EN中文

Journal of Geographical Sciences >

2023 , Vol. 33 >Issue 4: 845 - 862

DOI: https://doi.org/10.1007/s11442-023-2109-7

Research Articles

Separating the effects of two dimensions on ecosystem services: Environmental variables and net trade-offs

  • ZUO Liyuan , 1, 2 ,
  • JIANG Yuan 1, 2 ,
  • GAO Jiangbo , 1, * ,
  • DU Fujun 1, 2 ,
  • ZHANG Yibo 1, 2
Expand
  • 1. Key Laboratory of Land Surface Pattern and Simulation, Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China
  • 2. University of Chinese Academy of Sciences, Beijing 100049, China
*Gao Jiangbo, PhD and Professor, specialized in integrated physical geography, mountain ecosystem services, climate change impact and adaptation. E-mail:

Zuo Liyuan, specialized in mountain ecosystem services. E-mail:

Received date: 2022-10-24

  Accepted date: 2022-11-30

  Online published: 2023-05-11

Supported by

National Natural Science Foundation of China(42071288)

National Natural Science Foundation of China(41671098)

Kezhen-Bingwei Excellent Young Scientists of the Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences(2020RC002)

Fold

Abstract

Spatial and temporal changes in ecosystem services (ESs) are driven by two types of factors: environmental factors and trade-offs/synergies between services. In the ecological conservation red line (ECRL) area, in which national ecological security and social sustainable development are guaranteed, it is particularly important to clarify the driving mechanism of ESs for the management of ecosystems. In this study, soil conservation, water yield, and carbon sequestration in Beijing’s ECRL area are quantified, and GeoDetector is used to identify the factors influencing the trade-offs/synergies between ESs. Moreover, partial correlation analysis is used to calculate the net trade-offs/synergies and compare them with the extent to which environmental variables contribute to ESs. The results are as follows: environmental variables and trade-offs/synergies have different effects on the changes in ESs, and their interactions can enhance the determinative power of the corresponding individual variable. The land use intensity is an extremely important factor affecting the trade-offs/synergies between the three services, indicating that rational land use planning in Beijing’s ECRL area is crucial for avoiding the negative impacts of trade-offs and enabling coordinated optimization of ESs. After the elimination of the cross-influence of environmental variables, the trade-offs/synergies change significantly, and the impact of environmental variables on ESs is compared with the net trade-offs/synergies. Environmental variables are the driving forces of the spatiotemporal changes in soil conservation. Precipitation and carbon sequestration have similar effects on water yield. Spatiotemporal changes in carbon sequestration are closely related to the other two services, with smaller influences from environmental variables.

Key words： ecosystem services; trade-offs and synergies; driving mechanisms; quantitative separation; ecological conservation red line area

Cite this article

ZUO Liyuan , JIANG Yuan , GAO Jiangbo , DU Fujun , ZHANG Yibo . Separating the effects of two dimensions on ecosystem services: Environmental variables and net trade-offs[J]. Journal of Geographical Sciences, 2023 , 33(4) : 845 -862 . DOI: 10.1007/s11442-023-2109-7

