]. Compared to other models, the coupling coordination model not only assesses the current state of the system but also tracks the evolution of system coordination through time series data. Therefore, the model provides a theoretical basis for an in-depth understanding of the relationship between economic development and environmental protection and helps governments and policymakers formulate more refined policies [23,24,25]. The coupling coordination model has been widely applied in studying the relationship between the economy and the environment. Many scholars have used this model to explore the coupling and coordination relationships between economic, ecological, and environmental systems in different regions. For example, Ariken et al. analyzed the coupling coordination between urbanization and the ecological environment along China’s Silk Road Economic Belt [26]; Liu et al. studied the coupling between the spatial distribution coordination and dynamic evolution mechanisms in the Nansi Lake Basin of Shandong [27]; and Yao et al., using the entropy method and the coupling coordination model, investigated the coupling coordination status of urbanization and ecological environmental systems in 30 regions across China [28]. However, most existing studies focus on the macro-level relationships between economic development and environmental protection at the national, provincial, or regional levels. While these studies provide valuable insights into macroeconomic and environmental policies, they often overlook the economic structural differences between cities within a region and their unique impact on the environment. They also fail to adequately consider the diversity and disparities in economic development models among different cities. Therefore, this paper starts from the city region level, classifies each city according to its level of economic development, analyzes the differences in the degree of coordination between economic development and environmental protection and their coupling in various types of regions, and seeks to reveal the different impacts of different economic structures and development modes on environmental protection, with a view to providing theoretical support and empirical evidence for the formulation of more precise regional policies.

Moreover, it has been shown that factors such as industrial restructuring, technological innovation, and foreign investment have an important impact on the harmonization of the economy and the environment [2]. Industrial restructuring reflects the trend of economic transformation from resource-intensive to technology-intensive, which is an important path to realizing green development. Technological innovation can reduce the pressure on the environment while promoting economic growth by enhancing the efficiency of resource utilization and reducing pollution emissions [29]. Furthermore, foreign investment is often accompanied by technological spillover effects, which help to introduce advanced green technologies and management concepts, while the degree of marketization provides an endogenous impetus for coordinated development by optimizing resource allocation and enhancing economic efficiency. Optimization of the industrial structure and technological innovation are key factors in achieving coordinated development of the economy and the environment [29]. China’s rapid industrialization has led to the expansion of resource-intensive industries, which has brought about serious environmental problems, whereas the negative impact on the environment can be effectively reduced by promoting technological innovation and tertiary industry [29,30]. The economic growth of Yunnan Province relies on resource-intensive industries such as agriculture, mining, and tourism, and the rapid expansion of these industries has put greater pressure on the ecological environment [31]. However, the specific drivers affecting the degree of coupling coordination of economic development and environmental protection in various regions of Yunnan Province and the extent of their influence are not yet known and need to be further analyzed and quantified. Currently, the commonly used methods for analyzing drivers include traditional regression methods, support vector machines (SVM), neural networks, and random forest models. Compared with other methods, the random forest model can identify which factors have the greatest impact on the target variables by analyzing the importance of the variables so as to provide a decision-making basis for the formulation of corresponding policies or regulatory measures. Therefore, this paper chooses the random forest model for driver analysis.

The contribution of this paper is reflected in three aspects. Firstly, there is a lack of current research on the coupling and coordination of regional economic development and environmental protection in highland mountainous regions. Highland mountainous regions are usually characterized by unique geographical locations, significant differences in resource allocation, and unbalanced socio-economic development. Therefore, this paper chooses Yunnan Province as the study area, which fills the research gap in this field and provides theoretical support for the sustainable development of similar areas. Second, most existing studies have focused on the national, provincial, or overall regional level to explore the relationship between economic development and environmental protection. However, these studies usually ignore the differences in economic structures among different cities and their unique impacts on the environment. For this reason, this paper starts from the perspective of urban regions, classifies 16 cities in Yunnan Province according to their level of economic development, analyzes in depth the differences between different types of cities in terms of economic development and environmental protection, and then reveals the different impacts of various types of cities’ economic structures on environmental protection. Finally, the specific drivers of the coupling coordination degree of economic development and environmental protection in different regions of Yunnan Province and the degree of their influence have not been fully analyzed and quantified. Unlike previous studies, this paper adopts a random forest model to systematically analyze the specific impacts of industrial structure, technological innovation, degree of marketization, and foreign investment on the degree of economic and environmental coordination in Yunnan Province. Through this approach, this paper is able to more accurately assess the importance of each factor in the coordination of the economy and the environment and provides a new perspective and methodological support for quantifying the impact of these factors on the degree of coupling coordination.

In summary, The objectives of this paper are as follows: (1) to classify various cities in Yunnan Province according to their level of economic development; (2) to statistically and analytically characterize the spatio-temporal changes in the key indicators of economic development and environmental protection in different cities from 2000 to 2019; (3) to quantify and analyze the degree of coupling coordination between economic development and environmental protection and its spatio-temporal changes in the various types of districts of Yunnan Province in the last 20 years, by using the coupling coordination model; and (4) use the random forest model to quantify the degree of influence of the four key drivers on the coupling coordination of the economy and the environment in Yunnan Province. The purpose of this paper is to explore the coupled and coordinated relationship between economic development and environmental protection in Yunnan Province and its mechanism, and it seeks to reveal the intra-regional variability and its causes so as to provide a theoretical basis and reference for the green development of underdeveloped regions.

2. Experimental Data and Sources

2.1. Overview of the Study Area

The study area of this paper encompasses the urban regions of 16 prefectures in Yunnan Province. Situated on the southwestern border of China, Yunnan shares borders with Myanmar, Laos, and Vietnam, making it a crucial frontier and window for China’s southwestern opening [11]. Over the past two decades, urbanization and industrialization in Yunnan have accelerated, driving rapid economic growth. However, this economic expansion has been accompanied by increasingly significant environmental challenges, particularly in urban areas, where air pollution—especially rising PM2.5 concentrations—and declining vegetation cover are pressing issues. In response, Yunnan Province has gradually strengthened its environmental protection policies in an effort to balance economic development with ecological preservation [32]. Therefore, exploring the spatial and temporal changes of key indicators related to economic development and environmental protection in Yunnan’s urban areas, as well as the degree of their coupling and coordination and the underlying driving factors, is vital for achieving green development in the province.

2.2. Data Sources

In this paper, GDP, fractional vegetation cover (FVC), and PM2.5 concentration in urban areas are selected as the core indicators reflecting economic development and environmental protection. The GDP data is sourced from the dataset constructed by Chen et al. (2022), which utilizes calibrated nighttime light data to create a globally gridded GDP dataset at a 1 km resolution, covering detailed economic information from 1992 to 2019 [33]. This dataset has undergone rigorous calibration to more accurately represent the actual level of economic activity in each region.

