Landscape Ecological Risk Assessment and Zoning Control Based on Ecosystem Service Value: Taking Sichuan Province as an Example

Abstract

The escalation of urbanization in Sichuan Province has resulted in irrational land use and excessive resource exploitation. These developments have consequently diminished the functionality of ecosystem services and exacerbated landscape fragmentation in the region. These challenges pose significant threats to the ecological security of the area. In this study, we computed the ecosystem service value and the landscape ecological risk index of Sichuan Province from 2005 to 2018. We analyzed the spatial autocorrelation between the ecosystem service value and the landscape ecological risk index, constructing a framework for landscape ecological risk assessment and zoning control based on ecosystem service value. The results show the following: (1) Between 2005 and 2018, the total value of ecosystem services in Sichuan Province increased from CNY 10,261.17 × 10⁸ to CNY 10,310.43 × 10⁸, with forest land and grassland being the primary contributors to the ESV. (2) High- and higher-risk areas within the landscape ecology of Sichuan Province are expanding, forming a pattern of high-level risk concentration from west to east. (3) There exists a negative correlation between the ecosystem service value and the landscape ecological risk index in Sichuan Province. (4) In the ecological conservation zone, the principle of low-impact development must be upheld. In the ecological cultivation zone, adjusting the proportion of land use types is necessary to enhance the rationality of the land use structure. The ecological agricultural zone should fully utilize the advantages of agriculture, while the ecological improvement zone requires focused attention to ecological restoration and land remediation.

1. Introduction

Landscape patterns and ecosystem services form the foundation of sustainable research [1]. Ecosystem services not only offer a variety of ecological products for humanity but also encompass the essential elements for production and life, providing natural conditions vital for maintaining ecosystems and the processes essential for human life [2]. The value of ecosystem services represents the monetary worth of these services, aiding humans in comprehending the consequences of their actions regarding ecosystems and their well-being, thereby enhancing ecological awareness [3]. Nevertheless, rapid urbanization has spurred the expansion of construction land, altering land use and landscape patterns, diminishing ecosystem services and habitat quality, resulting in adverse ecological impacts, and ultimately leading to the degradation of ecosystem services [4].

Evaluating ecosystem service value and assessing landscape ecological risk are integral components of regional ecological security evaluation. These assessments epitomize the sustainable development of the intricate relationship between humans and their environment [5] and contribute to the sustainable growth of regional economies [6,7]. Research on ecosystem service value primarily delves into the impact of changes in land use types and landscape patterns on ecosystem service functions, alongside examining the spatial and temporal evolution and driving factors influencing these services. Landscape ecological risk studies concentrate on evaluating and quantifying landscape ecological risk as well as understanding their spatial and temporal evolution. This research also involves constructing an ecological security pattern based on landscape ecological risk [8]. As research into ecosystem service value and landscape ecological risk advances, some scholars have proposed a landscape ecological risk assessment system rooted in ecosystem services. They explore the spatial and temporal correlations and clustering patterns between the two [9], advocating for their integration from an independent standpoint. This integration primarily involves three aspects. First, it entails analyzing the correlation between ecosystem service value and landscape ecological risk. Various methods, such as using the coupling coordination degree model and spatial autocorrelation analysis, are employed to investigate this relationship. The study area is categorized into different types based on value and risk, followed by the implementation of stringent zoning control [10,11,12]. Second, it involves incorporating ecosystem services into the ecological risk assessment framework, thereby enhancing the assessment’s scientific rigor [13,14,15]. The third aspect delves into studying the impact of changes in landscape patterns on ecosystem service value [3].

With the progression of territorial spatial planning and the construction of an ecological civilization, landscape ecological risk assessment has progressively entered the domain of territorial spatial ecological restoration. Certain scholars have concentrated on the spatial efficacy of policies in regional planning [16] or proposed spatial planning strategies to address risks [17]. However, in most existing studies, ecosystem service value and landscape ecological risks often exist independently, leading to insufficiently targeted risk management and control. There has been limited integration of these elements. In practical management and control, integrating these aspects into urban and county land space ecological restoration planning poses challenges, making the implementation of plans complex.

Sichuan Province serves as a vital economic hub in western China and acts as a significant ecological shield in the upper reaches of the Yangtze River [18]. During the ‘Twelfth Five-Year Plan’ period, Sichuan Province’s urbanization rate surpassed the national average, undergoing rapid urbanization. This swift urban expansion escalated the demand for natural resources, leading to notable alterations in land use types and landscape patterns. Consequently, these changes have severely impacted the region’s ecosystem service function. Situated at the confluence of the first and second terrain levels of mainland China, Sichuan faces a high susceptibility to natural disasters due to its considerable terrain variations and increasing population density. These external stress factors have amplified interference with the local ecosystem. Presently, the existing ecological zoning management and control strategy insufficiently addresses the intricate relationship between ecosystem services and landscape ecological risks. Hence, effective ecological protection and restoration efforts in Sichuan Province have become imperative tasks.

This study aims to (1) improve the landscape ecological risk assessment framework based on the perspective of ecosystem services (Figure 1); (2) analyze the spatial and temporal dynamics of landscape ecological risk in Sichuan Province from 2005 to 2018; and (3) assess the overall ecosystem status in Sichuan Province by integrating ecosystem services and landscape ecological risks, with the findings contributing to the field of ecological risk management and control. The evaluation system is anticipated to enrich and refine the existing landscape ecological risk assessment framework, offering valuable insights for sustainable landscape development and ecological civilization construction in Sichuan Province.