1 Introduction

The benefits that ecosystems provide to humans are called ecosystem services (ESs) (MEA, 2005; Costanza et al., 2017). As a bridge connecting natural ecosystems and human society, ESs are affected by various factors such as the natural environment, social economy, and human demands in the formation and development processes (Guerry et al., 2015; Li et al., 2020). Moreover, there are also complex non-linear changes among ESs (Carpenter et al., 2009). The increase or decrease of one ES will affect the supply capacity of another service, resulting in trade-offs or synergies (Rodriguez et al., 2006). However, the influences of environmental factors and the trade-offs/synergies between ESs are not independent of each other, and the trade-offs/synergies are also driven by environmental factors (Feng et al., 2017; Zuo et al., 2021). Bennett et al. (2009) summarized the driving factors into two types: one is that the driving factor affects a service alone, and the other is that ESs are affected by a common driver. Eliminating the cross-influences of environmental factors on the trade-offs/synergies and comparing the different driving mechanisms of ESs helps to eliminate the negative impacts of trade-offs and to achieve coordinated and sustainable development of social-ecological systems (Howe et al., 2013; Cord et al., 2017).
The analysis of the driving forces of ESs and their trade-offs is a major challenge in current research on ESs (Bennett et al., 2015; Zhang et al., 2020). Gao et al. (2021) showed that vegetation coverage determines the spatial heterogeneity of soil loss in karst areas. A previous study demonstrated that increased land use intensity increases not only food production but also the risk of soil erosion (Xu et al., 2016). Chen et al. (2020) focused on 10 ESs and found that human activities strongly affect the spatial distribution of the trade-off bundles in Beijing and its surrounding areas. As a spatial boundary that must be strictly protected, the natural environment is the core variable that affects the changes in ESs in this region (Bai et al., 2016; Ouyang et al., 2016). For example, Gao et al. (2020) reported that vegetation coverage and land use are the dominant factors that affect soil loss and water yield in Beijing’s ECRL area, respectively. Precipitation significantly affects the spatial distribution of the trade-off relationship between soil conservation and carbon sequestration in Beijing’s ECRL area (Zuo and Gao, 2021).
The above researchers carefully analyzed the responses of ESs or trade-offs/synergies to temporal and spatial changes in environmental variables, but the cross-influences of the relationships between environmental variables and trade-offs/synergies have not been fully considered, and the separation of the driving forces is still lacking (Bennett et al., 2009; Dade et al., 2019). Accurately identifying and eliminating the influencing variables of trade-offs/synergies is the key to quantitative separation of the driving forces. Affected by temporal and spatial variations in the environmental variables and ESs, the spatial heterogeneity of the trade-offs/synergies is prominent (Qiu et al., 2013). GeoDetector uses the hierarchical heterogeneity of the spatial data to reveal the driving mechanism behind it, and GeoDetector is widely used in social sciences, natural sciences, environmental health, and other fields (Wang and Xu, 2017) for tasks that include the evaluation of suitability of land for urban development (Wang et al., 2021), the quantitative identification of impact factors (Yang et al., 2019), and the detection of human health risk factors and risk areas (Wang et al., 2010). In this study, GeoDetector was used to quantitatively identify environmental variables that affect the trade-offs/synergies and to prove that there is an interaction between these two driving forces, laying a foundation for the subsequent quantitative separation of the driving forces to changes in ESs.
As the bottom line and lifeline of national and regional ecological security, the ECRL aims to maintain the leading functions of China’s important ecosystems and enhance the ecological support capacity of social and economic development (CMEP, 2015). Beijing’s ECRL was officially released by the People’s Government of the Beijing Municipality in 2018, including areas with important ecological functions such as water retention and soil conservation, ecologically sensitive areas of soil erosion, and protected areas such as municipal drinking water sources, forest parks, and national key ecological public welfare forests (SCC, 2018). Beijing’s ECRL area is located in the Yanshan Mountains in the north and the Taihang Mountains in the west. The instability of high mountains and steep slopes and the vulnerability and low recovery of broken habitats make the red line area extremely vulnerable (Gong et al., 2017; Grêt-Regamey and Weibel, 2020), and the ecological environment is sensitive to climate change and land use changes (Jiang et al., 2019). In addition, the huge spatial differences in the natural geographical environment of mountainous areas and the drastic changes in environmental conditions affect the formation and development of ecological processes, which in turn leads to significant variability in the supply capacity and trade-offs/synergies of ESs (Briner et al., 2013; Yu et al., 2021). For example, as the elevation and vegetation coverage increase, the trade-offs between water yield and net primary productivity (NPP) increase significantly (Liu et al., 2019). The creation of forest management policies should take into account the effects of climate change along the elevation gradients (Mina et al., 2017). Therefore, the quantitative interpretation of the driving mechanism of ESs under heterogeneous landscape conditions in the ECRL area is still a key research topic (Zuo and Gao, 2021). Integrated analysis of the impacts of environmental variables and trade-offs/synergies on changes in ESs can provide targeted scientific references for strictly adhering to the red line (Ouyang et al., 2016; Xu et al., 2018).
In this study, three ESs (soil conservation, water yield, and carbon sequestration) that are closely related to the dominant ecological functions in Beijing’s ECRL area were considered (Figure 1). The factors influencing the trade-offs/synergies between these three ESs were quantitatively identified using GeoDetector, and the interactions between environmental variables and service variables were investigated. Furthermore, through partial correlation analysis, the impacts of net trade-offs/synergies and environmental variables on the spatiotemporal changes in ESs were compared to provide a scientific research framework for the quantitative separation of the driving forces of ESs.
Figure 1 Framework for quantitative separation of ES drivers

