Research on Driving Factors of Forest Ecological Security: Evidence from 12 Provincial Administrative Regions in Western China

Abstract

Forests are associated with countrywide ecological security, and there are significant differences in the forests of different regions. Based on the DPSIR model, 25 indicators were selected from five dimensions to determine the index system, and the entropy-weighted TOPSIS method and gray correlation were applied to determine the index of western China’s forests. The spatial distribution map was used to observe the spatial changes of forests. The results show that first, Inner Mongolia (0.466) has the best forest ecological security status and Ningxia (0.124) has the worst forest resource status. Second, the first and most frequent correlation is the area of planted forests (I1). The last and most frequent correlation is sulfur dioxide emission (P2). Thirdly, Inner Mongolia and Szechwan belong to the high ecological safety–high economic level, Yunnan, Guangxi, and Tibet belong to the high ecological safety–low economic level, and Gansu and Guizhou belong to the low ecological safety–low economic level. The rest of the regions are classified in the low ecological security–high economic level. Fourth, the forest ecological security in western China has gradually become better, with the security index increasing from 0.417 to 0.469, with an average annual increase of 12.47%.

1. Introduction

Forest ecological safety assessment is a current urgent international issue and has become an important tool for forest condition assessment and forest resource management. Its evaluation results are the basis for developing better sustainable management programs and measures. Currently, forests are facing quality degradation, overuse of resources, and reduction of biodiversity, which seriously affect the security of forest ecosystems [1]. Based on this background, it is particularly important to maintain the sustainability of forest ecosystems. Western China is a vast and sparsely populated region, with a land area that accounts for about 70% of China’s national territory [2]. However, due to high-intensity human activities and unreasonable land use, the forest ecological space in western China has been heavily occupied and the regional forest ecology has been deteriorating. These phenomena have led to the generation of problems such as increased atmospheric pollution, frequent acid rain, and the decline of biological resources [3].

Forests are the core of natural ecosystems, and they are related to the ecological security of a country. Discussions on finding solutions to forest ecological problems have resulted in global conventions, local policies, and measures such as ecological parks (such as nature reserves, forest parks [4,5], and wetland parks), with the aim of achieving forest ecological sustainability. For example, REDD+ curbing deforestation is a cost-effective strategy with a significant impact on global GHG reductions [6]. Similarly, the 19^th Party Congress report has repeatedly emphasized the concepts of “green development”, “ecological civilization construction”, and "two mountain theory", which shows that China attaches great importance to economic efficiency and environmental issues. The report shows the urgency and relevance of establishing a monitoring and assessment system on forest ecological security. In this regard, the biggest problem in China today lies in the uneven and unstable development of its regions. Western China has previously been the least developed region in China. To date, western China has gradually become more aware of forest ecology issues. Although China has made many efforts to develop and implement forest ecological policies in the western region, the results are not satisfactory in view of frequent forest ecological vulnerability.

In recent years, scholars have studied ecological security with different perspectives, research scopes, and research methods [7]. Research perspectives have involved forests [8,9], land [10,11], and landscape [12]. The scope of the study covers urban agglomerations [13], watersheds [14,15], and cities [16], etc. Research methods include principal component analysis, PSR (Pressure-State-Responses) model [17], hierarchical analysis, ecological footprint [18,19,20], the DPSIR (Driving force–Pressure-State–Impact-Responses) model, and the value of ecosystem services [21,22]. Among them, the DPSIR model describes the human–environment interaction relationship under the natural environmental pressure state and is the most widely used evaluation index system at present [23]. The model effectively reflects the causal relationship of the system and integrates the factors of resources, development, environment, and human health, providing a scientific and theoretical basis for the construction of forest ecological safety indicators in western China. Linking socioeconomic, ecological, and human activities (destructive and restorative activities) in the DPSIR model for use in forest condition studies is a current scholarly concern. According to Ref. [24], the model is currently applied to ecological indicator construction, technical methodological research, and ecological assessment. For example, Nobre et al. [24,25], Sarmin et al. [26], and Adams et al. [27] provided a better framework for an indicator system in western China by synthesizing and describing the change patterns of forests from the perspective of DPSIR.

Researchers have made great progress in evaluating the status of Forest Ecological Safety [28]. However, the understanding of Forest Ecological Safety (FES) [29] has not been unified and includes aspects such as FES (in a narrow sense) and FES (in a broad sense). First, the narrow definition of FES refers to its own characteristics of integrity, goodness, and sustainability in terms of structure, function, and ecological processes within the forest ecosystem [30,31]. Second, the broad definition of FES is based on the development, production, management, and maintenance of forest ecosystems in many aspects, and should not only focus on the security of forest ecosystems in providing sustainable ecological services for human survival and industrial development [32], but also needs to focus on the repercussion security of human economic activities that pose threats to forest ecosystems.

A comprehensive analysis revealed that scholars have evaluated forest ecological security in China mainly in the Zhujiang Delta (nine cities) [33,34], the Yangtze River Economic Zone [35,36,37], the Beijing–Tianjin–Hebei area (Beijing, Tianjin, and Hebei Province) [38,39], and provincial areas [40,41,42], with fewer studies in western China. The establishment of indicators shows regional zonality, poor availability, and operability of indicator data. Therefore, to address the above deficiencies, this study constructed 25 indicators from 5 dimensions. The research team used entropy-weighted-Ideal Solution (TOPSIS) to calculate the ecological security index of western Chinese forests, which was used to determine the importance ranking of each indicator factor. Using the gray correlation method and ArcGIS10.8 technology, the correlation and spatial change status of western Chinese forests were calculated over a ten-year period.

2. Methods

2.1. Western China

Western China is shown in Figure 1. The high and low legends in the figure refer to the vegetation cover index (NDVI, Normalized Difference Vegetation Index), which can accurately reflect the surface vegetation cover status. This dataset effectively reflects the distribution and change status of vegetation cover at spatial and temporal scales in various regions of western China. This is a very important reference for the monitoring of vegetation cover change status, the rational use of vegetation resources, and other ecological- and environmental-related fields of research. The scope of the western region was delineated in the strategic plan for western development announced by the State Council. The western provinces in the traditional sense include the five southwestern provinces and urban areas (Chongqing, Szechwan, Yunnan, Guizhou, and Tibet) and five urban provinces in northwest China (Qinghai, Xinjiang, Ningxia, Gansu, and Shaanxi), Inner Mongolia, and Guangxi, with an area of about 6,781,600 square kilometers [43]. The time span studied was from 2012–2021.

Western China is a vast and sparsely populated region, covering about 70% of China’s territory, but containing only 29% of the country’s total population. In terms of local forest preservation, the forest area of the entire study area is approximately 124,807,300 hectares. Among them, the forest coverage of the northwest region is about 76,805,400 hectares, accounting for about 61.54% of the whole. The forest area in the southwest region is about 48,019,900 hectares, accounting for about 38.46% of the western region. As of 2021, soil erosion and desertification areas in western China already account for about 80% of the country’s soil erosion area and about 95% of the desertified land area. The amount of sediment flowing into the Yangtze and Yellow Rivers each year is up to more than 2 billion tons, thus leading to serious siltation of rivers, lakes, and reservoirs. At the same time, the frequent occurrence of flooding causes serious impacts on the national economy and production.

