Global Versus Local? A Study on the Synergistic Relationship of Ecosystem Service Trade-Offs from Multiple Perspectives Based on Ecological Restoration Zoning of National Land Space—A Case Study of Liaoning Province

Abstract

Clarifying the trade-offs and synergies of ecosystem services in Liaoning’s ecological restoration zones is crucial for strengthening the positioning of ecological restoration zones and optimizing ecosystem services. This study is based on “Liaoning Provincial Land Spatial Planning (2021–2035)” and divides the area into ecological restoration zones. We utilized the InVEST model, ArcGIS Pro, and Geoda in this study to quantify five ecosystem services (Soil Conservation, Carbon Storage, Habitat Quality, Water Yield, and Food Production) and constructed an evaluation framework to assess the trade-offs and synergies of ecosystem services at both global and local levels. The conclusions are as follows: (1) The global relationships among ecosystem services in different ecological restoration zones are ranked as: strong trade-offs (35.51%) > weak trade-offs (33.17%) > low synergies (29.09%) > high synergies (2.24%); (2) The area exhibiting synergistic relationships between pairs of local ecosystem services in ecological restoration zones is larger than the area exhibiting trade-offs; (3) The strongest synergy is observed between water yield and soil conservation, while the most significant trade-off occurs between food production and soil conservation. These relationships exhibit similar spatial characteristics in the WSFR, SWCR, and WCR zones; (4) The proportion of areas showing trade-offs and synergies differs between global and local scales.

1. Introduction

Ecosystem services comprise the range of benefits provided by ecosystem processes and functionalities that humans value, both directly and indirectly, and that are crucial for well-being [1,2,3,4]. Costanza et al. highlighted these services’ dual environmental and economic contributions, advancing the methods for their economic evaluation [5,6]. Ecosystems, characterized by their inherent environmental and socio-economic features [7], operate not in isolation but as interconnected systems. Thus, the interplay among various ecosystem services is complex [8,9]. Analyzing how these services relate is key to improving ecosystem functionality [10,11]. The dynamics of these interactions, specifically the compensations and mutual enhancements among services, hold particular importance [12,13,14]. As noted in the Millennium Ecosystem Assessment [15], a compensatory dynamic is evident when the enhancement of one service diminishes another [16]. Conversely, mutual enhancements occur when the improvement of one service coincides with the elevation of others [17]. In detail, a compensation happens when the increase in one service detrimentally affects another, whereas a mutual enhancement is observed when several services amplify or decline in unison [18]. The overarching aim of managing ecosystems is to foster simultaneous advancements and mutual enhancements in ecosystem services, navigating the challenges posed by the variety of service types, their functional intricacies, spatial variability, and uneven distribution [19,20,21]. Therefore, the interactions among ecosystem services manifest as spatially nonlinear and evolve dynamically over time [22,23].

In recent years, examining the trade-offs and synergies among ecosystem services has emerged as a key research area within the fields of ecology, geography, forestry, and biology. For global assessments and simulations of ecosystem services, Costanza et al. (1997) evaluated the economic value of global ecosystem services [6]. Focusing on ecosystem services at the national scale, Ouyang et al. conducted an assessment of China’s ecosystem services [24]. At the regional scale, researchers pay more attention to specific ecological regions or watersheds, such as the assessment of ecosystem services in the Yangtze River Basin [25]. Local-scale studies typically concentrate on ecosystem services in urban or rural areas, such as the study by Bolund and Hunhammar, who evaluated the ecosystem services provided by urban green spaces [26]. The research by de Groot et al. targets specific ecosystems, such as the assessment of ecosystem services in forests, wetlands, or grasslands [27]. Scholars have utilized a variety of methods, including spatial overlay analysis [28,29], comparative analysis [30], root mean square error [31], pixel-wise correlation analysis [32], and scenario simulation [33,34], to explore the extensive trade-offs and synergies across ecosystem services, assess the variability in spatial interactions, and generate significant insights. Despite the utility of these methods, they present certain challenges. Spatial overlay techniques often fail to discern the subtle distinctions between stable, beneficial, and adverse relationships [35]. Comparative methods may neglect the spatial dynamics between ecosystem services, occasionally resulting in diminished or contradictory trade-offs and synergies. Root mean square error (RMSE) is unable to reflect the cooperative interactions among ecosystem services, and pixel-wise correlation analysis does not consider spatial dependencies among variables, among other issues. Additionally, most studies have concentrated on the spatial variability of trade-offs and synergies either broadly across ecosystem services or more narrowly between specific service pairs [36,37,38]. Despite this, there is a scarcity of holistic research that contemplates the spatial variances in trade-offs and synergies encompassing all ecosystem services at both the global and local levels. To overcome these shortcomings, this research incorporates the local bivariate Moran’s I index with spatial overlay analysis [39,40]. This method emphasizes spatial relationships among ecosystem services, enhancing both the breadth and depth of the analysis by not only focusing on the global impacts of all ecosystem services but also exploring the trade-offs and synergies between pairs of local ecosystem services.