The PM2.5 data were sourced from the Global High Air Pollutant (GHAP) dataset constructed by Wei et al. (2021), which has a spatial resolution of 1 km [34]. This dataset offers high-resolution PM2.5 data that covers air pollution across China from 2000 to 2019, effectively reflecting changes in environmental quality in the urban areas of Yunnan Province.

The FVC data were obtained by processing the MOD13A3 dataset with the Google Earth Engine (GEE) platform using the image element dichotomous model. The MOD13A3 dataset provides monthly 1 km-resolution NDVI data on a global scale. Since the growing season can better reflect the vegetation growth status, in this paper, the mean NDVI values were extracted for Yunnan Province from 2000 to 2019. In this paper, the mean NDVI values of the growing season (May to September) were extracted, then the maximum synthesis method was used to synthesize the annual NDVI data, and finally, the image element binary model was used to calculate the vegetation cover data for each year.

In order to accurately extract GDP, PM2.5, and FVC data within each state’s urban area, this paper also uses the Global Citywide Annual Dataset (1992–2020) provided by the National Glacial Tundra Desert Science Data Centre (http://www.ncdc.ac.cn/portal/ (accessed on 7 August 2024)). This dataset, with a spatial resolution of 30 m, covers the dynamics of urban expansion on a global scale and provides information on the boundaries of individual city limits. By combining this dataset, it is possible to better define urban areas and accurately extract economic and environmental data within cities, providing a basis for subsequent analyses of the spatial and temporal changes in economic development and environmental pollution indicators and their coupled coherence in individual regions.

Finally, an analysis of the driving factors is conducted, focusing on four key aspects: industrial structure, technological innovation, degree of marketization, and foreign investment, with specific indicators outlined in

Table 1. Indicators affecting the degree of coupling coordination of economic development and environmental protection in cities in Yunnan Province.

Classification	Explanatory Variable	Formulae	Source
Industrial structure	Ratio of primary industry value-added	Primary industry value-added/GDP	China County Statistical Yearbook
	Ratio of secondary industry value-added	Secondary industry value-added/GDP	China County Statistical Yearbook
	Ratio of tertiary industry value-added	Tertiary industry value-added/GDP	China County Statistical Yearbook
Technological innovations	Ratio of R/D input	R/D input/GDP	China Urban Statistical Yearbook, China County Statistical Yearbook
Degree of marketization	Ratio of non-state-owned fixed assets	Non-state-owned fixed assets/Social fixed assets	China Urban Statistical Yearbook, Yunnan Provincial Statistical Yearbook
Foreign investment	Ratio of foreign direct investment	Foreign direct investment/Social fixed assets	China Urban Statistical Yearbook, China County Statistical Yearbook

below. These statistics were obtained from the local statistical yearbooks of each city and state, the Yunnan Provincial Statistical Yearbook (2001–2020), the China County Statistical Yearbook, and the China Urban Statistical Yearbook (2001–2020), and some of the missing data were filled in by interpolation.

3. Research Methods

3.1. Selection of Economic Development and Environmental Protection Indicators and Their Drivers

3.1.1. Selection of Economic Development Indicators

Economic development is one of the core topics of this paper, and GDP is chosen as the main indicator of economic level. GDP can comprehensively reflect the economic scale and production activity level of the region and is an important basis for analyzing the interrelationship between the economy and the environment. High GDP is usually accompanied by higher resource consumption and pollution emissions; therefore, through the trend of GDP, we can initially understand whether there is a conflict or synergy between economic growth and environmental protection [4]. In addition, it has been pointed out that changes in the GDP growth rate have a direct impact on the quality of the environment; especially in the process of industrialization, economic growth is often accompanied by increased environmental pollution [35].

3.1.2. Selection of Environmental Protection Indicators

In order to comprehensively assess the status of environmental protection, this paper chooses vegetation cover and PM2.5 concentration as the core indicators of environmental protection. Vegetation cover is an important indicator of the ecological health of a region, and a higher vegetation cover usually means a better ecological environment, stronger ecological restoration capacity, and less land degradation [36]. The FVC data can reflect the vegetation status and ecological restoration effect of the region. Meanwhile, PM2.5 concentration, as a major indicator of air pollution, directly affects air quality and public health. Higher PM2.5 concentration often indicates that the environmental protection measures in the region may be insufficient and the air pollution problem is more serious [37]. By analyzing the time series changes of these two indicators, the relationship between the state of regional environmental protection and the environmental burden in the process of economic development can be revealed.

3.1.3. Selection of Drivers

In addition to the core indicators of economic development and environmental protection, the driving factors affecting the degree of coupling and coordination between the two are also crucial. In this paper, four factors are chosen as drivers: industrial structure, technological innovation, degree of marketization, and foreign investment. In addition to these four factors, there are other potential influencing factors, especially environmental governance capacity and government policy implementation. However, due to the difficulty of collecting data on these potential factors in each state region, they were not included in this study. Industrial structure determines the nature of economic activities and their impact on resources and the environment. A high proportion of secondary industries, especially heavy industries, tends to exert greater pressure on the environment, while an increase in the proportion of tertiary industries may help to alleviate environmental pressure and promote green development [38]. Technological innovation, especially the introduction of green technology, can effectively improve the efficiency of resource utilization and reduce pollution emissions, thereby promoting the harmonious development of the economy and the environment [29]. Regions with a high degree of marketization usually optimize the allocation of resources and improve economic efficiency through market mechanisms while also being able to promote the implementation of environmental protection measures through competitive mechanisms [29]. Foreign investment, especially green investment, not only introduces advanced technology and management experience but also enhances the local level of environmental protection technology and helps to improve environmental quality [29]. Therefore, these four drivers may have an important impact on the degree of coupling coordination between economic development and environmental protection.

3.2. Calculation of Vegetation Cover in Yunnan Province

To accurately calculate the vegetation cover in Yunnan Province, this paper employs the pixel binary model (PBF) to derive the vegetation cover data. This is a classical remote sensing method for estimating vegetation cover [39]. The fundamental principle of the model involves using the normalized difference vegetation index (NDVI) as a distinguishing indicator between vegetation and non-vegetation [40]. The vegetation cover for each image was calculated by establishing the NDVI thresholds for vegetation and bare soil:

$$FVC={\displaystyle \frac{NDVI-NDV{I}_{soil}}{NDV{I}_{veg}-NDV{I}_{soil}}}$$

(1)

FVC represents the vegetation coverage of the image elements, indicating the proportion of vegetation within the entire image, with values ranging from [0, 1]. NDVI refers to the normalized difference vegetation index of the image elements. NDVI_soil denotes the NDVI value for pure bare soil or unvegetated areas, with a confidence interval set at 5%. NDVI_veg represents the NDVI value for purely vegetated areas, with a confidence interval established at 95%.