2. Materials and Methods

2.1. Study Area

Located in the upper reaches of the Yangtze River, Sichuan Province spans the first and second steps of mainland China’s terrain, characterized by high elevations in the west and lower elevations in the east, creating a substantial height variation (Figure 2) [19]. The province encompasses approximately 486,000 km², extending from a latitude of 26°03′ to 34°19′ N and a longitude of 97°21′ to 108°12′ E. Sichuan Province has a pronounced monsoon climate [20], with an annual precipitation level ranging from 1000 to 1200 mm. Sichuan has a remarkable collection of natural resources, strong biodiversity, and outstanding eco-strategies. Its terrain exhibits notable variations, and economic development within its urban areas is characterized by disparities. Furthermore, urban resident populations are steadily expanding, resulting in escalating consumption and heightened demands for natural resources. In addition, Sichuan province needs to dedicate 13.18% of its area as a region for national ecological protection and restoration, and more than half of this area is concentrated in the western Sichuan area, so the task of ecological civilization construction is still very difficult.

2.2. Data Source

The dataset includes administrative division data, land use data, normal difference of vegetation index data (NDVI), net primary productivity data, (NPP) and socio-economic data (

Table 1. Data resource.

Data	Years	Spatial Resolution	Source
Administrative division	2018	Vector data	http://www.resdc.cn/ (19 January 2023)
Land use	2005, 2010, 2015, 2018	30 m
DEM	2005	500 m
NDVI	2005, 2010, 2015, 2018	250 m	https://modis.gsfc.nasa.gov/ (20 January 2022)
NPP	2005, 2010, 2015, 2018	500 m
Socio-economic	2005, 2010, 2015, 2018	Statistical data	http://tjj.sc.gov.cn/ (26 January 2022)

). Land use data were resampled to 500 m resolution using ArcGIS10.2, and the primary land use types were reclassified into six first-level categories (cropland, forestland, grassland, water body, construction land, and bare land) based on the National Standard Land-Use Classification of China using supervised classification.

2.3. Methods

2.3.1. Calculation of Ecosystem Service Value

(1)

Determination of ESV equivalent factor per unit area.

As determined via using the ecosystem service unit area value equivalent factor method to calculate ESV, the equivalent factor of ESV per unit area is 1/7 of the economic value of average grain yield in a 1 hm² farmland ecosystem [21,22,23]. The average grain price was 1.78 CNY/kg, and the average grain yield per unit area was 6021.66 kg/hm². In order to eliminate the impact of rising prices on food prices, the consumer value index (CPI) was used to modify the ESV equivalent [24] (Formulas (2)–(4)), and, finally, the ESV equivalent factor per unit area of Sichuan Province was determined to be 1561.85 CNY/hm².

$$D{M}_0=\frac{1}{7}\times G\times \frac{1}{n}\sum _{i=1}^{n}S,i=\mathrm{1,2},3\cdots n$$

(1)

Here, DM₀ refers to ESV equivalent factor per unit area (CNY/hm²); G and S denote the average price of grain in Sichuan Province and the average grain yield in Sichuan Province (kg/hm²), respectively; and n is year of the study period.

$$B_i=\frac{\mathrm{CPI}}{100}$$

(2)

$$\frac{B}{B_0}\times \frac{B_2}{B_1}\times \frac{B_3}{B_2}\times \cdots \frac{B_n}{B_{n-1}}=\frac{B_n}{B_0}$$

(3)

$$D{M}_0=D{M}_0\times \frac{B_n}{B_0}$$

(4)

Above, CPI is the consumer price index, $B_i$ is the CPI of the elimination unit, $\frac{B_1}{B_0}$ is the first period of the CPI, and $\frac{B_n}{B_0}$ is the current year-on-year fixed base CPI.

(2)

Correction of ESV coefficient in Sichuan Province

In this study, the analysis of ecosystem service functions encompassed four primary-level services and eleven secondary-level services. The coefficient representing the value of the ecosystem service function was derived by converting the proportional coefficient using the equivalent factor (

Table 2. The ESV coefficient of each ecosystem service function in Sichuan Province (CNY/hm²).

	Land Use Type	Farmland	Woodland	Grassland	Watershed	Construction Land	Unutilized Land
Ecological Service
Providing service	Food production	1733.65	390.46	359.23	624.74	0	15.62
	Raw material production	390.46	905.87	531.03	187.42	0	31.24
	Water supply	−2046.02	468.56	296.75	8168.48	0	15.62
Regulating service	Gas regulation	1390.05	2983.13	1889.84	749.69	0	109.33
	Climate regulation	734.07	8918.16	4982.3	2217.83	0	78.09
	Environmental purification	218.66	2608.29	1639.94	4466.89	−4263.85	327.99
	Hydrological regulation	2342.78	5841.32	3654.73	85,417.58	−13,088.3	187.42
	Soil conservation	812.16	3623.49	2295.92	734.07	0	124.95
Supporting service	Nutrient cycling	249.9	281.13	171.8	62.47	0	15.62
	Biodiversity	265.51	3311.12	2092.88	1999.17	0	109.33
Cultural service	Aesthetic	124.95	1452.52	921.49	1546.23	0	46.86

). The calculation formula is shown below

$$D{M}_{ij}=c_{ij}D{M_0}^{\prime },i=\mathrm{1,2},3\cdots 6;j=\mathrm{1,2},3\cdots 11$$

(5)

where i is ecosystem services function type i, and j is land use type j.