Liaoning Province features diverse topography and climate, with a rich variety of ecosystem types. It is strategically located, with Hebei to the west, Inner Mongolia to the north, and Jilin to the east, making its geographical position extremely important in the northeastern region of China. In line with the implementation principles of the “Liaoning Provincial Land Spatial Planning (2021–2035)”, and considering documents such as the “Liaoning Provincial Main Functional Zone Planning”, the “Liaoning Provincial Land Spatial Ecological Restoration Planning (2021–2035)”, and the “14th Five-Year Plan for Ecological and Environmental Protection of Liaoning Province”, the province has been divided into five ecological restoration zones based on the land-sea ecological security pattern [41]. Soil and Water Conservation Restoration Area in the Hills of Western Liaoning (SWCR): Located in the low hills and mountains region of western Liaoning, we are focusing on advancing soil and water conservation and urban-rural collaborative governance in the western part of the Yiwu Mountains, as well as vegetation restoration and soil erosion prevention in the Songling-Heishan area. This area focusses on soil and water conservation in the hilly regions of western Liaoning. Windbreak and Sand Fixation Restoration Area in Northwest Liaoning (WSFR): In the western part of Liaoning Province, we are carrying out vegetation restoration and desertification control in the southern part of the Horqin Sandy Land, as well as vegetation restoration and desertification prevention in the Nuluerhulu Mountains. This zone aims to address wind erosion and sand fixation in northwest Liaoning. Comprehensive Restoration Area of the Central Plain of Liaoning (CR): The region includes the Shenyang metropolitan area and Anshan, Benxi, Liaoyang, and other areas, which conduct comprehensive management of mountains, waters, forests, fields, lakes, grasses, and sands. This area is designated for a broad range of ecological restoration activities in the central plains of Liaoning. Water Conservation Restoration Area of the Mountainous Areas of Eastern Liaodong (WCR): Strengthen the conservation and protection of the forest ecological barrier in the eastern mountainous areas of Liaoning Province, the southern part of the Changbai Mountains, and the eastern forest regions of the Qianshan Mountain range. This zone emphasizes water conservation in the mountainous regions of eastern Liaoning. Integrated Restoration Areas in the Yellow Sea and Bohai Sea (IR): Promote the ecological restoration of the coastal areas and adjacent sea regions around the Yellow Sea and Bohai Sea in Liaoning Province, including the coastal areas of Suizhong-Xingcheng, the mouth of the Liao River, and its adjacent sea areas, as well as the coastal areas of Dandong-Zhuanghe and the mouth of the Yalu River and its adjacent sea areas. This area focusses on the integrated restoration of land and sea ecosystems in the Yellow Sea and Bohai Sea regions.

To evaluate the current ecological restoration zoning in Liaoning Province, it is essential to analyze the spatial heterogeneity of trade-offs and synergies in ecosystem services. Resolving the challenges of ecosystem service trade-offs and synergies from both global and local perspectives is a pressing issue that demands immediate attention.

The InVEST 3.11.0 model was employed to quantify five key ecosystem services. Utilizing ArcGIS Pro 2.5.1, hotspot analysis, GeoDa 1.22, and the local bivariate Moran’s I index, the study assessed the trade-offs, synergies, and spatial heterogeneity of these services at both global and local scales. Specifically, five representative ecosystem services were selected for analysis: food production (FD), carbon sequestration (CS), water yield (WY), soil conservation (SC), and habitat quality (HQ). Spatial overlay analysis was conducted to investigate the spatial heterogeneity of trade-offs and synergies among these ecosystem services across different ecological restoration zones in Liaoning Province. Additionally, local spatial autocorrelation analysis was used to construct a framework for examining the interactions between global and local ecosystem services within various restoration zones. The overarching aim of this research is to synergistically optimize ecosystem services, thereby supporting major national projects focused on ecological security barrier protection and restoration while simultaneously enhancing the overall functionality of key ecosystem services in Liaoning Province.

2. Materials and Methods

2.1. Study Area Overview

Located in the northeastern segment of China, Liaoning Province spans from 38°43′ N to 43°26′ N latitude and 118°53′ E to 125°46′ E longitude. The province governs two sub-provincial cities—Shenyang and Dalian—along with twelve other prefecture-level cities: Anshan, Fushun, Jinzhou, Panjin, Yingkou, Huludao, Tieling, Liaoyang, Benxi, Dandong, Chaoyang, and Fuxin. Characterized by a temperate continental monsoon climate, Liaoning features a diverse and complex topography. Its coastal areas are rich in marine resources and exhibit unique coastal landforms, enhancing the geographical diversity of the region (Figure 1).

2.2. Data Sources

The foundational data used in this study includes: (1) Land Use Data, covering Liaoning Province with a spatial resolution of 30 m, obtained from the Chinese Academy of Sciences Resource and Environment Data Center (http://www.resdc.cn, accessed on 1 June 2024); (2) Meteorological Data, including precipitation and potential evapotranspiration for Liaoning Province, sourced from the National Qinghai-Tibet Plateau Science Data Center (https://data.tpdc.ac.cn, accessed on 6 June 2024); (3) Soil Database: Soil data for Liaoning Province are derived from the Chinese soil dataset based on the World Soil Database (HWSD); (4) GDP Raster Data: Liaoning Province GDP data are obtained from the Chinese Academy of Sciences Resource and Environment Data Center (http://www.resdc.cn, accessed on 15 June 2024); (5) Population Density Data, sourced from the WorldPop population density dataset (https://www.worldpop.org, accessed on 28 June 2024) with a spatial resolution of 100 m; (6) Statistical Data, sourced from the Liaoning Provincial Statistical Yearbook (https://tjj.ln.gov.cn, accessed on 1 July 2024) and The Liaoning Province Water Resources Bulletin (https://slt.ln.gov.cn, accessed on 10 July 2024); (7) NDVI Data, calculated from the 16-day composite product of the MODIS13Q1 dataset provided by NASA, with a spatial resolution of 250 m; (8) Digital Elevation Model (DEM) Data: DEM data for Liaoning Province are obtained from the Geospatial Data Cloud (http://www.gscloud.cn, accessed on 15 July 2024). All data are from the year 2020, and the spatial resolution has been uniformly adjusted to 100 × 100 m for analysis. The research analysis flowchart is as follows (Figure 2):

2.3. Ecosystem Service Assessment Methods

2.3.1. Food Production

This study utilizes grain yield data from Liaoning Province to illustrate the role of food production within ecosystem services. Existing research has demonstrated a significant linear relationship between regional grain yield and NDVI [42,43]. Therefore, this study integrates grain yield data from the Liaoning Statistical Yearbook with NDVI data to achieve spatial quantification of food production services [42,43].

$$C{P}_a=\left({\displaystyle \frac{NDV{I}_a}{NDV{I}_{sum}}}\right)\times P_{sum}$$

(1)

In the equation: CP_a is the crop yield per raster a, P_sum is the aggregate crop yield across the study area, NDVI_a is the NDVI for raster a, and NDVI_sum is the comprehensive NDVI for the entire study area.