Compared with other regions, the western region has certain characteristics. First, compared with other regions, the natural environment in the western region is fragile and its ecological carrying capacity is relatively low. The western region has significant differences in climatic conditions, variable geological conditions, diverse types of landforms and a harsh natural environment. The northwest is arid and has little rainfall, and the degree of desertification is serious. The southwest has high mountains and steep slopes, shallow soils, and heavy rainfall. The climate of the Qinghai–Tibet Plateau is cold and harsh, and the air is thin. Second, compared with other regions, the western region is rich in resources and the local ecological environment has not yet deteriorated. The western region is rich in biological, mineral, land, and tourism resources, and is a natural-resource-rich region in China. The biodiversity of the region is of extreme importance nationally and globally. The local ecological environment in the region, such as the Hengduan Mountains and the Yarlung Tsangpo Grand Canyon, has not yet deteriorated. Third, compared with other regions, the western region has an important strategic ecological position. The west is the birthplace of the Yangtze River, the Yellow River, and other major rivers. The western region is an important ecological environment barrier in the central and eastern part of China and a key area to ensure China’s ecological security, as well as a sensitive area for global climate, so its strategic position is extremely important.

The forest ecosystem is the location of organic unity of interaction and interdependence between the human and forest environments. As the country attaches more and more importance to the construction of environmental protection, forest ecological safety is receiving more and more attention. The forest resources in the western region are very poor, and the western region is the birthplace of three major rivers: the Yangtze, the Yellow, and the Pearl, and the strategic position of the ecological environment means that the environmental security of the west directly affects the environmental security of the east and the whole country. Western China is a forest-poor region, and too few forest resources not only cause shortage of timber and forest products, reduction or even extinction of rare plants and animals in China, but also cause destruction of ecosystems, deterioration of environmental quality, and intensification of soil erosion, which leads to desertification and frequent floods and poses a potential threat to the sustainable development of China’s economy; the conservation of forest resources in western China is therefore urgent. It is of great practical significance to study forest ecological security in western China.

2.2. Research Methodology

In this paper, a comprehensive evaluation method combining both entropy–TOPSIS and gray correlation method is used, and ArcGIS software [44] is applied to analyze the spatial change status. The main purpose is to first determine the index weights using the objective evaluation method entropy weight method, and then measure and evaluate the level of forest development in the west using the entropy–TOPSIS method. Most previous studies have used methods such as principal component analysis, AHP, and the entropy method. However, the explanatory significance of the principal component analysis method is not only ambiguous, but also requires a high cumulative contribution rate of the extracted principal components, which often cannot fully reflect the differences among the evaluation objects when the sample size is small and the degree of differentiation is not high. AHP shows difficulty in avoiding the influence of subjective factors when determining weights. Although the entropy method can better determine the differences between indicator states, it cannot reflect the gap between the actual and ideal levels of forest development when targeting the same evaluation object. For this reason, this paper adopts the entropy–TOPSIS method to improve the evaluation object and the formula of positive and negative ideal solution values on the one hand, so that the evaluation results can further match with the actual situation; on the other hand, it overcomes the shortcomings of AHP and the traditional TOPSIS method, which mainly rely on experts’ subjective opinions to determine the weights. The study of forest ecological security is a multi-attribute decision problem, and entropy-weighted TOPSIS and gray correlation method can be applied to transform multi-attribute optimization into single-objective optimization. The integrated evaluation method both transforms multi-attribute optimization into single-objective optimization, eliminates subjectivity, and provides better guidance.

The entropy–TOPSIS method first calculates the weights of indicators and then applies TOPSIS to determine the forest ecological safety index. The entropy weighting method [45] is a mathematical method that considers what each indicator conveys, objectively assigns weights to each indicator, and calculates a comprehensive index. The method uses entropy to calculate the weight values based on the information provided by each evaluation index value to overcome some drawbacks. The TOPSIS approach is used to define the most effective and inferior options in a scheme and to locate the distance between every comparison object and the top of the line and inferior solutions, respectively. By calculating the closeness of each evaluation object to each program, the programs are ranked as superior or inferior. Combining the two methods, an entropy-weighted–TOPSIS-integrated model is constructed, which not only reflects the importance of each indicator, but also comprehensively reflects the dynamic evolution trend of forest conditions.

The gray correlation method effectively measures the degree of similarity and correlation between evaluation objects [46]. The approach weighs the diploma of the relationship between indications primarily based on the diploma of dependence of the developmental dynamics between them. The gray correlation between each index and the research objectives were calculated separately by the formula of correlation degree with certain accuracy. The decision principles and bases of the above two methods are different and have good complementarity. The model decision-making process is as follows:

Determine the evaluation object, collect the raw data, and obtain the initial decision matrix ${X=\left(x_{\mathrm{i}\mathrm{j}}\right)}_{n\times m}$. Among them, $n=25$. $x_{\mathrm{i}\mathrm{j}}$ is the attribute value of the $\mathrm{i}$th evaluation unit under the $j$th indicator.

In the first step, the indicators are normalized to obtain the normalized decision matrix ${V=\left(v\right)}_{n\times m}$. Equations (1) and (2) are the calculation steps.

$$V_{\mathrm{i}\mathrm{j}}=\frac{x_{\mathrm{i}\mathrm{j}}-\underset{i}{\min }\left(x_{\mathrm{i}\mathrm{j}}\right)}{\underset{i}{\max }\left(x_{\mathrm{i}\mathrm{j}}\right)-\underset{i}{\min }\left(x_{\mathrm{i}\mathrm{j}}\right)}$$

(1)

$$V_{\mathrm{i}\mathrm{j}}=\frac{\underset{i}{\max }\left(x_{\mathrm{i}\mathrm{j}}\right)-x_{\mathrm{i}\mathrm{j}}}{\underset{i}{\max }\left(x_{\mathrm{i}\mathrm{j}}\right)-\underset{i}{\min }\left(x_{\mathrm{i}\mathrm{j}}\right)}$$

(2)

In the second step, the index weights are determined. In Equations (3)–(5), $w_{\mathrm{j}}=\left(w_1,w_2,\cdots ,w_{\mathrm{n}}\right)$

$${\mathrm{p}}_{\mathrm{i}\mathrm{j}}=v_{\mathrm{i}\mathrm{j}}/\sum _{\mathrm{i}=1}^{\mathrm{m}}{v}_{\mathrm{i}\mathrm{j}}$$

(3)

$$e_{\mathrm{j}}=-1/\ln \left(m\right)\sum _{\mathrm{j}=1}^{\mathrm{m}}{\mathrm{p}}_{\mathrm{i}\mathrm{j}}·\ln {\mathrm{p}}_{\mathrm{i}\mathrm{j}}$$

(4)

$$w_{\mathrm{j}}=1-e_{\mathrm{j}}/\sum _{\mathrm{j}=1}^{\mathrm{m}}1-\left(e_{\mathrm{j}}\right)$$