3.3. Urban Classification Methodology

In order to deeply explore the differences in economic structure among different cities in Yunnan Province and their unique impact on the environment, this paper classifies 16 prefecture-level cities in Yunnan Province according to their total cumulative GDP from 2000 to 2019 and analyzes the mechanism of their economic activities on the ecological environment. Total cumulative GDP is an important indicator for measuring the level of regional economic development and the scale of industrial activities. Using the cumulative total GDP of the last 20 years for classification can effectively smooth out the short-term fluctuations of the data in a single year, thus providing a more comprehensive assessment of the level of economic development of each city and its long-term impact on the environment. The UN human development index (HDI) classifies regions into four categories: very high human development, high human development, medium human development, and low human development [41]. Thus, based on the cumulative GDP of each city over the 20-year period from 2000 to 2019, all 16 cities and municipalities are divided into four regions by total GDP.

The first type of region (region I) includes economically developed cities, a portion of the cities with the highest cumulative GDP, which are usually the core area of economic development in Yunnan Province, with highly developed industries and services and the greatest contribution to the province’s GDP. The second type of region (region II) involves more developed cities, cities with the next highest total GDP, whose level of economic development is second only to the first category. Cities in this category are usually regional economic centers with relatively balanced industrial structures and stable economic and environmental performance. The third type of region (region III) contains moderately developed cities, cities with a medium level of cumulative GDP. These cities have a slower process of industrialization and a smaller economy but have grown in recent years with the tilting of national regional policies. The fourth type of region (region IV) groups the less developed cities, the cities with the lowest total GDP, which are mostly remote areas with more lagging economic development, a relatively homogeneous industrial structure, and better ecological and environmental conditions. With less economic activity, environmental pressure is relatively low.

3.4. Coupling Coordination Analysis Model

In socio-economic and ecological research, the coupling coordination model is often used to evaluate the relationship between multiple systems, such as economic development and environmental protection or resource utilization and economic growth [42]. Therefore, this paper uses the coupling coordination degree model to evaluate the coupling coordination relationship between economic development and environmental protection. The data need to be normalized before calculating the degree of coupling coordination, and the formula is as follows:

$$R_{ij}={\displaystyle \frac{X_{ij}-X_{\min }}{X_{\max }-X_{\min }}}$$

(2)

where R_ij is the standardized value of the jth evaluation indicator, X_ij denotes the initial value of the jth indicator of the ith city, X_max denotes the maximum value, and X_min denotes the minimum value.

Next, the coupling coordination degree is calculated using the following formula:

$$C=\sqrt{{\displaystyle \frac{U_1\times U_2\times U_3}{{\left[\frac{U_1\times U_2\times U_3}{3}\right]}^3}}}$$

(3)

$$T=a{U}_1+b{U}_2+c{U}_3$$

(4)

$$D=\sqrt{C\times T}$$

(5)

Equations (3)–(5) include the following: C represents the coupling degree, T is the integrated coordination value, and D is the coupling coordination degree. U₁, U₂, and U₃ correspond to the standardized values of GDP, FVC, and PM2.5, respectively, and a, b, and c represent the weights of system importance, respectively, and satisfy the condition a + b + c = 1. Referring to the analyses of the interrelationship between the economy and the environment in other studies, it is usually considered that the economy and the environment are of equal importance. Based on this, this paper considers that GDP, FVC, and PM2.5 are of equal importance in the coupling coordination degree model, and therefore, the weights of a, b, and c are all set to 1/3 [43,44]. The coupling coordination degree D ranges from [0, 1]. A higher D value indicates a greater degree of coupling coordination, reflecting more harmonized development between the two systems. Conversely, a lower D value indicates less coordination. Finally, the coupling coordination degree is divided into 10 levels, as shown in

Table 2. Criteria for classifying the coupling coordination degree.

Interval of D-Values for Coupling Coordination	Level of Coordination	Degree of Coupling Coordination
[0.0~0.1)	1	Extreme disorder
[0.1~0.2)	2	Severe disorder
[0.2~0.3)	3	Moderate disorder
[0.3~0.4)	4	Mild disorder
[0.4~0.5)	5	Borderline disorder
[0.5~0.6)	6	Barely Coordinated
[0.6~0.7)	7	Primary Coordination
[0.7~0.8)	8	Intermediate coordination
[0.8~0.9)	9	Good coordination
[0.9~1.0]	10	Quality coordination

.

3.5. Driver Analysis

In this paper, the random forest model is chosen as the method for driver analysis based on its several advantages. Firstly, random forest is an integrated learning algorithm that improves prediction accuracy and robustness through the integration of multiple decision trees [45]. Compared with traditional regression models, random forest models do not need to assume a linear relationship between variables [46] and thus can effectively deal with complex nonlinear relationships in economic and environmental problems. In addition, the random forest model has a significant advantage in assessing the importance of variables. It is able to intuitively identify the factors that have the greatest impact on the target variables by calculating the contribution of each variable to the predictive performance of the model [46], which provides researchers with a reliable policy basis.

Secondly, the random forest model performs excellently in handling missing values and outliers. Economic and environmental data often contain missing values or noise, and the random forest model improves the model’s tolerance to data noise and prediction performance through its ensemble learning mechanism and random sampling strategy, thereby enhancing the model’s stability and accuracy [47].

Compared to other commonly used methods, the random forests model outperforms traditional regression methods, SVM, neural networks, and geographic detector in several ways. Traditional linear regression, although simple and easy to use, has difficulties in handling nonlinear relationships and is sensitive to multicollinearity [48]. Support vector machines perform better in high-dimensional spaces but perform slightly worse in regression tasks and have poor model interpretation, making it difficult to quantify the importance of variables [49]. Neural networks are good at modeling nonlinear relationships but are complex and computationally expensive to train, and their ‘black box’ nature makes it difficult to interpret the importance of variables [50]. Geographic detector is capable of identifying spatial distribution differences of driving factors and assessing the impact of these factors on target variables by quantifying spatial correlations. However, geographic detector struggles with complex nonlinear relationships and cannot effectively quantify the relative importance of factors, making it difficult to capture more intricate variable interactions and dynamics. In contrast, random forests can handle nonlinear relationships and clearly identify which factors have the greatest impact on the model’s output through the evaluation of variable importance. This provides a more accurate quantitative analysis for a deeper understanding of the mechanisms behind these factors. Furthermore, random forests can process numerous variable interactions, offering a more comprehensive and detailed analysis. Therefore, this paper adopts a random forest model to quantify and analyze the drivers of coupled changes.