2.3.2. Habitat Quality

In this analysis, biodiversity is assessed using the InVEST model’s habitat quality module. This index quantifies habitat quality by evaluating habitat type sensitivity and external threat factors [42]. Initially, the primary stress factors that could potentially impact habitat quality within the study area were identified. For each stress factor, a maximum influence distance was established, representing the range over which the stress factor affects habitat quality. A weight was assigned to each stress factor to reflect its relative importance in influencing habitat quality. The type of decay for each stress factor was selected, with linear decay indicating that the impact of the stress factor decreases linearly with distance while exponential decay signifies that the impact diminishes exponentially with distance. The habitat suitability corresponding to different land use types was determined, as was the sensitivity of different land use types to each stress factor. These parameters were then input into the model’s formula to calculate habitat quality. The quality of habitat is thus derived:

$$Q_{\left(ab\right)}=H_b\times (1-{\displaystyle \frac{{D_{\left(ab\right)}}^2}{{D_{\left(ab\right)}}^2+k^2}})$$

(2)

Here, Q_(ab) measures the habitat quality for raster ‘a’ within land use type b in Liaoning Province; H_b reflects the habitat suitability for raster ‘a’ in land use type b in Liaoning Province; D_(ab) describes the extent of habitat degradation for raster ‘a’ in land use type b in Liaoning Province, and k is the half-saturation constant. The habitat quality score ranges from 0 to 1, with higher scores indicating better quality and less degradation. Parameters are derived from prior studies.

$$D_{xi}={\sum }_{r=1}^R{\sum }_{y=1}^{Y_r}\left({\displaystyle \frac{W_r}{{\sum }_{r=1}^R{W}_r}}\right)r_y{i}_{xy}{\beta }_x{S}_{jr}$$

(3)

Here, R represents the number of stress factors; W_r denotes the weight of each stress factor; Y_r is the number of grid cells for each stress factor; r_y stands for the number of stress factors on a grid cell; β_x indicates the accessibility level of grid cell x; S_jr represents the sensitivity of landscape j to the stress factors; and i_xy is the influence distance of the stress factors.

2.3.3. Carbon Storage

This study utilizes the carbon sequestration services of carbon stocks in Liaoning Province for the year 2020. Carbon stocks are represented by four fundamental carbon pools. The InVEST (Integrated Valuation of Ecosystem Services and Trade-offs) model is instrumental in assessing the energetic and economic value of ecosystem services across terrestrial, freshwater, and marine ecosystems. It quantifies carbon storage based on four critical carbon reservoirs: aboveground biomass, belowground biomass, soil organic carbon, and dead organic matter. These reservoirs are fundamental to ecosystem carbon cycling and play a significant role in global climate dynamics. Carbon density classification is predicated on field measurements, scholarly literature, or pre-existing datasets, with a preference for locally sourced data to enhance accuracy. Adjustments for temperature and precipitation are incorporated to account for environmental variability. The model estimates current carbon stocks by integrating land use and land cover data with carbon value tables for the four carbon pools. It categorizes carbon storage into aboveground biomass, belowground biomass, soil carbon, and dead organic carbon, evaluating their respective contributions to ecosystem services. By integrating land use data with corresponding carbon density classifications, we calculated the carbon stocks for these four pools and subsequently determined the overall carbon stock for Liaoning Province. The carbon density data for various land types were adjusted based on existing research in the study area and the specific conditions of the region. The formula for calculating carbon stocks is as follows:

$$C_t=C_a+C_b+C_s+C_d$$

(4)

In the equation: C_t represents the total carbon stock; C_a denotes the aboveground biomass carbon pool; C_b indicates the belowground biomass carbon pool; C_s refers to the soil carbon pool; and C_d stands for the dead organic matter carbon pool [44].

2.3.4. Soil Conservation

In this study, soil erosion control services within the ecosystem services of Liaoning Province are represented by soil retention quantity. This study quantifies soil conservation in ecosystem services using the RULSE equation, where soil conservation is assessed by the difference between potential soil erosion and actual soil erosion. Soil conservation is the actual soil erosion per unit area. By calculating the difference between potential soil erosion (potential erosion) and actual soil erosion (actual erosion), the amount of soil conservation (asoil conservation) can be obtained. The calculation of soil retention employs the Revised Universal Soil Loss Equation (RUSLE), which is widely used and scientifically validated by most researchers [45]. This approach quantifies soil erosion control services by calculating the difference between the potential soil erosion (assuming no vegetation cover, i.e., vegetation cover factor C = 1 and soil conservation factor P = 1) and the actual soil erosion. The difference is considered the soil retention quantity for the area. The formula for this calculation is as follows [46,47,48]:

$$SC=R\times K\times LS\times (1-C\times P)$$

(5)

In the equation: SC represents the soil conservation amount; R is the rainfall erosivity factor [49]; K is the soil erodibility factor [50]; C is the vegetation cover factor (dimensionless) [51]; P is the conservation practice factor (dimensionless) [51]; and LS is the slope length and steepness factor (dimensionless) [52]. The detailed calculation formulas for these factors can be found in the referenced literature.

The empirical formula for calculating the rainfall erosivity factor R using various rainfall amounts is as follows:

$$R={\sum }_{i=1}^{12}1.735\times {10}^{\left(1.5\times \mathit{lg}{\displaystyle \frac{{pi}^2}{p}}\right)-0.8188}$$

(6)

In the equation: R represents the rainfall erosivity factor for a grid cell; i denotes the month; pi is the rainfall amount for a specific month; and p is the annual rainfall amount.

The soil erodibility factor K is calculated using the formula developed in the EPIC model.

$$\begin{array}{c}K=0.2+0.3\mathit{exp}[-0.0256Sa(1-{\displaystyle \frac{Si}{100}}\left)\right]\times {\left({\displaystyle \frac{Si}{Cl+Si}}\right)}^{0.3}\\ \times [1-{\displaystyle \frac{0.25SOC}{SOC+\mathit{exp}(3.72-2.95SOC)}}]\\ \times [1-{\displaystyle \frac{0.7Sn}{Sn+\mathit{exp}(-5.51+22.9Sn)}}]\end{array}$$

(7)

In this formula, K represents the soil erodibility factor; Sn is defined as 1 s the soil sand content (Sa) divided by 100; Sa denotes the soil sand content; Cl is the soil clay content; and SOC stands for the soil organic carbon content.