2 Study area and methods

2.1 Study area

Beijing (39°28′-41°05′N, 115°25′-117°30′E) is located in the northern part of the North China Plain. It is surrounded by mountains to the west, north, and northeast. The mountainous area accounts for 62% of the total area of Beijing. The plains, shallow mountains, and deep mountains are sequentially distributed from southeast to northwest. Beijing’s ECRL area has an ecological pattern consisting of two barriers and two zones. The two barriers refer to the ecological barrier of the Yanshan Mountains in the north and the Taihang Mountains in the west, and the two zones are the ecological protection zones along the Yongding River and along the Chaobai River-Ancient Canal. ​​Beijing’s ECRL covers an area of 4290 km2, accounting for 26.1% of the total area of Beijing, with an elevation range of −5 to 2223 m (Figure 2). The study area has a warm temperate semi-humid semi-arid monsoon climate, with four distinct seasons. The annual average temperature is 9.98°C, the annual average precipitation is 594.24 mm, and 80% of the annual precipitation is concentrated between June and August. According to the dominant ecological functions, Beijing’s ECRL area is divided into four types of areas: soil conservation, water retention, biodiversity maintenance, and important river and wetland areas (SCC, 2018).
Figure 2 Elevation (a) and sub-regions (b) of the study area (Beijing)

2.2 Data sources

The following datasets were used in this study:
(1) The digital elevation model (DEM) data were downloaded from Google Earth v6.0.3, with a spatial resolution of 9 m.
(2) The meteorological data were obtained from the National Climate Center of the China Meteorological Administration (http://data.cma.cn/). In this study, the daily data from 35 meteorological stations in the study area and its surrounding areas were selected, and the spatial distribution of the meteorological data, including the precipitation and evapotranspiration data, with a resolution of 1 km, was interpolated using the professional ANUSPLIN v4.2 software.
(3) The soil property data were obtained from the Cold and Arid Regions Sciences Data Center at Lanzhou, China (http://westdc.westgis.ac.cn), which are based on the Harmonized World Soil Database (HWSD v1.2) with a spatial resolution of 1 km.
(4) The land use data were obtained from the Beijing Municipal Ecology and Environment Bureau, with a spatial resolution of 15 m.
(5) Based on the data from the HJ1A/B CCD (30 m), GF1 WFV (16 m) (http://www. cresda.com/CN/), and MODIS MOD09GQ (250 m) (https://lpdaac.usgs.gov), the normalized difference vegetation index (NDVI) data were obtained via a linear combination of reflectance values (near-infrared, red band), with a spatial resolution of 30 m.
(6) The vegetation type data were obtained from the Data Center for Resources and Environmental Sciences, Chinese Academy of Sciences (RESDC) (http://www.resdc.cn), with a spatial resolution of 1 km.
(7) The map of Beijing’s ECRL was released by the Beijing Municipal Ecology and Environment Bureau in 2018 (http://sthjj.beijing.gov.cn/bjhrb/index/xxgk69/zfxxgk43/fdzdgknr2/zcfb/szfgfxwj/325832692/index.html), and in this study, we obtained the sub-regions of Beijing’s ECRL (Figure 2) by digitization.