4. Results

4.1. Urban Classification Results

Through statistical analysis, it is found that the main indicators of economic development and environmental protection of each city show a certain pattern in the change of time series. In order to better analyze and show the pattern, as well as to explore the unique impact of economic structural differences in different types of cities on the environment, we tried to classify the 16 cities and municipalities. Through the attempt, it was found that there is an important correlation with the degree of economic development, so this paper divides the 16 cities and municipalities into four different categories based on the size of their cumulative GDP during the period of 2000–2019. The classification results are shown in Figure 1.

Region I includes the four cities and towns of Kunming, Qujing, Yuxi, and Honghe. These localities have the highest total GDP, relatively rapid economic development, and occupy an important position in the economic structure within Yunnan Province. Kunming, as the capital city of the province, has a significant industrial agglomeration effect, while Qujing, Yuxi, and Honghe drive economic growth within the region by virtue of their industrial, mineral, and agricultural resources. Region II includes Dali, Chuxiong, Dehong, and Lijiang. These cities and municipalities are in the second tier of the cumulative GDP rankings and have relatively stable economic development, relying mainly on tourism, agriculture, and primary processing and manufacturing. As famous tourist cities, Dali and Lijiang have seen remarkable development in their tourism economy, which has led to the prosperity of service and related industries in the region. Region III includes Baoshan, Pu’er, Wenshan, and Xishuangbanna. The economic development of these localities mainly relies on agriculture, forestry, and their related processing industries. Although their total GDP is not as high as that of the first two categories, they have also made great progress in green economy and sustainable development in recent years, especially in the ecological protection zones such as Pu’er and Xishuangbanna. Region IV includes Zhaotong, Lincang, and Diqing. These localities have a weaker economic base and a relatively low cumulative GDP. The geographical remoteness of these areas, the relative backwardness of their infrastructure, and the fragile ecological environment have made their economic development less rapid than that of other cities and towns, but in recent years, they have also been gradually upgrading their economies through the development of specialty agriculture and tourism.

Due to the small city scope of the Nujiang region during the period of 2000 to 2003, the data is incomplete, which leads to its inability to be included in the analysis of city classification in this study. Therefore, the main research object of this experiment is the 15 cities and towns in Yunnan Province. By dividing Yunnan Province into four regions with different levels of economic development, it is possible to better compare and analyze the characteristics of the time series changes in the GDP growth rate, FVC, and PM2.5 concentration, as well as the differences in the degree of coupling coordination of the different types of cities during the 20-year period.

4.2. Temporal Variation Characteristics of Economic Development and Environmental Protection Indicators

The temporal variation characteristics of multiple indicators, i.e., GDP, FVC, and PM2.5 concentration, for each city in Yunnan Province from 2000 to 2019 are shown in Figure 2 below. In region I, the GDP growth rate of several cities other than Honghe showed an overall fluctuating downward trend during 2000–2019. Kunming’s GDP growth rate has less fluctuation and a smoother trend, declining at a rate of 0.3259 per year. Yuxi had the fastest declining rate of GDP growth at 0.8752/year. It is worth noting that although each city’s GDP growth rate is generally on a downward trend, its economy is still growing, just at a slower rate. In terms of vegetation cover, the four cities in the first category show a “U” shaped trend, decreasing and then increasing, during the period of 2000–2019. Yuxi has a higher FVC level, with less fluctuation, and Kunming has the lowest FVC level, with less fluctuation. In terms of PM2.5 concentration, the four cities in the class I area show an inverted U-shaped trend, increasing and then decreasing, especially after 2013. Red River had the highest PM2.5 concentration. The PM2.5 concentrations in Kunming and Qujing are close to each other, with similar fluctuation trends. Yuxi has the lowest PM2.5 concentration.

In region II, the GDP growth rate of each city showed an overall decreasing trend, with the fastest decreasing rate in Chuxiong (1.0248/year) and the slowest decreasing rate in Dali (0.2211/year). The FVC of the second type of area generally shows a “W” trend, decreasing, then increasing, then decreasing, and then increasing, with the inflection points in 2007, 2010, and 2013, respectively. From 2000 to 2007, the vegetation cover showed a certain degree of decrease, during which Dehong’s FVC was the highest, and Dali’s was the lowest. From 2000 to 2010, there was a decrease in FVC, it rebounded from 2007 to 2010, then declined from 2010 to 2013, and then showed an increasing trend after 2013, with Chuxiong having the highest FVC and Dehong having the lowest FVC. The PM2.5 concentrations in the second category of areas also showed a U-shaped trend, increasing and then decreasing from 2000 to 2019, especially after 2013, when PM2.5 concentrations generally decreased. The PM2.5 concentration in Dehong is relatively high and fluctuates greatly, while the PM2.5 concentration in Lijiang is the lowest and maintains a relatively stable downward trend.

The GDP growth rates of the four cities in region III also show a decreasing trend, with Baoshan showing the slowest rate of decline (0.05/year) and Pu’er showing the fastest rate of decline (0.9859/year). Except for Baoshan, where the FVC showed a decreasing trend, several other areas showed a linear increasing trend, growing at a rate of 0.0032 per year. The PM2.5 concentrations in the four cities also showed an inverted “U” shaped trend of increasing and then decreasing between 2000 and 2019, especially after 2013, when PM2.5 concentrations in all cities generally decreased, indicating a significant improvement in air quality.

In region IV, Zhaotong’s GDP growth rate showed an increasing trend (0.5237/year), Diqing and Lincang’s GDP growth rate showed a declining trend, and Lincang had the fastest decreasing rate (0.2132/year). Zhaotong’s vegetation cover is relatively low, fluctuating between 0.2 and 0.3 overall, and with a large change over the 20-year period, but with a slight upward trend. The vegetation cover in Lincang is relatively high and stable, with less fluctuation, and overall maintained between 0.3 and 0.35, showing better vegetation cover. Vegetation cover in Diqing is lower and decreased significantly around 2004, and then fluctuated at a low level, showing the instability of vegetation cover. The PM2.5 concentrations in the three cities showed an inverted “U” shaped relationship, increasing and then decreasing over the 20-year period, with Diqing having the lowest PM2.5 concentration and Zhaotong having the highest PM2.5 concentration.

As a whole, the temporal variation of economic development and environmental protection indicators for the four types of areas are shown in Figure 3 below. The GDP growth rates of the four types of areas show a fluctuating downward trend, with the slowest decline in GDP growth rate found in region I (0.2746/year) and the fastest decline in GDP growth rate found in region IV (0.7775/year). FVC shows a U-shaped trend, decreasing and then increasing, with the lowest FVC in region I, the highest FVC in region II before 2010, and the highest FVC in region III after 2010. PM2.5 shows an inverted U-shaped trend, with the lowest PM2.5 in region II and the highest PM2.5 in region I.