2.3.5. Water Yield

Water yield is defined as an ecosystem’s ability to intercept rainfall and store freshwater resources [51]. This study quantifies the water yield service in Liaoning Province using the water yield module (Water yield) of the InVEST model.

$$Y_a=(1-{\displaystyle \frac{AE{T}_a}{P_a}})\times P_a$$

(8)

$$D_{wy}=D_{pp}\times {\rho }_a$$

(9)

In the equation: Y_a represents the annual water yield for raster a; AET_a is the annual evapotranspiration for raster a; and P_a is the annual precipitation for raster a. To standardize the units of water yield, service supply, and demand, the raster area was multiplied by Y_a to obtain the water yield volume. D_wy represents the water yield service demand for raster a; D_pp is the per capita water consumption; and ρ_a is the population density.

2.4. Analysis of Ecosystem Service Trade-Offs and Synergies

2.4.1. Global Ecosystem Service Trade-Offs and Synergies

This study employs spatial overlay analysis to identify the types and regions of ecosystem service trade-offs or synergies at a global scale. From a global perspective, the spatial overlay method allows for a clear visualization of spatial differentiation in trade-offs or synergies between multiple services. This approach helps in effectively implementing trade-off management decisions for various services in specific spatial areas. The specific calculation method employed is as follows:

• Ecosystem Service Capacity Classification

Because there are different units of measurement for various types of ecosystem services, it is impossible to directly relate or compare them on the same scale without a standardized approach. Therefore, we first normalize each service by comparing the ecosystem service value of each raster unit to its maximum value within the study area. Then, using the natural breaks method, the capacity of each service is classified into three levels, low, medium, and high, which are assigned the corresponding codes 1, 2, and 3 (

Table 1. Classification of ecosystem services capacity.

Ecosystem Services Type	Low (1)	Medium (2)	High (3)
Carbon Sequestration	0–0.137	0.137–0.537	0.537–1
Food Production	0–0.153	0.153–0.275	0.275–1
Water Yield	0–0.275	0.275–0.478	0.478–1
Soil Conservation	0–0.059	0.059–0.227	0.227–1
Habitat Quality	0–0.392	0.392–0.710	0.710–1

).

2.

Ecosystem Service Spatial Overlay

Since different types of ecosystem services have different units, conducting associations and comparisons on the same scale is not possible. Therefore, the first step is to standardize each service. This article uses spatial overlay analysis to analyze and display the trade-offs and synergies between different ecological zones and types of ecosystem services within the study area. This method can clearly reflect the spatial distribution characteristics of trade-offs/synergies between different ecosystem services [53]. The specific operation method involves the following steps: To begin, compare the ecosystem service value of each grid cell to the highest value of that service in the study area. Then, divide the data into three levels, which are low, medium, and high, represented by the numbers 1, 2, and 3. Next, use the ArcGIS Pro 2.5.1 software to produce a spatial overlay on the classified ecosystem service grid data to see how the different levels of ecosystem services are spread out in space.

After standardizing and classifying the raster data for the five ecosystem services, a spatial overlay analysis is performed. The rules for this analysis are as follows:

$$CODE=FD\times \mathrm{10,000}+CS\times 1000+SC\times 100+WY\times 10+HQ$$

(10)

In the equation: FD represents food production services; CS represents carbon sequestration services; SC represents soil conservation services; WY represents water yield services; and HQ represents habitat quality services.

3.

Classification Standards for Service Trade-offs and Synergies

Based on the overlay results of the five ecosystem services and the definitions of trade-offs and synergies [54,55], a classification standard for trade-offs/synergies is established (

Table 2. Classification criteria and statistics of trade-offs/synergies among ecosystem services.

Ecosystem Service Relationships	Proportion of Study Area	Subclass	Proportion of Study Area	Type	Example
Trade-off	68.67%	Strong Trade-off	35.50%	High (1) Medium (0) Low (4)	11,311 11,113
				High (1) medium (1) Low (3)	11,321 12,113
				High (1) Medium (2) Low (2)	12,312 12,321
				High (1) Medium (3) Low (1)	22,312 32,212
		Weak Trade-off	33.17%	High (2) Medium (0) Low (3)	11,133 11,313
				High (2) Medium (1) Low (2)	23,113 31,123
				High (2) Medium (2) Low (1)	23,213 23,123
				High (3) Medium (0) Low (2)	33,113 13,313
				High (3) Medium (1) Low (1)	33,213 23,313
				High (4) Medium (0) Low (1)	33,313 33,133
Synergy	31.33%	High Synergy	2.24%	High (5) Medium (0) Low (0)	33,333
				High (4) Medium (1) Low (0)	33,233 23,333
				High (3) Medium (2) Low (0)	33,223 32,332
				High (2) Medium (3) Low (0)	22,332 32,322
				High (1) Medium (4) Low (0)	22,322 22,232
				High (0) Medium (5) Low (0)	22,222
		Low Synergy	29.09%	High (0) Medium (1) Low (4)	11,211 12,111
				High (0) Medium (2) Low (3)	12,112 12,211
				High (0) Medium (3) Low (2)	12,212 22,112
				High (0) Medium (4) Low (1)	22,212 22,122
				High (0) Medium (0) Low (5)	11,111

). Strong trade-offs refer to a state where one service has a high supply capacity while the other services have medium-to-low supply capacities. Weak trade-offs refer to a state where two, three, or four services have high supply capacities while the others are medium-to-low. Synergies are divided into high synergies and low synergies. High synergies mean that all five services have high and medium supply capacities, which is the most harmonious state and the ultimate goal of ecosystem management. Low synergies mean that all five services are medium-to-low, which is the least desirable state.

2.4.2. Quantification of Local Ecosystem Service Trade-Offs and Synergies

• Ecosystem Service Cold and Hot Spot Analysis

Cold spots and hot spots refer to regions where the distribution of ecosystem service values is either significantly dispersed (cold spots) or highly concentrated (hot spots). A hot spot is identified when a given service exhibits high values that are spatially surrounded by similarly high values. In spatial statistics, the G_i^* statistic, introduced by Getis and Ord [56], is widely utilized as a local spatial autocorrelation measure. This statistic, based on a complete distance matrix, is employed to examine the degree of spatial clustering for high or low values of ecosystem services within ecological restoration zones. The calculation formula is as follows:

$$G_i^{*}={\displaystyle \frac{{\sum }_j^n{W}_{ij}{x}_j}{\sum _j^n{x}_j}}$$

(11)