2.3 Methods

2.3.1 Quantifying the ES supply

In this study, the revised universal soil loss equation (RUSLE) model (Renard et al., 1997), integrated valuation of ecosystem services and trade-offs (InVEST) model (Sharp et al., 2016), and Carnegie-Ames-Stanford approach (CASA) model (Monteith and Moss, 1977, Potter et al., 1993) were used to quantitatively evaluate the spatial patterns of soil conservation, water yield, and carbon sequestration in the study area, respectively. To eliminate the influences of inter-annual differences in climate and extreme climate events (such as the July 20 Heavy rainstorm in Beijing) on the identification of the driving forces, in this study, ESs in Beijing’s ECRL area from 2015 to 2018 were evaluated and analyzed as a four-year average. The required parameters and calculation process of the model are presented in Table 1.
Table 1 Models used to quantify ESs
ESs Models Calculation method
Soil conservation RUSLE model $A=R\times K\times LS\times \left( 1-C\times P \right)$ (1)
where A denotes soil conservation (t·ha−1·a−1), and R is the rainfall erosion factor. Owing to the significant seasonal differences in the precipitation in the study area, the monthly R factor is calculated and then integrated to the annual scale (Arnoldus et al., 1980). K is the soil erodibility factor, which is calculated based on the soil property data using the erosion productivity impact calculator (EPIC) (Williams et al., 1989). LS is the slope length and steepness factor. High-resolution DEM (9 m) data were used to calculate the LS to fully characterize the complexity of the elevation in the study area (Naipal et al., 2015). C is the cover and management factor, which is calculated based on the NDVI data (Cai et al., 2000). P is the practice factor, which is assigned according to the type of land use with reference to previous research on the North China Plain (Xu et al., 2012). LS, C, and P are dimensionless.
Water yield InVEST model $Y(x)=\left( 1-\frac{AET(x)}{P(x)} \right)\times P(x)$ (2)
where Y(x), AET(x), and P(x) are the annual WY, annual actual evapotranspiration, and annual precipitation in grid unit x, respectively.
Carbon sequestration CASA model $NP{{P}_{t}}=APA{{R}_{t}}\times {{\varepsilon }_{t}}$ (3)
where t is the period over which the NPP accumulated; and APARt and εt are the photosynthetically active radiation (MJ·m−2) and the actual light-use efficiency (gC·MJ−1) absorbed by the vegetation in pixel x in time t, respectively.

2.3.2 GeoDetector

In this study, GeoDetector was used to analyze the spatially stratified heterogeneity of ESs and to reveal the driving mechanisms behind it. This method has two main advantages: the factor detector can quantitatively determine the explanatory power of each influencing variable; and the interaction detector can judge whether there is an interaction between two influencing variables, as well as the strength of the interaction, its direction, and whether it is linear or nonlinear (Table 2) (Wang et al., 2016). Thus, it can provide a scientific basis for the next step of the partial correlation analysis. GeoDetector uses the q statistic to measure the degree of interpretation of the independent variable X (influencing variables) to the spatial differentiation of the dependent variable Y (ESs). The q statistic is expressed as follows:
$q=1-\frac{\sum\limits_{h=1}^{L}{{{N}_{h}}}{{\sigma }_{h}}^{2}}{N{{\sigma }^{2}}}$
Table 2 Types of interactions between pairs of variables
Description Interaction
q(X1∩X2) < Min[q(X1), q(X2)] Nonlinear weakening
Min[q(X1), q(X2)] < q(X1∩X2) < Max[q(X1), q(X2)] Single factor nonlinear weakening
q(X1∩X2) > Max[q(X1), q(X2)] Double factor
q(X1∩X2) = q(X1) + q(X2) Independent
q(X1∩X2) > q(X1) + q(X2) Nonlinear enhancement
where h=1, 2, …, L refers to the strata of variables; N and σ2 are the total number of sample units and the variance, respectively; and Nh and ${{\sigma }_{h}}^{2}$ are the number of sample units and the variance in stratum h, respectively. The term $\sum\limits_{h=1}^{L}{{{N}_{h}}}{{\sigma }_{h}}^{2}$ is the sum of the strata variance, and Nσ2 is the total sum of the variance.

2.3.3 Partial correlation analysis

To avoid the cross-influence of environmental variables, in this study, the variation characteristics of the spatial relationships between pairwise ESs were quantitatively assessed using a spatiotemporal statistical framework by conducting partial correlation analysis at the pixel scale (Tomscha and Gergel, 2016). The advantage of this method is that it can analyze the changes in the time series of ESs, and it can also clarify the spatial relationship and change characteristics of ESs (Wang et al., 2017). A positive coefficient indicates a synergistic relationship between the two ESs, and a negative coefficient indicates a trade-off relationship. The corresponding formulas for calculating the correlation coefficient and partial correlation coefficient between ESs are as follows:
${{r}_{12\left( ij \right)}}=\frac{\sum\limits_{n=1}^{n}{\left( ES{{1}_{n\left( ij \right)}}-\overline{ES{{1}_{\left( ij \right)}}} \right)\left( ES{{2}_{n\left( ij \right)}}-\overline{ES{{2}_{\left( ij \right)}}} \right)}}{\sqrt{\sum\limits_{n=1}^{n}{{{\left( ES{{1}_{n\left( ij \right)}}-\overline{ES{{1}_{\left( ij \right)}}} \right)}^{2}}\sum\limits_{n=1}^{n}{{{\left( ES{{2}_{n\left( ij \right)}}-\overline{ES{{2}_{\left( ij \right)}}} \right)}^{2}}}}}}$
${{r}_{12\cdot 3\left( ij \right)}}=\frac{{{r}_{12\left( ij \right)}}-{{r}_{13\left( ij \right)}}{{r}_{23\left( ij \right)}}}{\sqrt{\left( 1-{{r}^{2}}_{13\left( ij \right)} \right)\left( 1-{{r}^{2}}_{23\left( ij \right)} \right)}}$
${{r}_{12\cdot 34\left( ij \right)}}=\frac{{{r}_{12\cdot 3\left( ij \right)}}-{{r}_{14\cdot 3\left( ij \right)}}{{r}_{24\cdot 3\left( ij \right)}}}{\sqrt{\left( 1-{{r}^{2}}_{14\cdot 3\left( ij \right)} \right)\left( 1-{{r}^{2}}_{24\cdot 3\left( ij \right)} \right)}}$