4.3. Characterization of Temporal and Spatial Variations in Coupling Coordination

The coupling coordination degrees of GDP, FVC, and PM2.5 concentration in urban areas of each state and city in Yunnan Province calculated using the coupling coordination model (Figure 4) show significant differences in time and space during 2000–2019.

As a whole, the coupling coordination of economic development and environmental protection in Yunnan Province is at a low level, and there is a strong correlation between the degree of coupling coordination and the degree of urban economic development, with the highest degree of coupling coordination found in region I and the lowest degree of coupling coordination found in region IV (Figure 4a,c). The coupling coordination degree of region I is between 0.54 and 0.66, which generally achieves the “barely coordinated” and “primary coordination” stages, while the region II and region III categories are between 0.44 and 0.55, which are generally on the verge of “borderline disorder” and “barely coordinated”. The coupling coordination degree of region IV is only between 0.31 and 0.43, which is in the stage of “mild disorder” and “borderline disorder” (Figure 4c).

From a temporal perspective, there is a slight downward trend in the coupling coordination degree in Yunnan Province from 2000 to 2019. This phenomenon is likely the result of multiple factors working together, including unreasonable industrial structure, insufficient policy implementation, and limited environmental governance capacity [51,52]. Firstly, Yunnan’s industrial structure is still dominated by resource-dependent industries, especially agriculture, mining, and traditional manufacturing, which, while contributing to economic growth, often place significant pressure on the environment, leading to poor coordination between the economy and the environment. Secondly, although Yunnan has strengthened the formulation of environmental protection policies in recent years, the implementation of these policies has been insufficient, with delays in enforcing environmental protection measures in some areas, preventing the timely realization of environmental protection outcomes. Meanwhile, Yunnan still lags behind in technological innovation, with a low penetration rate of green and environmental technologies, which, to some extent, limits the potential for coordinated economic and environmental development. Finally, Yunnan’s environmental governance capacity is relatively limited, especially in high-resource-consuming industries and sectors, where an effective environmental management and regulatory system has not yet been formed, resulting in slow progress in pollution control and hindering improvements in coupling coordination. To achieve coordinated economic and environmental development, Yunnan needs to increase efforts in several areas to promote green and low-carbon development. Firstly, industrial structure adjustment must be accelerated, with a shift from resource-intensive industries to technology-intensive and knowledge-intensive industries, particularly by strengthening the development of the tertiary sector and green industries. Secondly, Yunnan should increase investment in technological innovation, especially in environmental protection technologies, clean energy, and green technologies, to promote the research and application of low-carbon technologies and enhance the province’s environmental governance capacity. Thirdly, the implementation of policies must be strengthened to ensure that environmental protection policies are effectively enforced in all sectors and regions through enhanced supervision and raising environmental standards, thereby promoting the participation of society as a whole in environmental protection. Additionally, Yunnan should accelerate market-oriented reforms, optimize resource allocation, and use market mechanisms to promote a win−win situation for green development and environmental protection. In this process, the implementation of energy quota trading policies is a key tool to promote green development [53]. The government should implement differentiated regulatory policies based on the characteristics of different industries: for heavy industries and high-pollution sectors, strengthen energy quota control and support green technology research and development; for light industries and clean industries, increase investment in R&D and improve production efficiency [53]. Through these measures, Yunnan can effectively promote industrial upgrading and green transformation, significantly improve the coupling coordination between the economy and the environment, and achieve sustainable development. Except for the third category of areas, where the coupling coordination degree shows an upward trend (0.0012), several other categories of areas show a downward trend, with the fastest rate of decline found in the first category of areas (−0.0044) and the slowest rate of decline found in the fourth category of areas (−0.0008) (Figure 4c).

From the spatial dimension, the spatial differentiation of the coupling coordination degree of economic development and environmental protection in Yunnan Province is obvious in different regions (Figure 4b). The coupling coordination degree is higher in economically developed regions in central and southwestern Yunnan, such as Kunming, Qujing, and Yuxi, where the mean value of the coupling coordination degree is higher due to their higher level of economic development and relatively better environmental governance measures, reflecting better coordination between economic development and ecological environmental protection. On the other hand, the coupling coordination degree is lower in northwestern Yunnan, northeastern Yunnan, and southeastern Yunnan, in localities such as Diqing, Lijiang, Zhaotong, and Wenshan, where the coupling coordination degree is relatively low, the economic development is lagging behind the other areas, and the coordination ability of ecological environmental protection is weaker. The remoteness of these regions and their weak economic foundation, coupled with their fragile environment, have led to a low degree of coupling, making it difficult to realize the double enhancement of the economy and the environment.

4.4. Contributive Analysis of Impact Factors

The contribution of the influencing factors in all types of cities and the province is analyzed through the random forest model, as shown in Figure 5. The results show that industrial structure is the most important factor affecting the coupling coordination degree of various types of cities, accounting for more than 70%, but there are still significant differences in the factors affecting the coupling coordination in various types of cities (Figure 5a). For region I, the ratio of secondary industry is the most important factor affecting the coupling coordination, with a contribution rate of 24.4%. The largest contribution to the coupling coordination of region II is foreign investment (24.1%). Technological innovation contributes the most to the coupling coordination of region III, with a contribution rate of 23.9%. Finally, the main factor affecting the coupling coordination of region IV is the degree of marketization, with a contribution rate of 25.8%.

From a province-wide perspective, industrial structure is the factor that contributes the most to the change in the coupling coordination degree of economic development and environmental protection in Yunnan Province, with a contribution rate of 74.9%, of which the ratio of the tertiary industry contributes 35.8% to the coupling coordination degree (Figure 5b), which suggests that the adjustment of the economic structure has a key role in promoting the process of green development. As Yunnan Province gradually transforms from an economic model dominated by the secondary industry to the tertiary industry, the optimization of the industrial structure helps to alleviate the negative impact of economic activities on the environment. Second is scientific and technological innovation, with a contribution rate of 9.6%, indicating that technological innovation promotes the application of environmental protection technologies and the enhancement of resource utilization and plays a positive role in promoting the coordinated development of the economy and the environment. In contrast, the degree of marketization and foreign investment have relatively small impacts, accounting for 8.1% and 7.2%, respectively. Although market allocation improves economic efficiency and foreign investment brings economic growth and employment opportunities, their overall contribution to the degree of coupling coordination is not as significant as that of industrial restructuring.

5. Discussion

This study examines the interaction between economic development and environmental protection, as well as the spatio-temporal changes in their coupling coordination degree, based on GDP, FVC, and PM2.5 concentration data from Yunnan Province from 2000 to 2019. By analyzing the coupling coordination degree and its influencing factors across different regions, this research reveals the coordination mechanisms between economic growth and ecological environmental protection in Yunnan Province, along with the regional disparities. The following is a further discussion of the main research findings.