(3)

Calculation of Ecosystem Service Value based on grid unit

The 5 km × 5 km grid unit consistent with the landscape ecological risk evaluation was used to calculate the ESV of the grid unit in Sichuan Province. The formula is as follows:

$$\mathrm{ESVI}=\sum _{i=1}^{11}\sum _{j=1}^6(A_j\times D{M}_{ij}),i=\mathrm{1,2},\cdots 11;j=\mathrm{1,2},\cdots 6$$

(6)

Here, ESVI represents the ESV (CNY) of each grid; A_j and DM_ij refer to the area of land use type (hm²) and ESV coefficient of i ecosystem service function of j land use type (CNY/hm²), respectively; i is ecosystem services function type I; and j is land use type j.

2.3.2. Calculation of Landscape Ecological Risk Index

(1)

Calculation of risk index

The risk index was calculated by multiplying the probability index by the loss index, and the weight of each index layer of the probability index was determined using the entropy method (

Table 3. The ESV of each ecosystem service function in Sichuan Province (CNY/hm²).

	Dimension Layer	Indicator Layer	Property	Formula	Remarks
LS				$LS=\frac{ES{V}_a-ES{V}_b}{\mathsf{\Delta }T}$	LS is the ESV loss index; ESVa and ESVb are the ESV of periods a and b, respectively; and ∆T is the time interval between the two periods.
PROB	Terrain factors	Terrain index (T)	Positive	$T=\mathit{lg}\left[(\frac{E}{\overline{E}}+1)\times (\frac{S}{\overline{S}}+1)\right]$	$\mathrm{In}\mathrm{the}\mathrm{second}\mathrm{equation},T\mathrm{is}\mathrm{Terrain}\mathrm{index};E\mathrm{and}\overline{E}$$\mathrm{are}\mathrm{the}\mathrm{elevation}\mathrm{value}\mathrm{of}\mathrm{any}\mathrm{point}\mathrm{in}\mathrm{the}\mathrm{region}\mathrm{and}\mathrm{the}\mathrm{average}\mathrm{elevation}\mathrm{value}\mathrm{of}\mathrm{the}\mathrm{area}\mathrm{where}\mathrm{the}\mathrm{point}\mathrm{is}\mathrm{located},\mathrm{respectively};\mathrm{and}S\mathrm{and}\overline{S}$ are the slope value of any point in the region and the average slope value of the area where the point is located, respectively. The larger the elevation value, the larger the slope value and the Terrain index, and vice versa.
	Human stress	Proportion of construction land area (Jg)	Positive	$J_g=\frac{A_{g^{\prime }}}{A_g}$	Human stress is used to indicate the degree of harm human disturbance factors inflict on an ecosystem. In this study, the proportion of construction land area is used to indicate this measure.
	Ecological resilience	NPP	Negative	$Z_{ij}=\frac{X_{ij}-{\mathit{min}}_i{X}_{ij}}{{\mathit{min}}_j{X}_{ij}-{\mathit{min}}_j{X}_{ij}}$	Ecological resilience is expressed in terms of NPP and NDVI, and the original data must be normalized.
		NDVI	Negative
	Landscape vulnerability	Probability of connectivity (PC)	Negative	$PC=\frac{{\sum }_{i=1}^n{\sum }_{j=1}^n{a}_i\times a_j\times {P_{ij}}^{*}}{{A_L}^2}$	$\mathrm{In}\mathrm{this}\mathrm{equation},\mathrm{the}\mathrm{range}\mathrm{of}\mathrm{PC}\mathrm{is}[0,1]$, and the larger the value, the larger the connectivity, while Pij* is the maximum possibility of the connection between patch i and patch j.In this formula, PCremove represents the landscape connectivity index of the remaining patches after removing a patch, and dPC represents the importance index of landscape connectivity.
				$dPC=\frac{PC-P{C}_{remove}}{PC}\times 100\%$
		Landscape vulnerability index (LVI)	Positive	$LVI=LSI\times (1-LAI)$	Here, LSI is the Landscape Sensitivity Index, and LAI is the landscape fitness index. Ui is the Landscape Disturbance Index, which represents the degree of interference of external factors in the landscape pattern; it is obtained by weighting fragmentation (FN), the reciprocal of the fractal dimension (FD), and dominance (DO) as a, b, and c-weights, which are 0.5, 0.3, and 0.2, respectively.Combined with the land use situation in Sichuan Province, the landscape vulnerability of land ecological types is divided into the following weights: construction land—1, water—2, forest land—3, cultivated land—4, grassland—5, and unused land—6.
				$LSI={\displaystyle \sum _{i=1}^n}{U}_i\times V_i$
				$U_i=aF{N}_i\times bF{D}_i\times cD{O}_i$
				$LAI=PRD\times SHDI\times SHEI$
	Entropy method			$E_j=-\frac{1}{\mathit{ln}(m)}{\displaystyle \sum _{i=1}^m}{P}_{ij}\mathit{ln}(\frac{1}{P_{ij}})$	In this last equation, Ej is the uncertainty of index j (entropy value), m is the number of evaluation indexes, Dj is the certainty of index entropy, and Wj is the weight of index j.
				$D_j=1-E_j$
				$W_j=\frac{D_j}{{\sum }_{i=1}^m{D}_j}$

). The formula is shown below

$$PROB=\sum _{j=1}^n(W_j\times PRO{B}_j)$$

(7)

$$ESRISK=LS\times PROB$$

(8)

where ESRSIK stands for landscape ecological risk index; LS and PROB are the loss index and probability index, respectively; PROB_j is the probability value of j; and W_j is the weight of j.