3 Results and analysis

3.1 Spatial distributions and characteristics of ESs

The spatial distributions of and dynamic changes in ESs are shown in Figure 3. The high soil conservation values were mainly concentrated in the high-elevation mountainous areas in the northeastern and southwestern parts of the study area. Forests and grassland are widely distributed in these mountainous areas, and the erosion resistance of vegetation roots and litter strengthen the soil conservation ability in the mountainous areas (Wiśniewski and Märker, 2019). From 2015 to 2018, the soil conservation ranged from 0 to 1700 t·ha−2·a−1, with small changes over time, but the mean values of the different years varied greatly. By comparing red line areas with different dominant ecological functions, it was found that the mean soil conservation value was the highest in the soil conservation ECRL area (Table S1), which confirms the rationality of the dominant function of this red line area. The land use types such as rivers, lakes, and wetlands were distributed in the important river and wetland ECRL area, which was affected by runoff erosion and had the lowest soil conservation values (Table S1).
Figure 3 Spatial distribution of ESs in Beijing’s ECRL areas from 2015 to 2018
Table S1 Statistics of the mean value of ESs in Beijing’s ECRL areas
ESs Year Beijing’s ECRL area Water
retention
ECRL area		 Soil
conservation
ECRL area		 Biodiversity maintenance
ECRL area		 Important river
and wetland
ECRL area
Soil conservation (t·ha‒1·a‒1) 2015 345.27 379.34 510.74 357.53 42.74
2016 542.72 487.59 878.95 713.40 50.46
2017 438.77 411.77 677.98 566.56 43.49
2018 298.54 351.53 394.99 302.66 32.68
Water yield (mm) 2015 131.92 157.08 153.93 75.57 124.24
2016 330.21 330.62 423.48 336.18 212.62
2017 246.92 256.64 311.55 250.39 138.20
2018 143.28 184.97 135.16 89.99 114.69
Carbon sequestration (gC·m‒2) 2015 452.95 479.10 392.44 513.27 349.34
2016 516.91 532.89 451.76 618.76 379.72
2017 480.61 498.50 436.29 580.33 316.44
2018 419.44 445.56 374.40 493.48 275.07
The water yield gradually increased from north to south, and its spatial distribution exhibited clear heterogeneity. The value and spatial distribution of water yield were closely related to precipitation, and the Spearman correlation coefficient between water yield and precipitation in each year was greater than 0.64. In 2016, heavy rainfall with a long duration and large total amount of precipitation occurred in Beijing. In that year, the water yield of Beijing’s ECRL area ranged from 0 to 900.17 mm, with an average value of 330.21 mm, which was significantly higher than in the other years. By comparing the four sub-red line areas, it was found that the highest water yield values occurred in the water retention ECRL area and the soil conservation ECRL area (Table S1).
The spatial distribution of carbon sequestration was closely related to the land use types in the study area. The vegetation in the forest and grassland was densely distributed, and the cumulative carbon sequestration was significantly higher than for other land use types. From 2015 to 2018, the carbon sequestration value in Beijing’s ECRL area ranged from 30 to 1200 g C m−2. This result is consistent with those of previous studies conducted in the mountainous area of Beijing (Yin et al., 2015). The high carbon sequestration values were mainly concentrated in the biodiversity maintenance ECRL area (Table S1), which provided good environmental conditions for the habitat and survival of organisms, and played an important role in maintaining the species richness, vegetation diversity, and high carbon sequestration performance of this area.