5.1. Analysis of Changes in Vegetation Cover and PM2.5 Concentration

The four types of areas reached the lowest vegetation cover around 2010, especially in region I, which may be related to climate change, extreme weather events, and lagging policy and environmental protection measures. It has been shown that extreme weather events such as droughts, high temperatures, and floods can significantly affect vegetation cover and growth [54]. Yunnan region experienced severe droughts during 2009–2010, which negatively affected vegetation growth and led to a significant decrease in vegetation cover. In addition, prior to 2010, environmental protection policies in Yunnan Province and other parts of China were relatively lagging behind, and environmental governance was weak, which was also a major cause of the decline in vegetation cover. Although China began to gradually strengthen its environmental protection policies in the early 2000s, the implementation and effects of these policies were not fully realized before 2010. With rapid economic development, the protection of the vegetation cover failed to follow in time, leading to over-exploitation of natural resources and environmental degradation.

Another interesting phenomenon is that PM2.5 concentrations in various cities peaked in 2013 and gradually began to decrease from 2013 onwards, and the reason for this phenomenon may be related to policy and energy restructuring. The year 2013 was an important turning point in China’s environmental governance, and in order to deal with the increasingly serious air pollution problem, especially the rapid increase in PM2.5 concentrations, the government introduced the Air Pollution Prevention and Control Action Plan in that year. Through strict environmental regulations and policy measures, the Chinese government has effectively controlled PM2.5 pollution in a short period of time and significantly reduced the emission of air pollutants [55]. In addition, the improvement of energy structure and the use of clean energy have significant effects on reducing PM2.5 concentrations, and after 2013, China accelerated the adjustment of its energy structure, reducing the share of coal in energy and shifting to the use of clean energy [56].

5.2. Mechanisms for Coordinating Economic Development and Environmental Protection in the Four Types of Regions

In region I, early economic growth relied on a high-pollution and high-resource-consumption model, resulting in a low coupling between economic development and environmental protection. GDP growth was relatively fast in 2000–2010, but it relied mainly on traditional heavy and resource-based industries, leading to large-scale energy consumption and environmental pollution. Especially in resource extraction industries such as mining and coal, the emissions of waste gas and wastewater have caused a serious pressure on the environment [57]. With the increase in industrial emissions, PM2.5 concentrations have increased significantly, while the vegetation cover (FVC) has been damaged. The contradiction between economic growth and environmental degradation intensified, the coupling coordination was low, and economic development and environmental protection failed to form a good coordination. Around 2010, the state’s attention to environmental protection gradually increased, and green transformation became the core of the policy [58]. The government increased its support for clean energy and green industries, which promoted industrial restructuring and reduced pollution emissions. With the implementation of environmental protection policies, FVC has rebounded, and PM2.5 concentration has gradually decreased, indicating a gradual improvement in the environment. Despite the progress made in green transformation, the recovery process of coupling coherence is still slow due to the slowdown in economic growth, increased resource consumption, and insufficient local environmental protection efforts.

In region II, the GDP growth rate generally shows a decreasing trend, the FVC shows a more volatile ‘W’-shaped change, and the PM2.5 shows an inverted U-shape change, which reflects the complex relationship between the economy and the environment. From 2000 to 2007, the economic growth of the second category of areas was relatively rapid, and the rapid development of tourism and infrastructure construction contributed to the increase in GDP [59]. However, due to the acceleration of urbanization, the expansion of tourism and the rapid development of transport, accommodation and other services, environmental pollution problems gradually came to the fore, with a rise in PM2.5 concentrations and a decline in FVC. In this stage, despite the remarkable economic growth, measures in environmental management and sustainable development were still insufficient, and the ecological environment faced some deterioration. From 2007 to 2013, the FVC in the second category of areas showed a fluctuating trend of change with a rise and then a fall, and PM2.5 still fluctuates and rises, reflecting the impact of short-term adjustments in policies on the environment. During this period, some areas may have carried out some environmental protection measures or had localized ecological restoration projects, but their effectiveness has not yet been durable and stable, and FVC shows fluctuating changes. Despite some policy regulations, the environmental pressure has not been fundamentally relieved, and PM2.5 concentration has fluctuated and increased. From 2013 onwards, with the promotion of green economic transformation, the government strengthened the control of environmental protection and introduced a series of policies to promote green development [55,56]. At this time, PM2.5 concentration gradually decreased and FVC rebounded. However, the slowdown in economic growth and excessive resource consumption in some areas resulted in the coupling coordination remaining at a low level and a slow recovery.

In region III, the GDP growth rate shows a decreasing trend, but the FVC rises linearly, and the PM2.5 concentration shows an inverted U-shape change, with a gradual increase in the degree of coupling coordination. The economy of the third category of areas mainly relies on low-carbon and low-pollution industries, such as agriculture, forestry, and eco-tourism. Before 2005, the economic development of these regions relied on traditional agriculture, resource-based industries, and the export of some primary products [60], and economic growth was relatively rapid. Although the economic growth of these regions relied on traditional agriculture and ecological industries, there was a certain amount of mineral extraction, burning of agricultural waste, etc., which led to an increase in the concentration of PM2.5. However, the ecological protection policies in the third category were more stringent and effective relative to other regions, such as returning farmland to forests and the delineation of nature reserves [61], and therefore, the FVC showed a rising trend. Before 2013 it was mainly the rise of FVC that promoted the rise of coupling coordination. From 2013 onwards, with the national emphasis on the green economy, the economy of the third category of areas gradually transformed into green and low-carbon [55]. Eco-tourism, green agriculture, and eco-protection industries gradually became new economic growth points, the government strengthened the implementation of environmental protection policies, the PM2.5 concentration began to gradually decline, and FVC continued to rise in this period, and the two together promoted the coordinated development of the economy and the environment.

Region IV relies mainly on traditional resource-based industries, including agriculture and mineral resource extraction. Before 2013, the rapid development of resource-consuming industries promoted economic growth [62], but it also brought environmental problems, which, together with insufficient environmental protection measures, led to the increase of PM2.5 concentration. FVC showed a slight fluctuating upward trend, but it was not enough to change the fact that the coupling coherence was on a downward trend. After 2013, with the strengthening of environmental protection policies, the concentration of PM2.5 gradually declined, and FVC continued to rise slowly. Through market-oriented reforms, these regions have used market mechanisms to allocate resources more effectively, optimize part of the industrial structure, and improve economic efficiency [63]. However, these regions still have the problem of a slower economic structural transformation, as GDP growth continues to slow down, and the recovery process of coupling coordination is still slow.