(5)

In the third step, in Equation (6), the matrix $Y$ is obtained:

$$Y={\left(y_{\mathrm{i}\mathrm{j}}\right)}_{m\times n}={\left(w_{\mathrm{j}}{v}_{\mathrm{i}\mathrm{j}}\right)}_{m\times n}$$

(6)

In the fourth step, calculate$f_j^{+}$ and $f_j^{-}$ for Equations (7) and (8), respectively:

$$f_j^{+}=\underset{i}{\max }\left(y_{\mathrm{i}\mathrm{j}}\right)$$

(7)

$$f_j^{-}=\underset{i}{\min }\left(y_{\mathrm{i}\mathrm{j}}\right)$$

(8)

In the fifth step, the gray correlation is calculated. Equation (9) to Equation (12) calculate the correlation degree ${r_{\mathrm{i}}}^{+}$ and ${r_{\mathrm{i}}}^{-}$:

$${{\gamma }_{0\mathrm{i}\mathrm{j}}}^{+}=\frac{\underset{i}{\mathrm{m}\mathrm{i}\mathrm{n}}\underset{j}{\mathrm{m}\mathrm{i}\mathrm{n}}\left|{f_{\mathrm{j}}}^{+}-y_{\mathrm{i}\mathrm{j}}\right|+\rho \underset{i}{\mathrm{m}\mathrm{a}\mathrm{x}}\underset{j}{\mathrm{m}\mathrm{a}\mathrm{x}}\left|{f_{\mathrm{j}}}^{+}-y_{\mathrm{i}\mathrm{j}}\right|}{\left|{f_{\mathrm{j}}}^{+}-y_{\mathrm{i}\mathrm{j}}\right|+\rho \underset{i}{\mathrm{m}\mathrm{a}\mathrm{x}}\underset{j}{\mathrm{m}\mathrm{a}\mathrm{x}}\left|{f_{\mathrm{j}}}^{+}-y_{\mathrm{i}\mathrm{j}}\right|}$$

(9)

$${{\gamma }_{0\mathrm{i}\mathrm{j}}}^{-}=\frac{\underset{i}{\mathrm{m}\mathrm{i}\mathrm{n}}\underset{j}{\mathrm{m}\mathrm{i}\mathrm{n}}\left|{f_{\mathrm{j}}}^{-}-y_{\mathrm{i}\mathrm{j}}\right|+\rho \underset{i}{\mathrm{m}\mathrm{a}\mathrm{x}}\underset{j}{\mathrm{m}\mathrm{a}\mathrm{x}}\left|{f_{\mathrm{j}}}^{-}-y_{\mathrm{i}\mathrm{j}}\right|}{\left|{f_{\mathrm{j}}}^{-}-y_{\mathrm{i}\mathrm{j}}\right|+\rho \underset{i}{\mathrm{m}\mathrm{a}\mathrm{x}}\underset{j}{\mathrm{m}\mathrm{a}\mathrm{x}}\left|{f_{\mathrm{j}}}^{-}-y_{\mathrm{i}\mathrm{j}}\right|}$$

(10)

$${r_{\mathrm{i}}}^{+}=\frac{1}{\mathrm{n}}\sum _{\mathrm{j}=1}^{\mathrm{n}}{{\gamma }_{0\mathrm{i}\mathrm{j}}}^{+}$$

(11)

$${r_{\mathrm{i}}}^{-}=\frac{1}{\mathrm{n}}\sum _{\mathrm{j}=1}^{\mathrm{n}}{{\gamma }_{0\mathrm{i}\mathrm{j}}}^{-}$$

(12)

In the sixth step, the Euclidean distances ${d_{\mathrm{j}}}^{+}$ and ${d_{\mathrm{j}}}^{-}$ are calculated. Equations (13) and (14) are calculated as:

$${d_{\mathrm{j}}}^{+}=\sqrt{{\sum _{\mathrm{j}=1}^{\mathrm{n}}\left(f_j^{+}-y_{\mathrm{i}\mathrm{j}}\right)}^2}$$

(13)

$${d_{\mathrm{j}}}^{-}=\sqrt{{\sum _{\mathrm{j}=1}^{\mathrm{n}}\left(f_j^{-}-y_{\mathrm{i}\mathrm{j}}\right)}^2}$$

(14)

In the eighth step, the composite index is calculated. When $T_{\mathrm{j}}=1$, it is the highest; when $T_{\mathrm{j}}=0$, it is the lowest:

$$T_{\mathrm{j}}=\frac{{D_{\mathrm{j}}}^{-}}{{{{D_{\mathrm{j}}}^{+}-D}_{\mathrm{j}}}^{-}}$$

(15)

3. Evaluation Index System Construction and Data Calculation

3.1. Indicator System Construction

The DPSIR model is a continuous improvement of the PSR model, DSR (Driving force–State–Responses) model and PSIR (Pressure–State–Impact–Responses) model, which can better reflect the causal relationship between factors [47]. It is a conceptual model used to measure the environment and sustainable and healthy development, which is comprehensive, systematic, holistic, and flexible. This suggests that the DPSIR model is a more desirable evaluation model to assess the evolutionary trends of forest conditions.

Forest ecological security research mainly relies on methods such as the PSR model, DPSIR model, and SD (system dynamics) model. Among them, most of the studies using the PSR model have used provinces as the basic unit to observe the changes in the forest condition index between different provinces. The PSR model captures the relationship between human pressures on the ecology, the impact of the pressures on the ecology, and the feedback from humans on changes in the state of the ecosystem. However, as the natural environment becomes more and more closely related to social and economic aspects, as well as the complexity of its own characteristics and properties, the PSR model lacks the ability to make scientifically accurate judgments, and the inherent problems that exist. Therefore, the lookup crew delivered the driver and influence elements to the PSR mannequin to set up the DPSIR model (Figure 2). The indicators cover social, economic, resource, and environmental aspects, and describe the interconnection between numerous factors and forest ecological security in five dimensions.

The indicator system should be comprehensive and objective, and must contain a series of different measurement dimensions, focusing on specific and observable indicators. The following principles should be possessed: First, the principle of comprehensiveness. The construction of ecological safety evaluation indicators should be comprehensive in both content and structure. The comprehensiveness of content means that the index system should cover as many factors related to forest ecological safety as possible. The comprehensiveness of the structure means that the index system should pay attention to the hierarchical requirements, differences, and connections between regions and the whole, between different regions, and between different levels in the same region, to ensure an all-round, multi-faceted, and comprehensive assessment of the whole process and effects of the development of forest ecological safety. Second, the principle of operability. When constructing the index system, the accessibility and assessment ability of the indexes must be fully considered. When selecting the indicators, the use of existing information resources from the current statistical reports and relevant statistical departments to obtain data, or the use of existing data and indirect calculations, can be obtained. Indicators should be as simple and clear as possible and select representative indicators, not only to facilitate the collection and calculation of analysis, but also to save manpower, material, and financial resources. Third, the principle of comparability and independence. Comparability can be divided into vertical and horizontal comparability. In terms of vertical comparability, it must be able to dynamically reflect the construction status and development trend of the ecological safety level development process. In terms of horizontal comparability, the construction of the index system can measure and assess the variability of the development level of ecological safety in different regions. Independence means that indicators at the same level do not have an inclusion relationship and are not substitutable for each other.