The G_i^* test yields the following Z(G_i^*) values [56]:

$$Z{(G}_i^{*})={\displaystyle \frac{G_i^{*}-E{(G}_i^{*})}{\sqrt{VAR\left(G_i^{*}\right)}}}={\displaystyle \frac{\sum _j{W_{ij}x}_j}{\sqrt[s]{\frac{n\sum _j^n{W}_{ij}^2-{\left(\sum _j^n{W}_{ij}\right)}^2}{n-1}}}}$$

(12)

$$X={\displaystyle \frac{\sum _j^n{x}_j}{n}},S=\sqrt{{\displaystyle \frac{\sum _j^n{x}_j^2}{n}}-X^2}$$

(13)

In the equation: w_ij is the spatial weight matrix; x_j is the attribute value; x is the mean of all attribute values; and n is the number of patches. The classification based on Z(G_i^*) is as follows: Z(G_i^*) > 2.58: Extreme hot spot; 1.96 < Z(G_i^*) ≤ 2.58: Hot spot; 1.65 < Z(G_i^*) ≤ 1.96: Sub-hot spot; Z(G_i^*) < −2.58: Extreme cold spot; −2.58 ≤ Z(G_i^*) < −1.96: Cold spot; −1.95 ≤ Z(G_i^*) ≤ −1.65: Sub-cold spot; −1.65 ≤ Z(G_i^*) ≤ 1.65: Non-significant area.

2

Pearson Correlation Analysis

This study utilizes the Pearson correlation coefficient method to analyze the trade-offs and synergies between local ecosystem services. Based on the significance test, a positive correlation coefficient indicates synergy, while a negative correlation coefficient reflects a trade-off relationship.

$$R_{xy}={\displaystyle \frac{{\sum }_i^n(x_i-x_i)(y_i-y_i)}{\sqrt{({\sum }_i^n{(x_i-x_i)}^2}\sqrt{({\sum }_i^n{(y_i-y_i)}^2}}}$$

(14)

In the formula: R_xy is defined as the correlation coefficient; the sample count is denoted by n; x_i and y_i are the individual values of variables x and y at the i-th position, respectively; and x and y indicate the average values of variables x and y, respectively.

3.

Spatial Differentiation Analysis Method (GeoDa)

The analysis of cold and hot spots typically provides insights into the local spatial autocorrelation of ecosystem services, pinpointing clusters where services have similar characteristics. However, this approach does not effectively highlight areas with significant differences in attributes on a local scale. The spatial differentiation analysis method addresses this gap. Using the bivariate local Moran’s I, this study delves into the local interrelationships among ecosystem services within the study area. To assess the dynamics of trade-offs and synergies among pairs of ecosystem services at the regional level in Liaoning Province, this research quantifies the values for five key ecosystem services and utilizes the GeoDa 1.22 software for a detailed bivariate spatial autocorrelation analysis in 2020. During this process, spatial formations such as high-low and low-high clusters are indicative of trade-offs between two specific services, whereas high-high and low-low clusters are representative of synergistic interactions among them.

3. Results

3.1. Differences in Ecosystem Service Capacities and Spatial Differentiation Characteristics Across Ecological Restoration Zones in Liaoning Province

In 2020, the total amounts of carbon sequestration, soil conservation, food production, and water yield in Liaoning Province were 2.09 × 10⁹ t, 2.44 × 10⁹ t, 5.22 × 10⁹ t, and 2.33 × 10⁷ t, respectively. Analyzing the average service capacities per unit area across the five ecological restoration zones, the WCR exhibited the highest capacities for carbon sequestration, soil conservation, water yield, and habitat quality. Conversely, the WSFR demonstrated the weakest soil conservation capacity, with an average value of 20.77 t/hm². The CR had the lowest capacities for habitat quality and carbon sequestration, while the IR had the lowest food production capacity. The SWCR exhibited the weakest water yield capacity.

The contribution of ecosystem services across different ecological restoration zones exhibits distinct spatial distribution patterns (Figure 3). The spatial proportions of water yield (WY), carbon sequestration (CS), habitat quality (HQ), and soil conservation (SC) are notably similar across the five zones (Figure 4). Among these, the WCR stands out, showing the highest contribution across all five services, with soil conservation accounting for 72.24%—the largest share. This can be attributed to the region’s dense vegetation cover, which enhances its soil conservation capacity. In contrast, the WSFR records the lowest contribution for soil conservation, habitat quality, and water yield, with soil conservation contributing only 1.61% and both habitat quality and water yield registering values below 10% (

Table 3. Ecosystem services of different ecological restoration zones in Liaoning province.

Ecosystem Services Type	Area	Per Area Service Capability			Proportion of Liaoning Province
		MIN	MEAN	MAX
Soil Conservation (t hm−2 a−1)	IR	0	100.35	3534.84	12.15%
	WSFR	0	20.77	809.62	1.61%
	CR	0	42.55	2564.03	3.43%
	WCR	0	379.33	5911.32	72.24%
	SWCR	0	88.13	1971.16	10.57%
Water Yield (t hm−2 a−1)	IR	0	376.24	1043.78	21.56%
	WSFR	0	222.55	531.66	8.16%
	CR	0	314.57	848.62	11.84%
	WCR	0	520.52	1090.59	46.35%
	SWCR	0	213.78	538.11	12.09%
Food Production (t hm−2 a−1)	IR	3.35	123.67	664.27	16.09%
	WSFR	9.71	257.89	419.85	21.02%
	CR	12.07	240.18	954.00	20.18%
	WCR	12.94	126.14	362.70	25.15%
	SWCR	3.51	139.42	394.70	17.56%
carbon sequestration (t hm−2)	IR	0	127.98	215.84	19.34%
	WSFR	0	111.22	215.84	10.24%
	CR	0	101.20	215.84	9.53%
	WCR	0	177.25	215.84	39.64%
	SWCR	0	149.74	215.84	21.25%
Habitat Quality	IR	0	0.41	1	18.23%
	WSFR	0	0.39	0.97	9.36%
	CR	0	0.32	0.97	8.77%
	WCR	0	0.66	0.99	43.13%
	SWCR	0	0.50	0.87	20.51%

). These low figures are likely due to WSFR’s proximity to the Horqin Sandy Land, where significant wind erosion diminishes the effectiveness of soil and water retention relative to other restoration zones. In the CR, the lowest contributions for carbon sequestration and habitat quality can be attributed to the region’s high degree of urbanization. The conversion of ecological land into construction land and more frequent human activities in the central area further reduce the effectiveness of ecosystem services.