5.3. Analysis of Differences in Coupling Coordination Among the Four Types of Regions

The results of this study show that the coupling coordination degree of all types of regions except region III is in a declining trend, and the decline rate of region I is higher than that of the other types of regions. The reasons behind this phenomenon may be related to the economic development mode, resource and environmental pressure, policy implementation, technological and industrial structural adjustment, and other factors. Firstly, the reason why the coupling harmonization degree of region I declines faster is that its rapid economic growth has brought about greater environmental pressure. Pollutant emissions and resource consumption increase while economic growth tends to exert great pressure on the environment, especially before comprehensive industrial transformation and application of environmental protection technology are realized [9]. Region I has rapid economic development and accelerated industrialization and urbanization; its cities have strong economic strength but also face greater environmental pressure, with the expansion of the economic scale, industrial production, energy consumption, transportation, and other areas of emissions that increased significantly exacerbating atmospheric pollution, water pollution, and ecological environment deterioration. In this situation, the environmental carrying capacity has reached its limit, making the coordination between economic development and environmental protection decline. Furthermore, the lag in policy implementation and insufficient enforcement are other important reasons for the decline in coupling coordination in region I. Although environmental protection policies in this region are more comprehensively formulated, they face obstacles in specific implementation in terms of fund allocation, enforcement supervision, and local governments’ priority orientation towards economic development. For example, in order to achieve short-term economic growth targets, environmental enforcement may be lax or deflated in some localities, thus exacerbating environmental problems [64]. The industrial structure of the first group of areas favoring industrialization is also a key factor. Although these industries can bring high economic benefits, they also generate a large amount of pollution and resource consumption, and coupled with the lag in industrial transformation and structural optimization, the environmental pressure in these cities has failed to be effectively mitigated, leading to a decline in the degree of coupling coordination.

The decrease in the coupling coordination degree of the other two regions may be related to the intensification of the contradiction between economic growth and environmental carrying capacity and the poor implementation of policies. Although the economic development of these cities is relatively slow, their environmental protection technology, infrastructure construction, and environmental governance capacity are also relatively limited, leading to the intensification of the contradiction between economic development and environmental protection and a downward trend in the degree of coupling coordination. In addition, these two types of areas have a relatively weak economic base and limited policy implementation capacity, and the implementation of environmental protection policies is not as strong and effective as in the first type of areas. Insufficient resources in environmental protection infrastructure, pollution monitoring, and technical inputs also lead to poor environmental governance, and the contradiction between economic growth and environmental protection is difficult to resolve. In addition, these cities may also suffer from insufficient coordination among local governments in policy implementation, especially in cross-regional pollution prevention, leading to unsatisfactory governance [65].

The possible reasons for the increase in the coupling coordination of region III are mainly the upgrading of the industrial structure and the promotion of the green development model, as well as the support of ecological protection policies. The economic base of region III is relatively weak, but in recent years, the economic growth of these cities has relied more on the green economy and sustainable development model. Agriculture, forestry, and industries related to ecological protection are the economic pillars of these cities. These industries have a relatively low impact on the environment, and as awareness of ecological protection increases, the growth of the green economy has led to an increase in the degree of coupling coordination. Region III localities are often listed as national or provincial ecological protection areas (e.g., Pu’er, Xishuangbanna, etc.), which provides stronger policy support for their environmental protection. These cities have more resources and policy support for ecological environmental protection, especially in forest protection and biodiversity conservation, and the government has taken active measures. For example, through the promotion of ecological compensation mechanisms and the upgrading of ecological and environmental assessment standards [66], these regions have been able to effectively control ecological damage and enhance the coordination between the economy and the environment. The implementation of these policies has effectively controlled ecological damage and enhanced the coordination between the economy and the environment.

5.4. Analysis of the Drivers of the Four Types of Regions

According to the results from the random forest analysis, the industrial structure plays a dominant role in the coupling coordination between economic development and environmental protection in Yunnan Province. For region I, the proportion of the secondary industry is the primary factor influencing their coupling coordination degree. In region II, foreign investment is the main factor, while in region III, technological innovation takes precedence. In region IV, the degree of marketization is the key influencing factor.

The industrial structure plays a dominant role in the coupling coordination between economic development and environmental protection in Yunnan Province. This is primarily because the industrial structure directly determines the nature of economic activities and their impact on the environment. Different industries have significantly varying effects on resource consumption and environmental pollution, and the optimization and adjustment of the industrial structure directly affect the balance between economic growth and environmental protection. Economic research indicates that optimizing the industrial structure, particularly transitioning towards the tertiary sector and high-value-added industries, can alleviate the negative impact of economic growth on the environment to a certain extent and promote coordinated development between the economy and the environment [67].

For region I, the proportion of the secondary industry is the dominant factor mainly because in region I, the proportion of the secondary industry is high, and the economic development of these cities depends on manufacturing, mining, and other heavy industries. The high proportion of secondary industries has led to a high degree of industrialization in this region, but it has also brought about more serious environmental pollution. For example, the accelerated industrialization process has led to a significant increase in energy consumption and a rise in the emission of pollutants, especially when environmental protection technologies are insufficiently applied, resulting in a sharp increase in environmental pressure. It has been shown that an over-representation of heavy industries and highly polluting industries often leads to a decrease in the degree of coupling coordination, as they directly increase atmospheric and water pollution as well as the excessive consumption of resources [68].

Foreign investment becomes the dominant factor affecting the coupling coordination degree of region II because foreign investment in these areas is mainly concentrated in the tourism and service industries, especially in Dali and Lijiang, where the development of tourism benefits from foreign investment, and these industries put relatively less pressure on the environment, thus promoting the coordinated development of the economy and the environment. It has been shown that foreign investment in developing countries is often able to improve local environmental quality through the introduction of green technologies and the enhancement of management efficiency [69]. However, the distribution of foreign investment is uneven, and certain regions may rely on resource-exploiting foreign investment, a pattern of investment that may have fueled economic growth in the short term but may have put some pressure on the environment in the long term. Therefore, foreign investment should be further encouraged to enter green and sustainable industries as a way to promote a win–win situation for both the economy and the environment.