First, the target layer is the general objective of this study. Second, the criteria layer is a subdivision and specification of the goal layer. They are Response (R), Pressure (P), State (S), Impact (I), and Drive (D), respectively. Third, each criterion layer is in turn composed of several indicator layers. Twenty-five indicators were selected through induction to build the indicator system (

Table 1. Forest ecological security indicator system in western China.

Target Level	Guideline Layer	Indicator Layer	Unit	Characteristic	References	Weights
A: Western China Forest Ecological Safety Index	D: Driving force	D1: Per capita GDP	RMB/person	+	Wu, H. et al. (2018) [48]	0.0446
		D2: Annual precipitation	Millimeters	+		0.0306
		D3: Natural population growth rate	People/square kilometer	−		0.0251
		D4: Gross annual product	Billions of dollars	+		0.0450
		D5: Secondary industry value added	Billions of dollars	+		0.0458
	P: Pressure	P1: Population density	%	−	Quinn, C. et al. (2017) [49]	0.0384
		P2: Sulfur dioxide emissions	Million tons	−		0.0760
		P3: Urbanization rate	%	−		0.0261
		P4: Nitrogen oxide emissions	Million tons	−		0.0585
		P5: Total energy consumption	Million tons of standard coal	−		0.0420
	S: Status	S1: Forest cover	%	+	Rudianto, R. et al. (2020) [50]	0.0365
		S2: Total afforestation area	Thousands of hectares	+		0.0235
		S3: Total standing wood accumulation	Billion cubic meters	+		0.0353
		S4: Average percentage of good days	%	+		0.0295
		S5: Water resources per capita	Cubic meter/person	+		0.0268
	I: Impact	I1: Area of planted forests	Million hectares	+	Baker, B. et al. (2022) [51]	0.0191
		I2: Area of forest pest and rodent infestation	Million hectares	−		0.0383
		I3: Number of forest fires	Times	−		0.0265
		I4: Total fire area	Hectares	−		0.0448
		I5: Area of affected forests	Hectares	−		0.0537
	R: Response	R1: Forestry investment	Million dollars	+	Zhang, X. et al. (2021) [52]	0.0296
		R2: Closure of mountains for forestry	Thousands of hectares	+		0.0359
		R3: Number of environmentally sound treatment plants	Seats	+		0.0416
		R4: Domestic waste removal volume	Million tons	+		0.0564
		R5: Industrial pollution control completed investment	Million dollars	+		0.0705

Note: “+” represents a positive indicator, and “−” represents a negative indicator.

).

As the “starting point” of the forest ecosystem, the driving force layer (D) is the potential cause of changes in forest ecosystem security. When the GDP shows rapid positive development, it will enhance the investment in ecological protection and make the regional forest ecological security level higher. Annual precipitation reflects the natural weather conditions of the forest. Excessive population increase can influence the grant of sources and the exceptions of the ecological environment. Economic development is gradually becoming a way for forests to enhance environmental quality, and the growth of annual gross product and secondary industry assets drives forest resource consumption. Therefore, the driving force layer is measured using GDP per capita (D1), annual precipitation (D2), natural population growth rate (D3), annual gross product (D4), and secondary industry value added (D5).

The stress layer (P) is progressively derived from the driver layer and is the epiphenomenal cause of forest ecosystem security problems, reflecting the pressure carried by the ecosystem. Population density can reflect the density of population activities in western China. The higher urbanization rate index reflects the higher interference of urban residents on the forest environment, which makes the forest environment react adversely. Pollutant gas emissions and total energy consumption reflect the pressure on the forest air environment, water environment, and land environment, and the indicator values measure the negative impact on the forest ecosystem. Therefore, pressure (P) is measured using population density (P1), sulfur dioxide emissions (P2), urbanization rate (P3), nitrogen oxide emissions (P4), and total energy consumption (P5).

The state layer (S) is formed by the impact of pressure and drive and reflects the state of the forest ecosystem’s own security. The forest cover, total afforestation area, and total standing wood accumulation reflect the richness of forest resources, and the higher the value, the better the forest resources in western China can be maintained. The average ratio of good days directly measures the state of forest security and reflects the state of regional atmospheric quality. The water resources of the forest are the core resources for optimizing the forest’s condition, and the larger the area of forest water resources, the more difficult it is to be disturbed by the outside world. Therefore, the state (S) is measured by forest cover (S1), total afforestation area (S2), total live wood accumulation (S3), average good days ratio (S4), and per capita water resources (S5).

Impact layer (I) is the interaction between human society and ecological security changes, reflecting the impact from forest ecosystem changes. Forest ecological security is, in most cases, affected by fires, so the quantity of woodland fires, fire area, and affected woodland area are chosen as the affect warning signs of forests. Forest biological disasters mainly include diseases, insects, and rodents, so the area of occurrence of forest diseases, insects, and rodents is selected as the impact indicator of a forest. Therefore, impact (I) is measured by using the area of plantation (I1), the area of forest pest and rodent occurrence (I2), the number of forest fires (I3), the total area of fires (I4), and the area of affected forests (I5).

The response stratum (R) is the active and effective response of forests to multiple stresses and can express the level of human governance needed to manage and protect forest ecosystems. Forestry investment and forest closure and are effective measures to protect forests. The number of environmentally friendly treatment plants refers to various domestic waste treatment facilities designed, constructed, operated, maintained, and managed in accordance with relevant technical, environmental, and sanitary standards and specifications, mainly including sanitary landfills, composting plants, and incineration plants, etc. Thus, forest stability is restored, reflecting local efforts to maintain the state of the forest. Additionally, the completed investment in industrial pollution control (R5) reflects the positive response made by the local people in protecting ecology and preventing environmental pollution. Therefore, response (R) is measured by using forestry investment (R1), forest closure (R2), number of environmentally friendly treatment plants (R3), the amount of domestic waste removal (R4), and the completed investment in industrial pollution control (R5).

3.2. Data Source

This paper selects data related to forest ecology for 12 provinces (cities and districts) in western China from 2012–2021. The data were all panel data, mainly from the statistical yearbooks of each region. In addition to the statistical yearbooks, data were obtained from the official websites of local governments in western China, such as the Climate Bulletin of Tibet Autonomous Region (2012–2021), the Water Resources Bulletin of Guizhou Province (2012–2021), the National Bureau of Statistics Annual Data by Province (2012–2021), the Statistical Bulletin of Tibet Autonomous Region (2012–2021), etc. All data and tables were processed by Excel and Stata MP software.