3.2. Analysis of Global Ecosystem Service Trade-Offs and Synergies

In Liaoning Province, the interactions between various ecosystem services are predominantly marked by strong trade-offs, weak trade-offs, high synergies, and low synergies (Figure 5). The distribution of these relationships, ranked by proportion, is as follows: strong trade-offs (35.51%) > weak trade-offs (33.17%) > low synergies (29.09%) > high synergies (2.24%). The spatial distribution of global ecosystem service trade-offs and synergies demonstrates a clear pattern (Figure 6). With the exception of areas characterized by low synergies, the WCR exhibits the highest proportion of regions with the other three types of relationships, making it the most dominant among the five ecological restoration zones. In the five ecological restoration zones, low synergy relationships are primarily concentrated in the SWCR and the IR, with 33.93% of the low synergy areas located in SWCR and 27.73% in IR. Together, these two zones account for over half of the total low synergy areas. The proportion of low synergy areas in the CR and the WFSR is similar. High synergy areas are predominantly located in the WCR, which accounts for 68.15% of the high synergy regions, indicating a prominent presence of high synergy relationships in this zone. Following WCR, the IR and CR also exhibit significant high synergy, with the combined proportion of these three zones exceeding 90% of the total high synergy areas. In contrast, the high synergy regions in SWCR and WFSR are nearly nonexistent. Similarly, weak trade-off relationships are mainly concentrated in WCR, accounting for 59.79% of such regions, with the remaining weak trade-off areas distributed in SWCR and IR. Strong trade-off relationships, however, exhibit a more uniform spatial distribution across the five ecological restoration zones without any significant regional concentration, indicating a more balanced presence across the study area.

According to spatial characteristics, the distribution of the four types of relationships exhibits significant spatial heterogeneity. Low synergy relationships are widely distributed across SWCR, IR, and the southwestern part of WSFR, as well as the southern portion of CR, with a scattered distribution throughout the entire study area. High synergy relationships are only present in the southern part of WCR, near the boundary with IR, and in the northeastern part of WCR, with almost no occurrence in other regions. In contrast to synergy relationships, trade-off relationships display a more widespread global distribution. Weak trade-off relationships are primarily located in the eastern part of WCR, with a spatial pattern opposite to that of low synergy relationships. Strong trade-off relationships are concentrated in the northern part of WSFR, the northwestern region of WCR, and the northern boundary area of CR.

3.3. Analysis of Local Ecosystem Service Trade-Offs and Synergies

3.3.1. Spatial Cold and Hot Spot Analysis of Ecosystem Services

This segment builds on the initial examination of global trade-offs and synergies within ecosystem services by delving deeper into their spatial clustering within ecological restoration zones through cold and hot spot analysis. This method evaluates the local spatial autocorrelation properties of each service, as depicted in Figure 7. The findings reveal that the eastern region of the study area, particularly within the WCR, is predominantly characterized by hot spots for most services, while the western regions mainly feature cold spots. Except for food production services, the four other ecosystem services demonstrate a consistent spatial arrangement of hot spots across all restoration zones. These areas typically form a D-shaped cluster within the WCR, reflecting its role as a key area for water conservation.

Figure 7 illustrates the general consistency in the distribution of cold and hot spots across carbon sequestration and habitat quality services alongside the spatial patterns of soil conservation and water yield services, which display a high degree of similarity. Notably, the hot spots for food production are primarily situated in the expansive WSFR and its southeastern stretches, reaching into neighboring ecological zones. Conversely, significant cold spots for food production appear chiefly in the southwestern areas of IR and SWCR, and in the southeastern parts of WCR, surrounding the hot spots and creating a core-like pattern. The cold spots for water yield and soil conservation are predominantly found in WSFR, SWCR, and the northwestern sections of IR, marked by distinct boundaries. Soil conservation hot spots are prevalent throughout WCR and to a lesser extent, form narrow strips along the northeastern edges of IR. Meanwhile, water yield hot spots are broadly spread across IR. The cold spots for habitat quality and carbon sequestration services tend to gather in the southwestern regions of IR and SWCR, with less frequent occurrences elsewhere.

3.3.2. Ecosystem Service Trade-Offs and Synergies

A comprehensive Pearson correlation analysis was conducted on the five ecosystem services within the five ecological restoration zones, and the outcomes are depicted in (Figure 8). The correlations range from 0 to 0.8, highlighting substantial differences among the services. Predominantly, food production exhibits various degrees of trade-offs when compared to the other four services, while those services generally tend towards synergistic interactions. More specifically, the correlation metrics between food production and the quartet of other services vary between 0.2 and 0.6. The most pronounced trade-off appears between food production and water yield, marked by a correlation of 0.51. Conversely, the most minimal trade-off shows between food production and habitat quality, recorded at 0.28. Among the synergistic interactions, habitat quality and carbon storage present the strongest link, with a correlation coefficient of 0.86, the peak value observed. This is followed by robust synergistic correlations of 0.73 and 0.70 between soil conservation and habitat quality and soil conservation and carbon sequestration, respectively. These patterns correspond well with the observed spatial distributions of cold and hot spots for soil conservation, habitat quality, and carbon sequestration. In contrast, the interrelation between water yield and habitat quality registers the lowest correlation, below 0.1, suggesting a minimal synergy. This analysis reveals significant spatial variability in the interactions between ecosystem services across different zones, with clear distinctions in trade-offs involving food production and profound synergies among the other services, thus providing essential insights for managing ecosystem interactions effectively.