(2)

Non-dimensional normalized processing

Due to the different units and magnitudes of each index, each index was subjected to non-dimensional normalized processing according to the positive and negative attributes in the final calculation of landscape ecological risk. The calculation method is shown below

$$Z_i=\left(X_i-X_{min}\right)÷\left(X_{max}-X_{min}\right)$$

(9)

$$Z_i=\left(X_{max}-X_i\right)÷\left(X_{max}-X_{min}\right)$$

(10)

where Z_j, X_max, X_min, and X_i are the normalization index, the maximum value of the index, the minimum value of the index, and the specific value of the index, respectively.

2.3.3. ESV Risk Spatial Autocorrelation Analysis

In this study, the spatial relationship between ESV and landscape ecological risk in Sichuan Province is reflected by global Moran’s I using a bivariate spatial autocorrelation analysis model. Furthermore, local indicators of spatial associations (LISA) were employed to capture local agglomeration and dispersion. Spatial mapping was performed with Geoda1.20 software. The formula is shown below

$$I_{vk}=\frac{m{\sum }_{i=1}^m{\sum }_{j=1}^m{W}_{ij}\times (\frac{y_{i,v}-{\overline{y}}_v}{{\partial }_v})\times (\frac{y_{i,k}-{\overline{y}}_k}{{\partial }_k})}{(m-1){\sum }_{i=1}^m{\sum }_{j=1}^m{W}_{ij}}$$

(11)

where I_vk is the bivariate global autocorrelation coefficient of grid ecosystem service value v and landscape ecological risk k; y_i,v denote the ith-grid ESV; y_i,k is the ith-grid landscape ecological risk; ∂_v and ∂_k denote variance; and w_ij denotes the spatial weights.

2.3.4. Division of Zoning Management

Research on ecological zoning stems from natural zoning. With the continuous development of natural zoning, ecological zoning research has been applied around the world [25]. Researchers have conducted extensive research from the perspectives of ecological restoration [26], ecological risk, the ES supply and demand relationship [27], and ES clusters [28].

The ESV and risk have five local spatial autocorrelation aggregation types, namely, HH, HL, LH, LL, and NN, which represent the high and low aggregation characteristics of ESV and risk in space. Regarding the NN non-significant regions, which are subject to the polarization and diffusion effects from adjacent spatial types, they need to be combined and treated as one category. In this study, the following principles were applied for attribute merging: Firstly, the diffusion effect (diffusion zone) is represented by local spatial positive correlation types, including HH and LL; the adjacent NN type is classified into these types. Secondly, the polarization effect (polarization zone) is represented by local spatial negative correlation types, including HL and LH; the adjacent NN type is classified into these types.

3. Results

3.1. Spatiotemporal Changes Analysis of ESV in Sichuan Province

This study focuses on the years 2005, 2010, 2015, and 2018 to analyze ecosystem service values in Sichuan Province (

Table 4. The structure of and change in the ESV of various land use types in Sichuan Province from 2005 to 2018.

Land Use Type	ESV (CNY 108)				2005–2018
	2005	2010	2015	2018	The Change in ESV	The Change Rate of ESV
Farmland	735.88	729.12	721.01	707.37	−28.51	−3.87%
Woodland	5937.46	6001.73	6027.07	6100.25	162.79	2.74%
Grassland	3112.12	3080.63	3061.01	3046.05	−66.07	−2.12%
Watershed	518.42	554.66	542.95	525.03	6.61	1.28%
Construction land	−40.38	−51.15	−66.10	−72.87	−32.49	−80.46%
Unutilized land	3.67	3.36	4.18	4.59	0.93	25.35%
All	10,267.17	10,318.35	10,290.12	10,310.43	43.26	0.42%

). As indicated in Table 4, the total ecosystem service values for these years were CNY 10,267.17, 10,318.35, 10,290.12, and 10,310.43 × 10⁸, respectively. These figures display a fluctuating upward trend, marking a total increase of CNY 4.326 × 10⁸. Upon examining specific land use types, the ecosystem service value of farmland decreased by 3.87%, while that of grassland decreased by 2.12%. Notably, the value of ecosystem services for construction land plummeted by 80.46%, resulting in a total decrease of CNY 12.707 × 10⁸ for these three categories. Conversely, woodland, watersheds, and unutilized land experienced increases of 2.74%, 1.28%, and 25.35%, respectively, contributing to a total increase of CNY 17.033 × 10⁸. This upward trend in ecosystem service value was predominantly driven by expansions in woodland and watershed area. Upon analyzing the impact of various land use types on ecosystem service value in Sichuan Province, it becomes evident that forest land and grassland make the most significant contributions.

Based on the coefficients derived from Formula (6) and Table 2, the ecosystem service value assessments for Sichuan Province in 2005, 2010, 2015, and 2018 were computed. Employing the best natural breakpoint method, the results were categorized into five grades: lowest ESV, lower ESV, medium ESV, higher ESV, and highest ESV, respectively (Figure 3).