4. Results

4.1. Cross-sectional Comparison between Different Regions

Forests are crucial for sustainable human development [53]. Forest ecological protection has a necessary impact on retaining biodiversity, promoting sustainable financial development, and ensuring human survival. The relative proximity of the regions in western China ranges from 0.00 to 1.00. A higher relative proximity value indicates a better forest condition and the increased ability for the forest ecosystem to play its role. The relative proximity rankings derived from entropy-weighted TOPSIS (

Table 2. Relative proximity of regions of western China.

Region	${{\mathit{d}}_{\mathit{j}}}^{+}$	${{\mathit{d}}_{\mathit{j}}}^{-}$	Total Relative Proximity Value	Ranking
Chongqing	0.217	0.068	0.239	9
Szechwan	0.177	0.105	0.372	5
Yunnan	0.178	0.111	0.385	4
Guizhou	0.203	0.074	0.268	6
Tibet	0.188	0.144	0.434	2
Shaanxi	0.206	0.065	0.240	8
Gansu	0.224	0.037	0.141	10
Qinghai	0.224	0.034	0.133	11
Xinjiang	0.215	0.076	0.262	7
Ningxia	0.231	0.033	0.124	12
Inner Mongolia	0.169	0.148	0.466	1
Guangxi	0.174	0.125	0.419	3

) were calculated as follows.

The forest ecological safety index is ranked from high to low as follows: 0.466 (Inner Mongolia), 0.434 (Tibet), 0.419 (Guangxi), 0.385 (Yunnan), 0.372 (Szechwan), 0.268 (Guizhou), 0.262 (Xinjiang), 0.240 (Shaanxi), 0.239 (Chongqing), 0.141 (Gansu), 0.133 (Qinghai), and 0.124 (Ningxia). During 2012–2021, Inner Mongolia had the best forest ecological security, followed by Tibet and Guangxi. The relative proximity values are 0.466, 0.434, and 0.419, respectively, which are all higher than 0.400. This indicates that these three regions ranked high in terms of forest condition in western China. Yunnan, Szechwan, Guizhou, Xinjiang, Shaanxi, and Chongqing were ranked 4–9 in terms of relative proximity. This shows that the forest condition of these six regions is relatively weak and needs to improve ecological services such as regional landscaping and wind and sand control. Among the forest ecological security status of each region, Gansu, Qinghai, and Ningxia scored lower, with 0.141, 0.133, and 0.124, respectively, and the forest resource status was poor. This indicates that these three regions in western China had the lowest relative proximity value index and should increase forest ecological maintenance and enhance forest ecological construction and protection.

4.2. Longitudinal Comparison between Different Years

The ranking results of relative proximity from 2012 to 2021 were obtained based on the relative proximity derived from entropy-weighted TOPSIS. Considering the long-cycle nature of forest growth, to improve the accuracy and availability of data, 2012, 2015, 2018, and 2021 were selected as representative years with intervals of 3 years to analyze the evolutionary trends. The higher the rating index of the forest, the safer the condition of the forest. The evaluation values were divided into three evaluation levels: high security (0.36–0.60), medium security (0.21–0.35), and low security (0.00–0.20). ArcGIS 10.8 software was used to represent the differences more visually in the spatial distribution of forest ecological security indices among regions in western China from 2012-2021 (Figure 3). The lighter the color in the distribution chart, the lower the evaluation level, and vice versa. The evaluation levels of different areas correspond to different ecological safety conditions.

In 2012, in the distribution of evaluation grades in western China, the low-grade regions were Gansu, Qinghai, and Ningxia, the medium-grade regions were Chongqing, Guizhou, Xinjiang, and Shaanxi, and the high-grade regions were Yunnan, Tibet, Guangxi, Inner Mongolia, and Szechwan. The average value of comprehensive evaluation in this year was 0.283, and the regions above the average value were Szechwan, Yunnan, Tibet, Inner Mongolia, and Guangxi.

In 2015, the distribution of evaluation grades of the regions in western China included Gansu, Qinghai, and Ningxia in the low grade, Szechwan, Yunnan, Chongqing, Guizhou, Xinjiang, and Shaanxi in the medium grade, and Tibet, Inner Mongolia, and Guangxi in the high grade. The average composite rating for the year was 0.274, down 3.18% from 2012. Szechwan and Yunnan were downgraded from high to medium ranking due to pressure (P) and impact (I) indicators.

In 2018, in the distribution of evaluation grades of regions in western China, the low-grade regions were Gansu, Qinghai, and Ningxia. The medium-grade regions were Yunnan, Chongqing, Guizhou, Xinjiang, and Shaanxi. The high-grade regions were Szechwan, Tibet, Inner Mongolia, and Guangxi. The average composite rating for the year was 0.278, an increase of 1.46% compared to 2015. An increase of one high-ranking province was seen compared to 2015: Szechwan was upgraded from medium to high ranking, while other regions were evaluated with the same ranking. The reason for the upgrade of Szechwan’s ranking is the driver (D) indicator.

In 2021, the regions in western China with low rankings were Gansu, Qinghai, and Ningxia, while the regions with medium rankings were Inner Mongolia, Chongqing, Guizhou, Xinjiang and Shaanxi, and the regions with high rankings were Szechwan, Yunnan, Tibet, and Guangxi. The average composite rating for the year was 0.299, an improvement of 7.55% over 2018. There was an increase of one high-ranking province and a decrease of one high-ranking province compared to 2018, and the evaluation level of other regions remained unchanged. Yunnan was upgraded from medium to high ranking, and Inner Mongolia was downgraded from high to medium ranking. Szechwan was upgraded because of the driving force (D) indicator. Inner Mongolia was downgraded because of the pressure (P) indicator.

Overall, the forest ecological safety index improved by 5.65% between 2012 and 2021, and the forest ecological safety level in western China improved steadily in most areas. However, the Gansu, Qinghai, and Ningxia provinces were in the low grade, and the forest security grade in Chongqing, Guizhou, Xinjiang, and Shaanxi areas was not significantly improved. Geographically, the forest status of the five southwestern provinces was good, while that of the five northwestern provinces need to be improved. The forest status of Inner Mongolia and Guangxi was better. In the future, the five northwestern provinces should focus on strengthening forest restoration to increase forest resources and improve forest quality.

To provide a more detailed description of the characteristics of changes over different time spans, this study presents the relative proximity and ranking of each region in western China from 2012–2021 (

Table 3. Relative proximity and ranking of western Chinese provinces, 2012–2021.