3.3.3. Spatial Differences in Ecosystem Service Trade-Offs and Synergies

Figure 9 illustrates significant variations in the Moran’s I indices for ecosystem services across various ecological restoration zones, which are consistent with the patterns noted in the Pearson correlation analysis. The analysis reveals that food production is generally associated with negative Moran’s I indices when compared to the other four services, suggesting consistent trade-offs. Conversely, indices for soil conservation, water yield, habitat quality, and carbon sequestration are positive, indicating prevalent synergistic interactions across these services. This pattern underscores that synergistic interactions are more common than trade-offs in these zones. Notably, the interaction between soil conservation and water yield displays the strongest synergy with a Moran’s I index of 0.402, aligning with Pearson correlation findings to highlight a robust cooperative relationship. Following this, the synergy between habitat quality and soil conservation is also significant, with an index of 0.206. Other synergistic relations among these services have Moran’s I indices ranging from 0.1 to 0.2, showing less pronounced synergies. The synergy between habitat quality and carbon sequestration, though positive, is the weakest with an index of 0.119. On the contrary, the Moran’s I indices for food production in relation to other services vary from −0.3 to −0.1, indicating differing levels of trade-offs. The most substantial trade-off occurs between food production and soil conservation, with an index of −0.281. Following this, the trade-offs decrease in intensity with habitat quality, water yield, and carbon sequestration, respectively, where the least significant trade-off is observed with carbon sequestration, marked by an index of −0.183.

The bivariate local Moran’s I index reflects the trade-offs and synergies between paired ecosystem services and their statistical significance, aligning closely with the results from Pearson correlation coefficients. This consistency confirms that the bivariate local Moran’s I index effectively captures the spatial trade-offs and synergies within the layout of national ecological restoration areas. However, the correlations derived from the bivariate local Moran’s I index are generally lower than those from Pearson correlation coefficients. This primarily stems from the inclusion of the spatial weighting matrix W_ij in the formulation of the local Moran’s I index, which introduces spatial lags into the interactions between ecosystem services. However, this incorporation of spatial lags effectively captures the spatial dependencies and interconnections that exist among different ecosystem services.

As illustrated in the figure (Figure 10), there is significant spatial heterogeneity among the local ecosystem services across the five ecological restoration zones. To enhance the clarity of these results, the administrative districts within each ecological restoration zone were used as the fundamental unit of analysis for a bivariate local spatial autocorrelation assessment. This approach allowed for a more detailed investigation of the trade-offs and synergies among ecosystem services within each zone. Overall, the areas exhibiting synergies between local ecosystem services were more extensive than those showing trade-offs across the ecological restoration zones. Apart from food production, habitat quality, water yield, carbon sequestration, and soil conservation generally displayed varying degrees of synergy. The spatial distribution patterns of local and global ecosystem services effects exhibit distinct characteristics of spatial differentiation. In terms of synergistic relationships, global synergies, encompassing both high and low synergy, are predominantly located in the western and southwestern parts of Liaoning Province, excluding the Western Carpathian Region (WCR) and its adjacent areas. Conversely, local ecosystem service effects, particularly synergistic relationships, are more frequently observed in the southeastern part of Liaoning, specifically within the WCR, indicating a certain degree of spatial variability. Conversely, food production demonstrated stronger trade-offs than synergies with the other four services. In terms of the spatial manifestation of synergistic effects between ecosystem services, these were predominantly clustered in the southeastern region of the WCR, forming a pattern resembling the number “7”. This area encompasses Donggang and Kuandian Manchu Autonomous County in Dandong City, Huanren Manchu Autonomous County in Benxi City, and Xinbin Manchu Autonomous County in Fushun City. The spatial distribution of trade-off effects lacks a clear pattern, with notable significance only in the northeastern part of the WSFR. In most cases, trade-offs are interspersed irregularly across the five ecological restoration zones, showing no distinct clustering. Overall, synergies between services far exceed trade-offs across the regions. Specifically, in the WCR, carbon sequestration exhibits synergistic effects with soil conservation, habitat quality, and water yield in both the northern and southeastern areas. Similarly, the synergies between water yield, soil conservation, and habitat quality extend outward from the central WCR, though the core region does not show significant spatial clustering. In contrast, food production demonstrates trade-offs with the other four services, forming a diagonal band from the northeast to the southwest, highlighting the presence of trade-off effects. In the WSFR, the spatial interaction between water yield and soil conservation is predominantly synergistic, covering nearly the entire region with strong spatial significance. Carbon sequestration and soil conservation also exhibit a clustered synergy in the northwest of the WCR, while habitat quality and carbon sequestration, as well as water yield, show a block-like distribution of synergy in the central WSFR. There is a small synergistic area between food production and habitat quality at the boundary between the WSFR and SWCR, but most interactions between services are expressed as trade-offs or non-significant relationships. In the SWCR, services generally exhibit trade-offs or lack significant spatial effects, with the exception of water yield and soil conservation, which demonstrate synergistic effects in the eastern part of the SWCR. This area of synergy covers more than 50% of the region, while the spatial interactions of other services are more mixed and interspersed. In the CR, food production shows trade-offs with the other four services, and habitat quality, soil conservation, water yield, and carbon sequestration display similar spatial distribution patterns of synergy. Both synergy and trade-offs are concentrated in the northern part of the CR, where it intersects with the WSFR and SWCR. In the IR, there are no significant patterns of synergy or trade-offs between services, with only sporadic distributions in the southern part of the region. Notably, in the central urban area of Dalian, food production shows synergies with carbon sequestration and habitat quality. In contrast, in the Lushunkou District of Dalian, habitat quality, water yield, and food production exhibit trade-offs.

4. Discussion

In the context of this research, the hotspots for habitat quality, carbon storage, water yield, and soil conservation services are predominantly situated within the Western Carpathian Region (WCR). This distribution pattern is largely attributed to the eastern region’s extensive forest cover, which accounts for about 40% of the watershed’s total land area. The dense vegetation and forests play a crucial role in sequestering carbon dioxide, releasing oxygen, preserving soil, and preventing erosion. Consequently, eastern Liaoning Province exhibits higher levels of carbon storage and soil conservation. This observation corroborates the findings of Zhang et al., who noted that areas with greater vegetation coverage tend to have enhanced carbon sequestration and soil conservation capabilities [57]. The bivariate local spatial autocorrelation analysis reveals that soil conservation, habitat quality, carbon storage, and water yield exhibit synergistic relationships when examined in pairs at a local scale. This finding is consistent with numerous previously published studies [58,59]. The global ecosystem service regions predominantly display strong and weak trade-off effects, with low trade-off effects being less common and high synergistic effects being the least prevalent. This pattern is consistent with existing research findings [29,60].