Technological innovation is a key factor affecting the degree of coupling coordination of region III, mainly because the economies of this region rely on natural resources, and technological innovation, especially the introduction of green technology, can effectively reduce resource consumption and improve resource utilization. Therefore, technological innovation is particularly important in this region, which not only improves production efficiency but also promotes the recycling of resources and environmental protection. The introduction of green technologies, such as energy-saving and emission-reduction technologies and the use of environmentally friendly materials, can effectively reduce the consumption of natural resources and help these regions achieve the coordinated development of economic growth and environmental protection [70]. In addition, the ecological protection policies (e.g., forest protection and biodiversity conservation policies) in region III provide policy support for the promotion of technological innovation. With the enhancement of ecological protection awareness and the continuous development of green technology, the coordination between the economy and the environment in the third category of regions has been enhanced. As for region IV, its economic foundation is relatively weak, and the degree of marketization is the main factor affecting its coupling coordination degree. This is because, in economically underdeveloped areas, the increase in the degree of marketization helps to optimize resource allocation and improve economic efficiency [71]. For example, through market-oriented reforms, these cities can use the market mechanism to allocate resources more effectively, optimize the industrial structure, reduce the waste of resources, and improve economic efficiency. At the same time, the market mechanism can motivate enterprises to pay more attention to environmental protection, especially under the guidance of environmental regulations and incentives to promote the transformation of enterprises to a green development mode [63].

Based on the in-depth analysis of the reasons for the changes in the coupling coordination degree of different types of cities, the following policy recommendations are put forward. Firstly, industrial structure upgrading and green transformation should be promoted. For region I, the control of highly polluting industries should be strengthened, and the win–win situation of economic development and environmental protection should be promoted through the introduction of environmental protection technology and green transformation of industries. The government can support the transformation and upgrading of enterprises through green credit, tax incentives, and other measures to promote the rapid development of the service industry and high-value-added industries. For region II, support for green tourism and environmental protection services should be increased, while the foreign investment structure should be optimized to attract more foreign investment in green technology and environmental protection industries so as to promote the development of the green economy and low-carbon industries. For region III, support for green technology and environmental protection industries should be increased to promote the development of green industries, while ecological protection measures should be strengthened to ensure that economic growth does not come at the expense of the environment. Region IV should focus on market-oriented reforms to optimize resource allocation and improve economic efficiency. While upgrading the level of marketization, the implementation of environmental protection policies should be strengthened, and a sound environmental protection mechanism should be established. Second, policy implementation and cross-regional synergy should be strengthened. Regions, especially those in the second and fourth categories, should enhance the implementation of environmental protection policies and solve environmental problems through cross-regional synergy among local governments. The government can promote intra-regional cooperation in pollution control and resource sharing to enhance overall environmental governance. Finally, it should focus on strengthening technological innovation and green development. The government can support enterprises and local governments in introducing green innovative technologies through the establishment of green technology demonstration zones to promote the coordinated development of resource conservation and environmental protection.

The four-region division of Yunnan Province not only helps to accurately identify and respond to differences in economic development, environmental pressures, and policy implementation capacity among various regions but also provides an important basis for regional policymaking and optimal allocation of resources and has broad implications for other provinces in China and similar regions in other developing countries. Through regional division, development strategies and environmental protection measures can be tailored to each region, avoiding ‘one-size-fits-all’ policy implementation. Other provinces in China, such as Guizhou, Sichuan, and other resource-based provinces, as well as other developing regions (e.g., Southeast Asia and some resource-rich countries in Africa), can learn from Yunnan’s regional division model and formulate locally appropriate economic and environmental policies based on different economic fundamentals, resource endowments, and environmental pressure.

5.5. Limitations

This study reveals the interactive relationship between economic development and environmental protection in Yunnan Province through the coupling coordination degree model and the random forest algorithm. However, this paper mainly relies on GDP, FVC, and PM2.5 as the core indicators, and although these indicators are highly representative in reflecting the relationship between the economy and the environment, they may not be able to comprehensively capture other potential economic and environmental variables, and future studies may consider introducing more dimensional indicators to further enrich the breadth and depth of the analysis. In addition, the spatial and temporal scales of this research are relatively limited. Future studies could explore a larger geographical scope and a longer time frame to examine the interactive changes between the economy and the environment over extended periods, particularly focusing on the long-term effects following policy adjustments. In this paper, we select four major influencing factors –industrial structure, foreign investment, technological innovation, and degree of marketization– as the core factors for measuring the degree of coupling coordination between economic development and environmental protection. These factors help to reveal the main drivers of the differences in the coupling coordination degree between different regions. However, we also note that apart from these four factors, there are other potential influencing factors, especially environmental governance capacity, government policy implementation, and local culture. These factors may not only exacerbate or alleviate the contradiction between economic growth and environmental protection but also create a ‘superimposed effect’ of changes in the degree of coupling and coordination between different regions through multiple mechanisms, such as the government, the market, and technology. However, due to the challenges in collecting data on influencing factors across different states and regions, this study has only gathered a subset of influencing factors from most regions. Future research could incorporate more social and cultural factors to explore the coupling relationship between social development and environmental protection.

6. Conclusions

This paper classifies the regional GDP of each state and city in Yunnan Province into four types of regions. After that, firstly, it analyzes the time series change characteristics of economic development and environmental protection indicators in Yunnan Province in the past 20 years; secondly, it calculates the coupling coordination degree of each region of Yunnan Province in the past 20 years by using the coupling coordination model; and lastly, it calculates the contribution of the influence factors such as industrial structure and technological innovation to the coupling coordination degree by using the random forest algorithm, and draws the following conclusions.

(1)

The region I category includes Kunming, Qujing, Yuxi, and Honghe; the region II category includes Dali, Chuxiong, Dehong, and Lijiang; the region III includes Baoshan, Pu’er, Wenshan, and Xishuangbanna; and the region IV category includes are Zhaotong, Lincang, and Diqing.

(2)

The economies of all four types of areas are on a growing trend, but the growth rate has slowed down. The FVC of region I shows a “U” trend, the FCV of region II shows a “W” trend, region III, except for Baoshan, has been increasing at a rate of 0.0032 per year, and region IV has been increasing at a rate of 0.0003 per year. The PM2.5 concentrations in all four types of areas show an inverted “U” trend.

(3)

The degree of coupling coordination is closely related to the degree of economic development, with region I having the highest degree of coupling coordination, reaching the “barely coordinated” and “primary coordination” stages in general, region IV having the lowest degree of coupling coordination, being in the stage of “mild disorder” and “barely coordinated”, and region II and region III being in the range between “verging on disorder” and “barely coordinated”. The overall coupling coordination degree of economic development and environmental protection in Yunnan Province shows a decreasing trend; except for region III, the coupling coordination degree of the other categories of areas shows a decreasing trend during the study period.

(4)

Industrial structure is the most important factor influencing the degree of coupling coordination of economic development and environmental protection in Yunnan Province (74.9%), and the dominant influencing factor varies from region to region, with region I being most significantly influenced by the secondary industry (24.4%), region II being dominated by foreign investment (24.1%), and region III being more likely to see a greater contribution of technological innovation to the degree of coupling coordination (23.9%), whereas the less developed areas in region IV are mainly influenced by the degree of marketization (25.8%).