Region	2012	Ranking	2013	Ranking	2014	Ranking	2015	Ranking	2016	Ranking
Chongqing	0.223	9	0.215	9	0.219	9	0.234	8	0.231	8
Szechwan	0.376	5	0.348	5	0.347	5	0.310	5	0.353	4
Yunnan	0.445	1	0.434	1	0.453	1	0.318	4	0.290	5
Guizhou	0.259	7	0.237	8	0.250	7	0.283	6	0.259	7
Tibet	0.410	2	0.413	2	0.401	4	0.390	3	0.415	2
Shaanxi	0.236	8	0.242	7	0.238	8	0.221	9	0.211	9
Gansu	0.137	10	0.113	12	0.130	11	0.135	10	0.137	11
Qinghai	0.125	11	0.116	11	0.135	10	0.110	11	0.141	10
Xinjiang	0.274	6	0.293	6	0.288	6	0.248	7	0.261	6
Ningxia	0.123	12	0.120	10	0.122	12	0.102	12	0.129	12
Inner Mongolia	0.402	3	0.349	4	0.423	2	0.418	2	0.396	3
Guangxi	0.385	4	0.362	3	0.408	3	0.515	1	0.509	1
Region	2017	Ranking	2018	Ranking	2019	Ranking	2020	Ranking	2021	Ranking
Chongqing	0.192	8	0.232	7	0.256	9	0.268	8	0.262	9
Szechwan	0.279	4	0.378	4	0.369	4	0.499	1	0.386	4
Yunnan	0.213	7	0.271	5	0.337	5	0.400	4	0.449	2
Guizhou	0.245	5	0.251	6	0.273	7	0.316	6	0.287	7
Tibet	0.357	2	0.419	2	0.460	1	0.460	2	0.454	1
Shaanxi	0.190	9	0.231	8	0.306	6	0.283	7	0.289	6
Gansu	0.115	11	0.172	10	0.162	10	0.178	10	0.178	10
Qinghai	0.130	10	0.138	11	0.155	11	0.163	11	0.152	11
Xinjiang	0.232	6	0.228	9	0.261	8	0.242	9	0.263	8
Ningxia	0.105	12	0.112	12	0.146	12	0.125	12	0.143	12
Inner Mongolia	0.555	1	0.484	1	0.457	2	0.367	5	0.317	5
Guangxi	0.342	3	0.415	3	0.400	3	0.414	3	0.409	3

). Based on the relative proximity and ranking of each region in western China from 2012–2021, the following conclusions were drawn.

First, the province with the highest ranking is Yunnan, which ranked first in relative proximity in 2012, 2013, and 2014 with a relative proximity of 0.445, 0.434, and 0.453, respectively. From 2012–2014, Yunnan had the highest forest ecological security index compared to other regions in western China. The main reason may be that Yunnan is in the upper Yangtze River basin area, which has a better natural environment in terms of annual precipitation and annual sunshine hours and is also strongly supported by the government in terms of afforestation investment and forest ecology construction. However, since 2015, the forest ecological safety index in Yunnan has been on a decreasing trend. Yunnan has three characteristics: mountainous, ethnic, and frontier, and the livelihoods of local people are highly dependent on forests and their by-products, thus adding to the ecological pressure in these regions. This suggests that it appears necessary to adopt an effective forest ecological benefit compensation mechanism in Yunnan, China.

Secondly, the province with the second highest number of rankings was Tibet. Tibet is in the second position in terms of relative proximity in 2012, 2013, 2016, 2017, 2018, and 2020 with a relative proximity of 0.410, 0.413, 0.415, 0.357, 0.419, and 0.460, respectively. Tibet ranked second in forest ecological security six times in the calendar year. The main reason is that Tibet is a treasure trove of precious and intact natural forests with advantages in terms of landscape type, ecological type, biological population, forest type, and growth distribution.

Third, the Ningxia, Gansu, and Qinghai were ranked low in relative proximity. The relative proximity values of these four regions were low in the time dimension. Among them, Ningxia has the lowest ranking. In the calculated results, the relative proximity was 0.123, 0.120, 0.122, 0.102, 0.129, 0.105, 0.112, 0.146, 0.125, and 0.143, respectively. In the cross-sectional comparison between different regions, Ningxia was in the last place in terms of total relative proximity, with a relative proximity of 0.124. This indicates that the forest condition index in Ningxia was consistently at the lowest level compared to other regions in western China, with low levels of stand quality and more fragile forests.

4.3. Relative Comparison of Different Indicators in the Same Region

Gray correlation analysis measures correlations between factors that change with time or object. The basic principle is the quantitative analysis of dynamic process trends, comparing geometric relationships with sequences. During the development of the system, if the trends of the two elements are consistent, it can be said that the correlation between these two elements is very high. On the contrary, it was lower. In this study, the strengths and weaknesses of the indicators were judged by analyzing the degree of correlation between each indicator and the research objectives (

Table 4. Gray correlation of indicators in western China by region.

Indicator Level	Chongqing	Szechwan	Yunnan	Guizhou	Tibet	Shaanxi	Gansu	Qinghai	Xinjiang	Ningxia	Inner Mongolia	Guangxi
D1	0.749	0.727	0.730	0.675	0.799	0.762	0.727	0.673	0.721	0.731	0.774	0.617
D2	0.757	0.758	0.728	0.817	0.827	0.741	0.760	0.695	0.720	0.739	0.799	0.777
D3	0.821	0.849	0.840	0.757	0.863	0.864	0.783	0.772	0.736	0.831	0.863	0.769
D4	0.746	0.726	0.729	0.678	0.792	0.760	0.726	0.670	0.716	0.733	0.774	0.615
D5	0.768	0.738	0.712	0.686	0.782	0.757	0.690	0.648	0.666	0.719	0.762	0.609
P1	0.791	0.722	0.759	0.712	0.792	0.767	0.798	0.658	0.739	0.773	0.784	0.613
P2	0.622	0.635	0.661	0.585	0.842	0.660	0.585	0.514	0.659	0.702	0.684	0.461
P3	0.759	0.755	0.750	0.693	0.814	0.912	0.727	0.673	0.731	0.774	0.781	0.651
P4	0.653	0.810	0.665	0.609	0.851	0.719	0.627	0.619	0.651	0.670	0.690	0.698
P5	0.803	0.757	0.692	0.786	0.774	0.816	0.771	0.798	0.759	0.714	0.714	0.655
S1	0.795	0.705	0.809	0.637	0.858	0.830	1.00	0.567	0.827	0.669	0.854	0.711
S2	0.855	0.720	0.819	0.787	0.857	0.846	0.674	0.687	0.734	0.781	0.885	0.631
S3	0.816	0.815	0.809	0.760	0.858	0.830	0.796	0.757	0.803	0.825	0.837	0.711
S4	0.858	0.762	0.778	0.812	0.836	0.826	0.759	0.712	0.755	0.800	0.800	0.634
S5	0.685	0.753	0.756	0.788	0.804	0.781	0.741	0.699	0.669	0.773	0.783	0.775
I1	0.845	0.673	0.846	0.777	0.908	0.850	0.701	0.717	0.866	0.773	0.908	0.598
I2	0.627	0.813	0.729	0.708	0.828	0.762	0.779	0.798	0.756	0.737	0.760	0.638
I3	0.637	0.768	0.629	0.529	0.711	0.765	0.735	0.667	0.724	0.703	0.804	0.664
I4	0.653	0.722	0.642	0.523	0.769	0.730	0.715	0.716	0.663	0.703	0.763	0.673
I5	0.666	0.760	0.641	0.527	0.751	0.739	0.629	0.667	0.656	0.624	0.741	0.628
R1	0.663	0.785	0.719	0.554	0.808	0.826	0.639	0.689	0.733	0.808	0.844	0.627
R2	0.797	0.741	0.705	0.753	0.857	0.772	0.806	0.745	0.763	0.733	0.757	0.657
R3	0.727	0.832	0.748	0.614	0.823	0.765	0.731	0.675	0.722	0.801	0.863	0.613
R4	0.667	0.728	0.726	0.624	0.761	0.832	0.702	0.751	0.672	0.794	0.836	0.577
R5	0.793	0.727	0.720	0.653	0.744	0.808	0.648	0.661	0.779	0.644	0.790	0.598

). The mean values of gray correlation of the indicators were all greater than 0.460.