Under optimal conditions, a significant body of research suggests that “high synergy” or “positive synergy” among all ecosystem services within a single ecosystem signifies the maximization of benefits, embodying the ultimate aim of ecosystem management. However, in the current reality, only certain regions, and often a minority, can attain this ideal state. The majority exhibit a predominant “trade-off” dynamic, often involving one or two services, and in some cases, an “underperforming synergy” where ecosystem services are not optimally balanced. In these circumstances, through deliberate human intervention and the establishment of appropriate management regulations, transforming ecosystems primarily characterized by “low synergy” relationships into those with a dominant “strong trade-off” dynamic centered around one or two service functions can serve as a practical, short-term goal.

In Liaoning Province, the synergistic effects between pairs of local ecosystem services surpass the trade-offs. However, from a global perspective, areas characterized by low synergy account for 29.09% of the study region, with adjacent areas being in a state where ecosystem services are far from optimal. Transitioning from low synergy to strong trade-offs can effectively ameliorate the least desirable conditions in these regions. According to the classification of global ecosystem service trade-offs and strategies in this study, improving the relationship from low synergy to strong trade-offs is not particularly challenging; it merely requires elevating one “medium” or “low” ecosystem service to a “high” state. During this adjustment process, the optimal approach is to align the type of ecosystem service being enhanced with the functional orientation of the ecological restoration distribution. This alignment not only facilitates the transition of ecosystem service relationships from low synergy to strong trade-offs but also harmonizes with the local ecological function zone’s positioning, thereby achieving a mutually beneficial outcome.

This study delved into the local ecosystem service dynamics, establishing an analytical framework that spans from a global to a localized perspective. At the global level, the predominant dynamics are characterized by strong and weak trade-offs, alongside minimal synergies, with areas of high synergy representing a mere 2.24% of the study region. Viewed from the local scope, the spatial extent of synergistic relationships among ecosystem service pairs is greater than that of trade-offs, presenting a marked difference from the global perspective. This pattern indicates that, on a local scale, ecosystem services generally exhibit favorable synergistic relationships. Nonetheless, achieving an ideal state of synergy globally is hampered by the complex nature of service interactions across multiple levels. An analysis encompassing both global and local views yields deeper insights. Regions of synergy, both local and global, mainly occur in WCR, known for its dense forests, lush vegetation, and effective soil and water conservation [61]. Conversely, WSFR experiences notable trade-offs due to its closeness to the Horqin sandy lands [62], where wind erosion significantly undermines the synergy efficiency of ecosystem services [63], a reflection of the unique ecological traits of each restoration zone.

However, this study has certain limitations. Primarily, it relies on data collected within a single year, constraining the assessment of temporal and geographical dynamics in ecosystem service trade-offs and synergies. Future studies could extend the use of spatial overlay techniques across longer periods to better understand these dynamics on a temporal and spatial scale. Moreover, the focus on five specific ecosystem services might not capture the full spectrum of potential interactions. Expanding the analysis to include socio-economic services could diversify the outcomes concerning local and global trade-offs and synergies [64], meriting additional exploration. Finally, the current research does not fully explore the underlying mechanisms driving spatial variations in ecosystem service trade-offs and synergies [65]. A significant challenge is the absence of robust quantitative measures for trade-off and synergy indicators. Although approaches such as redundancy analysis and geographic detectors, supported by correlation coefficients or root mean square errors [66,67,68], are employed to probe these driving forces, they often do not adequately address the intricate factors that influence these relationships [69,70]. The nature and intensity of ecosystem service dynamics are influenced by various elements, including the rate and significance of relationship changes [71,72,73]. Future inquiries should, therefore, integrate diverse analytical methods [74] to thoroughly account for these influencing factors.

5. Conclusions

(1) The overall ecosystem services in Liaoning Province are predominantly characterized by four types of relationships: strong trade-offs, weak trade-offs, high synergies, and low synergies. Among these, strong trade-offs have the highest proportion at 35.51%, followed by weak trade-offs, which account for 33.17%. Low synergy regions constitute 29.09%, while high synergy areas have the smallest share, making up only 2.24%. As a result, the regions with high synergy between ecosystem services are very limited across the entire study area.

(2) In examining local ecosystem services, it is evident that areas with synergistic relationships among different ecosystem services outnumber those with trade-offs. Specifically, there are notable trade-offs between food production and the other four services, while soil conservation, water yield, habitat quality, and carbon storage generally display synergistic relationships. Notably, the spatial synergy between soil conservation and water yield is the most pronounced.

(3) Within each ecological restoration zone, the relationships between local services vary. The southwestern part of WSFR, the eastern part of SWCR, and the southeastern part of WCR exhibit the strongest synergistic effect between water yield and soil conservation. The trade-off between food production and soil conservation is the strongest, with spatial characteristics similar to those observed between water yield and soil conservation in WSFR, SWCR, and WCR.

(4) On a global scale, trade-off areas between ecosystem services in all land ecological restoration zones are significantly larger than synergistic areas, especially when compared to high-synergy areas, which are almost exclusively found in WCR. However, on a local scale, the synergistic areas between pairs of ecosystem services are larger than the trade-off areas, reflecting a notable distinction. WCR’s designation as an area with high water and soil conservation capabilities aligns with this phenomenon, which is particularly pronounced in each zone’s specific characteristics. These measures aim to align with the goals of the Liaoning Provincial Land Spatial Planning (2021–2035) and address the specific environmental challenges identified in our research, contributing to the sustainable development of the region. (SWR) Prioritize vegetation restoration and soil conservation, with a special emphasis on rejuvenating forests and grasslands in western Liaoning’s low mountains and hills. (SWCR) Fortify the northwestern forest belt, advance desertification control, and manage soil erosion, focusing on Horqin Sandy Land’s southern region and Nuru Erhu Mountains for vegetation restoration and desertification prevention. (CR) Enhance water conservation and ecological safeguards in the Liao River’s upper basin, emphasizing urban and rural ecological restoration and the integrated management of coal mining subsidence areas to optimize national land use. (WCR) Bolster forest conservation and restoration, augment carbon sequestration in eastern Liaoning’s forests, and protect the region’s ecological security, with a focus on water conservation and biodiversity in the southern Changbai Mountains and eastern Qianshan Mountains. (IR) Intensify efforts to curb marine and terrestrial pollution, expedite the restoration and protection of coastlines, islands, and wetlands, and prioritize the swift recovery of key habitats’ quality.