As illustrated in Figure 3, regions with low ESV designations were predominantly clustered in the eastern part of Sichuan Province, with the lowest areas concentrated around Chengdu and its neighboring counties. This area constitutes the economic and cultural core of Sichuan Province, characterized by low altitude, flat terrain, high population density, numerous cities, and a high level of urbanization. Consequently, the overall Ecosystem Service Value (ESV) for this region remained low. In contrast, high-ESV areas are concentrated in the Western Sichuan Plateau, characterized by high altitude, favorable ecological conditions, sparse population, and low urbanization levels, resulting in significantly higher ESV. Analyzing the patterns of change from 2005 to 2018 reveals that the low-ESV regions in the eastern part of Sichuan Province have generally expanded in the direction of Guangyuan, Bazhong, and Dazhou, forming a ‘C’-shaped trend. Addressing the rising trend of low ESV in the eastern part of Sichuan Province while maintaining the high-ESV trend in the western region poses a significant challenge in the ecological construction and governance strategies of Sichuan Province.

3.2. Spatiotemporal Change Analysis of Landscape Ecological Risk in Sichuan Province

3.2.1. Temporal Changes of Landscape Ecological Risk

Based on the aforementioned calculation methods, the landscape ecological risk evaluation results for Sichuan Province in 2005, 2010, 2015, and 2018 were obtained. Employing the optimal natural breakpoint method, the results were classified into five distinct grades, denoted as [0, 0.25], [0.25, 0.45], [0.45, 0.60], [0.60, 0.70], and [0.70, 1.0], which represent lowest risk, lower risk, medium risk, higher risk, and highest risk, respectively (Figure 4).

(1)

Changes in general risk transfer

Referring to the change of risk in the four periods, the lowest-risk area decreased by 49,160.79km², while the lower-risk area decreased by 5188.5km², and the dominance of low-grade risk decreased. However, the high-grade risk area increased, within which the higher-risk area increased by 30,008.48 km², while the highest-risk area increased by 21,014.15 km², and the risk level increased significantly compared with 2005. In addition, the medium-risk area increased by 3326.7 km². It can be seen from Figure 5 that landscape ecological risk showed a significant trend during the study period. From the beginning of the low-grade risk-dominated period to the middle-risk dominated period and finally the rapid growth in high-grade risk, the area of low-grade risk was the largest, showing the polarization of high- and low-grade risks.

Utilizing ArcGIS10.2, the spatial superposition calculation and analysis of landscape ecological risks of various grades during the periods 2005–2010, 2010–2015, and 2005–2018 were carried out, and the results regarding risk area transfer changes are shown in Figure 6. As can be gleaned from the change in risk transfer at each level, on the one hand, lower-risk and lowest-risk areas account for a large proportion; on the other hand, the areas of higher risk and highest risk exhibited a gradual increase. These findings suggest the emergence of a relatively stable trend in landscape ecological risk over time.

3.2.2. Spatial Changes in Landscape Ecological Risk

(1)

The overall spatial distribution characteristics

During the designated study period, landscape ecological risk in Sichuan Province revealed a pattern characterized by high risk in the western region and low risk in the eastern region (Figure 4). Compared to 2005, there was a slight increase in medium-risk, higher-risk, and highest-risk areas in 2010; furthermore, the area of lowest risk transformed into a lower-risk area and the medium-risk area became larger, while the area of medium risk transformed into a higher-risk area, and the risk was developing towards an increasing trend. Compared to 2010, there was an increase in medium-risk, higher-risk, and highest-risk areas in the original region in 2015. The risk shows a steady growth trend; only a few lower-risk areas were converted to medium-risk areas. Compared to 2015, the landscape ecological risk situation in 2018 increased significantly, especially in the transformation of medium-risk areas into higher-risk and highest-risk areas, which was primarily observed in the Ngawa Tibetan and Qiang Autonomous Prefecture, the Ganze Tibetan Autonomous Prefecture, and the central urban area of Chengdu. These findings emphasize the rapid expansion of landscape ecological risk areas and a heightened severity of risk distribution during this period, posing a substantial threat to the ecological security of Western Sichuan and the Chengdu Plain.

(2)

Analysis of center-of-gravity migration characteristics

The displacement of the risk center of gravity serves as a valuable indicator for understanding the spatial distribution, direction, and potential trajectory of risk. When attempting to intuitively depict the spatial characteristics and trends of elements, the utilization of the standard deviation ellipse is highly effective [29]. In this study, we employed landscape ecological risk grades as the focus of analysis and conducted an investigation utilizing the standard deviation ellipse tool available in ArcGIS 10.2, allowing us to obtain the results regarding the center-of-gravity migration characteristics shown in Figure 7.