From the top five parts of the gray correlation analysis (Figure A1 and Figure A2): First, the first and most frequent indicator in the gray correlation was the area of planted forests (I1), with a total of four times. The regions were Yunnan (0.846), Tibet (0.908), Xinjiang (0.866), and Inner Mongolia (0.908), respectively. The main reasons are the advantages of planted forests compared to natural forests are shorter forest maturity period, uniform distribution of standing trees, and the possibility of selecting tree species to meet production and livelihood needs. With socio-economic development, humans have put too much pressure on forests, making them overwhelmed. Therefore, plantation forestry facilitates intensive management and maintains high forest productivity, which is essential for sustainable human development. Second, the indicator with the highest number of occurrences in the top five gray correlations was the natural population growth rate (D3), with a total of ten occurrences. The regions were Chongqing (0.821), Szechwan (0.849), Yunnan (0.840), Tibet (0.863), Shaanxi (0.864), Gansu (0.783), Qinghai (0.772), Ningxia (0.831), Inner Mongolia (0.863), and Guangxi (0.769). The main reason is that the forest condition is closely related to the natural population growth rate, which means that the main pressure on ecological security comes from population. Excessive population growth leads to negative impacts such as deforestation for farming and deforestation for building houses, which destroy more and more forest resources.

From the last five parts of the gray correlation (Figure A1 and Figure A2): First, the last and most frequent indicator in the gray correlation is sulfur dioxide emissions (P2), with a total of six times. The regions were Gansu (0.585), Chongqing (0.622), Szechwan (0.635), Inner Mongolia (0.684), Shaanxi (0.660), and Guangxi (0.461), respectively. The main reason is that the effect of sulfur dioxide on plants is not the gas itself. Sulfur dioxide forms acidic raindrops containing sulfite, which damages the pores and stomata of plant leaves and affects photosynthesis. Second, the indicator with the highest number of occurrences in the bottom five parts of the gray correlation is sulfur dioxide emissions (P2), with a total of 11 occurrences. NOx emissions (P4) appear eight times in the bottom five ranking of gray correlations by region. The regions were Chongqing (0.653), Yunnan (0.665), Shaanxi (0.719), Gansu (0.627), Qinghai (0.619), Xinjiang (0.651), Ningxia (0.670), and Inner Mongolia (0.690). The main reason is with booming industrial development, the toxic substances in the air are increasing and the phenomenon of atmospheric pollution is becoming more and more serious. Acid rain is formed in the process of high air humidity and rainfall. Acid rain infiltrates into the soil, increases soil acidity, affects soil structure, and reduces soil fertility. At the same time, acid rain increases toxic ions such as aluminum, nickel, and copper in the soil and inhibits tree growth.

4.4. Type Analysis of Different Regions

There is an interactive mechanism between wooded area ecological safety and financial development. In the long run, ecological security promotes economic quality improvement, and economic development provides the necessary human, financial, technical, and institutional support for forest ecological security. Economic development cannot come at the expense of forest ecosystems, and the protection of forest ecosystems is for sustainable economic development. Based on this background, maintaining the sustainability of forest ecosystems is particularly important. The research team classified the regions in western China into four types based on the entropy power–TOPSIS calculation results and the economic level of each region in western China. The types were specifically classified as high ecological security–high economic level, high ecological security–low economic level, low ecological security–high economic level, and low ecological security–low economic level. The dimensional division criteria are ecological security index and GDP per capita. The numerical classification criteria are relative proximity ≥0.3 for high ecological safety and GDP per capita ≥40,000 for high economic level. Accordingly, the regions in western China were distributed in a four-quadrant map (Figure 4).

Using two types of data, the Ecological Safety Index and GDP per capita, it is possible to classify western China into different types. The results of the four-quadrant diagram show that, first, Inner Mongolia and Sichuan are of the high ecological security–high economic level. Inner Mongolia and Sichuan have relatively stable forest conditions and have the advantage of high economic development and good local fiscal revenues. The region should invest funds in eco-forestry construction projects such as low-efficiency forest renovation, ecological public welfare forest compensation, and coastal protection forest system construction. Second, Yunnan, Guangxi and Tibet belong to the high ecological security–low economic level. The forest conditions in Yunnan, Guangxi, and Tibet are relatively stable. For areas with sufficient forest resources, they can take advantage of their superior natural conditions based on maintaining forest resources and improve their economic status in combination with government policies at all levels. Third, Chongqing, Shaanxi, Xinjiang, Ningxia, and Qinghai belong to the low ecological security–high economic level. The regions with backward forest status should reduce human pressure on forests and guide construction projects not to occupy or occupy less arboreal forests, special shrub forests, or public welfare forest land. Fourth, Gansu and Guizhou belong to the low ecological security–low economic level. The above regions need to maintain their management of forest resources, especially the protection of natural forests. They should carefully implement the government’s forest land protection and utilization plan and take measures to strengthen the integrated management of degraded forest ecosystems.

4.5. Evolutionary Features of Western China

The comprehensive ecological safety index for western China from 2012 to 2021 was calculated (

Table 5. Relative proximity of western China, 2012–2021.

Year	${{\mathit{d}}_{\mathit{j}}}^{+}$	${{\mathit{d}}_{\mathit{j}}}^{-}$	Total Relative Proximity Value	Ranking
2012	0.160	0.115	0.417	10
2013	0.140	0.128	0.477	5
2014	0.116	0.145	0.556	1
2015	0.120	0.123	0.507	3
2016	0.135	0.097	0.419	9
2017	0.123	0.127	0.508	2
2018	0.126	0.112	0.470	7
2019	0.128	0.117	0.478	4
2020	0.140	0.125	0.472	6
2021	0.158	0.139	0.469	8

) and confirmed an upward style with small fluctuations from 2012 to 2021. The West China Composite Index expanded from 0.417 to 0.469: an ordinary annual expansion of 12.47%. It can be specifically divided into four stages: a rapid growth (2012–2014), rapid decline (2014–2016), small growth (2016–2017), and steady decline (2017–2021) evolutionary course (Figure 5). The overall trend of forest ecological safety level is increasing, but in recent years (2017–2021) the level of forest ecological safety still has large potential and space for improvement.

The fundamental reasons for this are: First, the accelerated urbanization and industrialization process has led to frequent floods in western China, the depletion of plant and animal resources, and the increasingly serious situation of forest ecological security. Second, forestry policy has shifted to focus on ecological construction and herbal wooded area aid protection, and the percentage of planted timber has decreased. At the same time, the government has increased the intensity of forest tourism development to meet people’s demand for leisure tourism and forest recreation, and human actions have significantly interfered with and affected forest ecosystems. Thus, it is recommended that the economic and ecological development of western China be coordinated, and forest ecological security be strengthened. In 2017–2021, the state of forests in western China was weak and there was negative agglomeration; therefore, the investment and protection of western forests should be increased to enhance the regional forest ecological security.