During the process of each risk grade’s center-of-gravity migration, a distinct pattern of high-grade risk emergence from west to east was exhibited. Analyzing the spatial trajectory of the center-of-gravity movement during four periods, it can be noted that the center of gravity of the lowest risk grade was primarily concentrated on the boundary between Chengdu and Ziyang; it shifted towards the southwest, then proceeded to the north, and finally relocated to the southwest again. The center of gravity of lower risk predominantly resided in the western region of Chengdu and the northwestern region of Ya’an. It followed a trajectory of northeastward movement initially; then, it moved to the east and ultimately moved to the west. The center of gravity of medium risk was concentrated around the boundary of Ganze Tibetan Autonomous Prefecture and Ngawa Tibetan and Qiang Autonomous Prefecture, moving to the east at first, then moving to the west, and finally moving to the east. The center of gravity of higher risk was concentrated in Ganze Tibetan Autonomous Prefecture, moving to the west at first and then moving to the east. The center of gravity of highest risk was predominantly located in Ganze Tibetan Autonomous Prefecture and Chengdu, with a consistent westward trajectory of risk migration. From the perspective of azimuth angle change during the four periods, it can be noted that the lowest-risk area exhibits a ‘southwest–northeast’ pattern. Additionally, the azimuth angle demonstrated a continuous increase, indicating the gradual stabilization of the ‘southwest–northeast’ directional pattern. Conversely, the pattern for the lower-risk area underwent more drastic changes. Initially, it rapidly transitioned from an approximate ‘north-south’ direction to a ‘north–southeast’ direction, followed by a gradual return to an approximate ‘north-south’ pattern. The medium-risk category exhibited changes in two directions, namely, ‘northwest–southeast’ and ‘west–east’; Additionally, the azimuth initially decreased, followed by an increase and subsequently another decrease. The distribution pattern in the ‘northwest–southeast’ direction showed an initial weakening, followed by a strengthening, and finally another weakening, approaching the ‘west–east’ direction pattern. The higher-risk category displayed a distribution pattern with a ‘northwest–southeast’ direction. The azimuth angle gradually increased and then decreased, indicating an initial increase and a subsequent decrease in the ‘northwest–southeast’ pattern. Notably, the highest-risk category underwent a significant pattern change, firstly transitioning from an approximate ‘south–north’ direction to an approximate ‘west–east’ direction and ultimately adhering to the ‘west–east’ pattern.

3.3. ESV-Risk Spatial Autocorrelation Analysis

3.3.1. Global Spatial Autocorrelation

The results showed that when the significance level was p = 0.05, the global Moran’s I of ESV and landscape ecological risk in 2005, 2010, 2015, and 2018 were −0.242, −0.233, −0.277, and −0.254, respectively, indicating that ESV and landscape ecological risk showed a negative correlation in space; that is, the landscape ecological risk was lower in areas with high ESV, and vice versa.

3.3.2. Local Spatial Autocorrelation

As shown in the LISA cluster diagram in Figure 8, the regions with significant local correlations with ESV risk are mainly in Ganze Tibetan Autonomous Prefecture, Northwest Sichuan, Chengdu, Deyang, Meishan, Ziyang, Neijiang, Zigong, etc. On the whole, the spatial distribution of ESV and landscape ecological risk over the four periods was significant, as was the degree of clustering.

3.4. Landscape Ecological Risk Zoning Control and Strategy in Sichuan Province

3.4.1. Division of Zoning Management and Control Types

By utilizing the local spatial autocorrelation results of ESV and landscape ecological risk from 2018, we employed a spatial superposition combination approach to delineate ecological zoning. ESV and risk have five local spatial autocorrelation aggregation types, namely, HH, HL, LH, LL, and NN, which represent the high and low aggregation characteristics of ESV and risk in space. According to the merging principle proposed in, 4 primary-level zonings and 14 secondary-level zonings were divided (

Table 5. Ecological management zoning of Sichuan Province based on local spatial autocorrelation.

First-Level Zoning	Second-Level Zoning	Local Spatial Autocorrelation		Control Strategy	Typical Region
		Risk	ESV
Ecological conservation zone	Core conservation zone	LL	HH	Focus on protection; prohibition of large-scale land development and utilization.	It is mainly distributed in most areas of Ngawa Tibetan and Qiang Autonomous Prefecture, Ganze Tibetan Autonomous Prefecture, Liangshan Prefecture, Ya’an, and Leshan.
	Priority conservation zone	NN	HH	Protect regional species diversity and maintain the stability of the ecosystem.
	Key conservation zone	LL	NN	Improve the ecological carrying capacity and maintain the existing ecological level.
Ecological cultivation zone	Priority cultivation zone	LH	HH	Focus on cultivating high-ESV areas; improve ecosystem service capacity.	It is distributed in the eastern plain of Sichuan, northeast of Sichuan, and southwest of Sichuan in the form of dots.
	Key cultivation zone	HL	HL	Adjust the proportion of land use types. Moderately promote the rationality of land use structure.
	Common cultivation zone	NN	LL	Ensure the adequate governance of low-ESV areas.
Ecological agricultural zone	Priority agricultural zone	LH	NN	Give priority to ecological agriculture; improve the output ratio of regional agricultural input.
	Key agricultural zone	NN	HL	Give full play to the advantages of agriculture and improve the management level of agricultural land.	It is distributed in the plains area of Eastern Sichuan in the form of flakes, mainly for Chengdu, Nanchong, Guang’an, Sunning, Ziyang, Yibin, Luzhou, and so on.
	Common agricultural zone	LL	HL	Promote the rationality of agricultural land structure and tap its potential.
	Moderate agricultural zone	LL	LL	Take into account the diversity of land use; engage in rational planning.
Ecological improvement zone	Priority improvement zone	HH	HL	Ensure habitat patch connectivity and ecological diversity.
	Key improvement zone	HH	LL	Engage in ecological network construction and land remediation	It is distributed in Liangshan Prefecture, Ganze Tibetan Autonomous Prefecture, Ngawa Tibetan and Qiang Autonomous Prefecture, Guangyuan, Bazhong, Dazhou, Yibin, and Zigong in the form of lakes.
	Common improvement zone	HH	NN	Carry out land consolidation; improve environmental conditions.
	Comprehensive improvement zone	NN	NN	Prevent soil loss; improve the land.

); The zonings were visualized to obtain the results shown in Figure 9 and Figure 10.

3.4.2. Strategy of Zoning Management and Control

Based on the findings regarding ecological zoning and considering the spatial distribution characteristics of regional economic development, the internal ecological environment, ecosystem service value, and landscape ecological risk, different control measures were developed for distinct ecological zones. These zones include the ecological conservation zone, the ecological cultivation zone, the ecological agricultural zone, and the ecological improvement zone (Figure 11).

(1)

Ecological conservation zone

Ecological conservation zone, including core conservation zone, priority conservation zone, and key conservation zone.

(a) Characteristics: The ecological conservation zone is predisposed to low landscape ecological risk while demonstrating high ESV. This zone is characterized by abundant natural resources, favorable water conditions, strong ecological diversity, and the presence of nature reserves and forest parks. Furthermore, it remains largely unaffected by urban land expansion.

(b) Strategy: The ecological conservation zone should implement key protection and forest land conservation measures, strictly prevent deforestation, and safeguard species diversity. In the core conservation zone, low-intensity forestry planting and management should be conducted to preserve diverse habitats and effectively serve as barriers to maintain landscape ecological functions. The priority conservation zone requires the protection of key forest land and the rigorous prevention of illegal occupation and excessive logging activities. Additionally, strict actions should be taken against any unlawful behavior that compromises the ecological environment for economic gain. In the key conservation zone, policy guidance and pilot protection measures should be effectively employed. On the one hand, this involves promoting the rational utilization of forest resources in mountainous areas, strengthening forest land protection, and limiting the development of non-productive forest land. On the other hand, collaboration with various departments is essential to actively promote forest protection and sand prevention policies and emphasize the need for green ecological development. Furthermore, the introduction of low-impact cultural tourism resources should be strategically undertaken to enhance the ecosystem’s cultural service function and foster the formation of an economically and ecologically sustainable landscape pattern.

(2)

Ecological cultivation zone

The ecological cultivation zone includes priority cultivation zones, key cultivation zones, and common cultivation zones.

(a) Characteristics: The ecological cultivation zone is associated with low landscape ecological risk or insignificant risk aggregation, coupled with high ESV. It is a relatively intact and symbolic ecological recreation green space with a favorable ecological environment.

(b) Strategy: In the ecological cultivation zone, efforts are focused on promoting habitat optimization, diversifying regional landscape types, enhancing ecosystem services, and gradually establishing a model area for regional landscape ecology and resources. In the priority cultivation zone, the aim is to develop green spaces characterized by abundant vegetation and diverse landscapes and enhance the ecosystem’s capacity to regulate the local climate. In the key cultivation zone, the protection of the ecological environment is emphasized, and the principles of green low-carbon-cycle development are embraced. Efforts in this area emphasize ecological cultivation practices with respect to rivers, shelter forests, wetlands, and meadows, with the objective of improving habitat quality and soil conditions within the region. In the common cultivation zone, the construction of urban parks and greenway networks should be strengthened. This zone provides recreational places for production and living space and enables the development of a green ecological security pattern.

(3)

Ecological agricultural zone

The ecological agricultural zone includes priority agricultural zones, key agricultural zones, common agricultural zones, and moderate agricultural zones.

(a) Characteristics: In the ecological agricultural zone, the landscape ecological risk shows a complex distribution, and the ESV is relatively low. It serves as the primary region for cultivated land distribution in Sichuan province, characterized by flat topography, excellent water quality, and favorable soil conditions. This area is dominated by agricultural cropland ecosystems.

(b) Strategy: In the ecological agricultural zone, agricultural protection and development strategies are implemented, with a particular focus on safeguarding cultivated land integrity and constructing high-standard farmland, and this zone promotes the scaling and modernization of agriculture and industry. This area has become an important source of ecological and economic development. Efforts regarding the priority agricultural zone aim to enhance production efficiency by promoting modern agricultural techniques and technologies, implementing high standards for farmland construction and improving land ecology. A rational configuration of the regional agricultural industry structure should be achieved, with strict regulations prohibiting the conversion of cultivated land for non-food to ensure the preservation of permanent basic farmland. In the key agricultural zone, a rational reduction in the use of pesticides and fertilizers is advocated, accompanied by efforts geared toward enhancing the quality and yield of ecological agricultural products such as crops and livestock. Simultaneously, it is imperative to improve the flood resistance and disaster prevention capabilities of farmland ecosystems, enabling the scaling of agricultural construction and strengthening the function of agricultural production. This approach maximizes the supply capacity of food and raw materials for ecosystem services. In the common agricultural zone, in addition to strengthening the construction of agricultural land infrastructure, it is necessary to realize the effective utilization of cultivated land resources so as to realize the integrated development of ecological agriculture. In the moderate agricultural zone, promoting the conversion of farmland back into forests and grasslands according to the premise of ensuring food security is necessary. In this zone, ecological manors and agricultural picking gardens can also be developed to enhance the cultural service function of the ecosystem.

(4)

Ecological improvement zone

The ecological improvement zone includes priority improvement zones, key improvement zones, common improvement zones, and comprehensive improvement zones.

(a) Characteristics: This area represents a high-value concentration of landscape ecological risk, characterized by low ESV. It is primarily marked by rapid urbanization or consists of plateau bare land with unfavorable ecological conditions, featuring sparse vegetation and a simplistic structure.