5. Discussion

(1) Among the dynamic trends in different regions, the indexes in the western region, from highest to lowest, were Inner Mongolia (0.466), Tibet (0.434), Guangxi (0.419), Yunnan (0.385), Szechwan (0.372), Guizhou (0.268), Xinjiang (0.262), Shaanxi (0.240), Chongqing (0.239), Gansu (0.141), Qinghai (0.133), and Ningxia (0.124). Inner Mongolia has the best forest ecosystem security. The second highest forest ecological safety index is Tibet, with 0.434. Ningxia ranks last among all regions with the worst forest ecological safety status. The Inner Mongolia and Tibet Forest ecosystems have been significantly improved, which to some extent indicates that the quality of forests in the region has been improved. The highest degree of Inner Mongolia is attributed to the forestry system dividends such as collective forest rights system reform and ecological public welfare forest compensation. However, there is still more room for the development of forest protection and afforestation in Qinghai and Ningxia. To a certain extent, Ningxia has a good ecological environment with mild climate, abundant rainfall, and fertile land. However, due to rapid population growth, some forests and grasslands have been turned into agricultural lands, leading to ecological deterioration.

(2) In the spatial distribution, most of the regions with the highest forest ecological safety grade in the four years 2012, 2015, 2018, and 2021 were distributed in Yunnan and Tibet, while most of the worst regions were distributed in Gansu and Qinghai. Among them, the highest forest ecological safety index in Yunnan may be attributed to the significant improvement in the quality of forests and grasses through the introduction of forest management planning and forest cultivation implementation rules, as well as precise supervision of forest and wetland resources. Tibet’s ecological security is in good condition due to the fact that more than 100.5 million hectares of public welfare forests have all been brought under protection in Tibet since 2012. Tibetan areas are under various forestry ecological management measures to achieve positive results.

(3) Among the statistical frequencies, the area of planted forests (I1) and the natural population growth rate (D3) have the highest correlation with forest ecological security in western China. Excessive population growth can cause overuse of resources, resulting in a smaller new environmental capacity than previously. All regions in western China should strengthen their control and management of human activities in forest areas and restrain illegal logging and deforestation. Sulfur dioxide emissions (P2) and nitrogen oxide emissions (P4) are the least associated with western China. In the context of the transformation of the countrywide monetary improvement mode, the administration of industrial and business waste in the western vicinity has been multiplied and emissions have been reduced. Therefore, there is plenty of space and room for improvement of forest ecological environment in the region.

(4) In the spatial correlation, the regions in western China are divided into different types of forest ecological safety-economic level. In response to the spatial distribution structure, local governments can adopt corresponding policies. The high ecological security–high economic level areas should actively cultivate emerging technologies and alternative products to protect the sustainable capacity of forestry development. In high ecological security–low economic level areas, species and ecological diversification should be implemented, and the actual value of forest resources should be enhanced by rational use and standardized development. The low ecological security–high economic level areas should slow down the pace of economic development and increase the forest maintenance work, such as reforestation of barren hills and wastelands. In low ecological security–low economic level areas, economic forestry should be created by artificial intervention to further increase economic benefits while maintaining ecology. For example: Jiang Yong County in China combines its own economic development system and the scale of forest resource development to build a base of famous and special fruits mainly fragrant pomelo. This not only provides an effective way to enhance the regional economy, but also lays the prerequisites for plantation forestry.

(5) In terms of temporal changes, the composite index of ecological protection of plantation forests in the western region increased from 0.417 in 2012 to 0.469 in 2021. To a certain extent, this indicates that although western China has relatively superior natural resource conditions, it also faces a series of challenges and the possibility of deterioration. Negative effects such as demand for forest ecotourism and forest-related products brought about by economic development and people’s improved living standards, pollutant emissions formed by industrial production, frequent natural disasters under extreme weather, and slower financial spending will all affect forest ecosystems. Western China should weigh the relationship between elements and ecology to forestall temporary conduct that sacrifices wooded area sources for financial growth.

6. Conclusions

China has carried out a great deal of of pioneering work to improve the forest ecological environment in the western region, with remarkable achievements. However, changing long-established ecological degradation, such as soil erosion, requires multiple considerations in order to maintain long-term stability. In view of this, the following recommendations are made.

(1) Build a virtuous cycle ecosystem and adhere to the greening of the forest. The study area has generally evolved from a state of surplus to a state of overload to a state of balance, but there are differences among regions. The wooded area insurance in the western area multiplied from 5.05% in 1977 to 13.02% in 2020, for example. Nevertheless, for the western region is an ecologically fragile area, it is necessary to follow the integrity and systemic nature of the ecosystem and its inherent laws to build a virtuous cycle ecosystem. For example, the Yan’an area in northern Shaanxi Province has carried out afforestation and other greening work, with remarkable achievements. Returning 7,215,900 hectares of land to forest, the vegetation coverage rate increased from 46% to 81.1% in 2017, and the green border was pushed north by more than four hundred kilometers. The greening experience of Yan’an deserves serious summary and promotion, and provides an example of ecological restoration for the world.

(2) Increasing policy support and investment in forest ecological construction. Sound ecological public welfare wooded area safety and compensation mechanisms according to the state of local financial resources to enhance the compensation standards and compensation efforts. Fully implement the ban on the commercial logging of natural forests and the management of logging of commercial plantations. Set up a special forest economy development to subsidize forest planting, forest farming, and forest landscape utilization projects.

(3) Explore the cost of wooded area ecological merchandise and rationalize the use of water resources. On the premise of protecting the integrity of forest ecosystems, develop forest ecotourism and cultural tourism, and cultivate forest recreation industry. Based on not destroying forest resources, develop woody oilseeds, forest economy and special economic forests. By exploring the mechanism of realizing the fee of woodland ecological products, the ecological benefits are modified into financial and industrial advantages, realizing ecological and monetary “win-win”. Ensure good public education and education to popularize and promote effective water conservation techniques. Change the long-standing bad habits of people who do not conserve water and take effective measures to carry out publicity and education work in a sustained manner. Empower digital technology and build a smart barrier. Make full use of digital technology in the information age to rationalize the layout of water resources by changing the current water tariff, volume-based taxation, and reasonable sewage treatment.

This study is based on the relationship between forest ecosystems and human social systems, whilst thinking about the contemporary scenario and traits of woodland assets themselves. The research team conducted an empirical study using entropy–TOPSIS and gray correlation analysis to measure the variability of forests in 12 regions and analyze the trends of the indices to ensure the accessibility of the data. In the future, we can strengthen the research on smaller-scale areas and explore the forest ecological security of specific research objects in different time periods, while balancing the criteria for judging forest ecological security among regions, so that the research results can have more targeted application value.