(b) Strategy: Within this region, ecological restoration and land consolidation should be undertaken. In the priority improvement zone, initiatives should encompass ecological land restoration, the enhancement of soil fertility, comprehensive ecological land management, and the improvement of the local biological habitat. In the key improvement zone, there is a need to emphasize land consolidation while restraining haphazard urban expansion. Given the low vegetation cover, high population density, frequent human activities, and industrial clusters in this zone, challenges involving the interplay between human activities, land use, and ecological concerns are pronounced. Therefore, prudent land development and utilization in the central urban areas should receive heightened attention. In the common improvement zone, it is imperative to establish a robust system for preventing and controlling pollution sources, coupled with the implementation of land protection and optimization projects. Attention should be paid to land consolidation and improvement efforts, utilizing effective measures for pollution prevention and control to rigorously curtail indiscriminate pollutant discharge, regulate livestock farming, and forestall ecological degradation. In the comprehensive improvement zone, measures should concentrate on enhancing soil fertility and improving soil quality. Vigilance is needed to prevent the excessive use of pesticides and fertilizers, which can lead to soil compaction and fertility loss. Additionally, efforts toward augmenting green areas and comprehensively enhancing the local microclimate should be pursued to elevate the overall environmental quality.

4. Discussion

As a critical ecological barrier in the upper reaches of the Yangtze River, Sichuan Province faces escalating external stress factors, necessitating urgent ecological protection efforts. This province has been segregated into ecological conservation zone, ecological cultivation zone, ecological agricultural zone, and ecological improvement zone, with meticulously implemented control strategies. Strengthening protection measures, enhancing governance, and refining the governance mechanisms are imperative to contribute significantly to the high-quality development of Sichuan Province. From 2005 to 2018, the ESV in Sichuan Province exhibited improvement. However, landscape ecological risks, especially those of medium and high magnitudes, have been on the rise, indicating persistent hidden threats to the ecological security situation in the province. Therefore, it is imperative to bolster monitoring and early-warning systems for ecological risks and develop proactive countermeasures. In general, improvements can be made by taking the following measures: (1) Proactively align with the guidelines and policies of higher-level planning, integrating ecological protection and restoration initiatives into territorial space planning and rigorously implementing zoning control strategies. (2) Utilize natural resources in Sichuan Province to establish scenic spots and protected areas. This not only preserves biodiversity but also fosters economic development within the province.

In this study, the spatial and temporal evolution as well as the spatial interaction of ecosystem service value and landscape ecological risk in Sichuan Province were analyzed. This analysis serves as a foundation for ecological protection, restoration, and the city’s sustainable development to some extent. However, due to challenges regarding data collection and other objective factors, there are limitations to this study: (1) The equivalent factor method was employed to calculate ESV, and for certain types of construction land, the ecosystem service value coefficient was estimated to be 0. In reality, human activities generate substantial pollutants that can impact ecosystem service functions. Hence, exploring a more precise calculation method is imperative. (2) While this study explored the correlation between ESV and landscape ecological risk, it only scratched the surface and did not delve into the coupling and coordination relationship between the two. Therefore, future research could employ diverse methods to comprehensively depict this relationship, providing a basis for more scientifically managing and controlling ecological protection zoning in this region.

5. Conclusions

Adopting an ecosystem services perspective, this study developed and enhanced the landscape ecological risk assessment system. It examined the spatial and temporal evolution and dynamic trends of ecosystem service value and landscape ecological risk in Sichuan Province from 2005 to 2018. Additionally, it analyzed their spatial interaction in Sichuan Province using the spatial autocorrelation model. Based on these analyses, Sichuan Province was categorized into four ecological management and control zones, leading to the following conclusions:

(1)

From 2005 to 2018, the value of ecosystem services in Sichuan Province exhibited a fluctuating upward trend, with forest land and grassland being the primary contributors, accounting for about 90% of Sichuan Province’s ecosystem service value. These two elements significantly influenced the balance of ecosystem services in the region. Spatially, there was a distinctive pattern of higher values in the west and lower values in the east. The low-value area in the eastern part of Sichuan Province took on a ‘C’ shape and extended northwards.

(2)

From 2005 to 2018, the low- and lower-risk areas in Sichuan Province consistently decreased over time, while high- and higher-risk areas increased, indicating a stable trend. Spatially, there was a continuous rise in overall risk, with a discernible pattern of high-level risk shifting from west to east as time progressed.

(3)

From 2005 to 2018, there was a notable negative correlation between ecosystem service value and landscape ecological risk in Sichuan Province. Areas with high ESV experienced lower landscape ecological risk, and vice versa. A significant correlation was observed in the local spatial autocorrelation, particularly concentrated in Ganzi Prefecture in Western Sichuan as well as in Chengdu, Deyang, Meishan, and Ziyang in eastern Sichuan. The distribution across the four periods exhibited clear patterns with significant clustering characteristics.

(4)

Based on the outcomes of the spatial autocorrelation analysis of ecosystem service value and landscape ecological risk in Sichuan Province, the ecological management and control zones for the province have been delineated: In the ecological conservation zone, stringent protection measures for the natural environment are imperative, and a commitment to low-impact development principles is crucial. In the ecological cultivation zone, a moderate adjustment of land use types is necessary. This involves promoting the rationality of land use structures and enhancing the region’s ecosystem service capacity. In the ecological agricultural zone, leveraging agricultural advantages is essential. This includes the development of ecological farms and modern agricultural practices alongside an elevation of agricultural land management standards. In the ecological improvement zone, the focus should be on constructing ecological networks and nodes. Emphasizing ecological restoration and land remediation efforts is of paramount importance.