3.2 Quantitative attribution of the spatial relationships of ESs

A prerequisite for the quantitative separation of the driving factors is the accurate identification and elimination of the influencing variables of trade-offs/synergies. In this study, it was assumed that if an environmental variable synchronously affected a pair of ESs, the environmental variable was considered to be an extremely important factor; if it affected only one ES, it was a generally important factor; otherwise, it was an unimportant factor. By comparing the q values of the individual environmental variables, the different types of environmental variables were determined. Elevation, land use intensity, precipitation, and slope were the dominant environmental variables affecting soil conservation. Water yield was affected by vegetation coverage, land use intensity, and precipitation. Elevation, vegetation coverage, land use intensity, and slope were the dominant environmental variables affecting carbon sequestration (Table 3). Among them, land use intensity and precipitation jointly affected the spatial distribution of soil conservation and water yield, and were extremely important factors controlling the spatial variations in these two ESs. Vegetation coverage explained 19.4% of the spatial variations in water yield. Moreover, water yield had an impact on the spatial heterogeneity of soil conservation. Thus, vegetation coverage affected the spatial heterogeneity of soil conservation by affecting water yield and was a generally important factor. Elevation and slope only affected the spatial distribution of soil conservation and were generally important factors.
Table 3 The q values of the variables influencing ESs
Environmental variables Services variables
Elevation Vegetation coverage Land use intensity Precipitation Slope Soil conservation Water yield Carbon
sequestration
Soil conservation 0.233 0.074 0.133 0.116 0.530 —— 0.115 0.116
Water yield 0.062 0.194 0.443 0.283 0.099 0.139 —— 0.173
Carbon
sequestration		 0.318 0.424 0.289 0.053 0.192 0.149 0.224 ——

Note: The level of significance (p value) is <0.01.

Vegetation coverage and land use intensity were extremely important factors that affect water yield and carbon sequestration, and their q values were greater than 0.4. Precipitation explained 28.3% of the spatial variation in water yield. Moreover, when the water yield acted as an independent variable, it explained 22.4% of the spatial distribution of carbon sequestration. Thus, precipitation indirectly affected carbon sequestration by affecting water yield, making it a generally important factor. Elevation and slope played a dominant role in the spatial distribution of carbon sequestration but had insignificant effects on water yield, making them generally important factors. Elevation, land use intensity, and slope all significantly affected the spatial distributions of soil conservation and carbon sequestration and were extremely important factors affecting these two ESs. Precipitation exerted an influence on carbon sequestration by affecting soil conservation, making it a generally important factor. Vegetation coverage had a strong determinative power on the spatial differentiation of carbon sequestration, with a q value of 0.424, but its explanatory power for soil conservation was not significant, making it a generally important factor.
By comparing the impacts of individual variables (environmental variable or service variable) and their interactions on ESs, it was found that the magnitude of the interactions between environmental variables and service variables was not simply the sum of the influences of the two types of variables (Tables 3 and 4), which demonstrates that these two types of variables were not independent of each other. Moreover, the interactions between service variables and environmental variables all had greater q-values than the two individual variables, indicating that the interactions between these two types of variables were enhancing interactions. The enhancement form is mainly manifested in two ways: the interaction between the service variable and the environmental variable being greater than the maximum value of the corresponding individual variable, i.e., double factor enhancement; and the interaction being greater than the sum of the two individual variables, i.e., nonlinear enhancement (Table 2).
Table 4 Interaction between ecosystem services and environmental variables
Dependent variable Water yield Dependent variable Soil
conservation		 Dependent variable Carbon
sequestration
Independent
variable		 Soil conservation Carbon sequestration Independent
variable		 Water yield Carbon sequestration Independent
variable		 Water yield Soil conservation
Elevation 0.172 0.217 Elevation 0.318 0.267 Elevation 0.466 0.363
Vegetation coverage 0.273 0.233 Vegetation coverage 0.164 0.145 Vegetation coverage 0.441 0.464
Land use intensity 0.502 0.485 Land use intensity 0.22 0.194 Land use intensity 0.332 0.335
Precipitation 0.353 0.459* Precipitation 0.206 0.242* Precipitation 0.306* 0.232*
Slope 0.165 0.218 Slope 0.579 0.566 Slope 0.309 0.233

Note: * denote that the two-factor interaction is nonlinear enhancement, and the others denote that the interaction is bivariate enhancement.

3.3 Net trade-offs/synergies between ESs and their spatial distributions

Table 5 shows that the mean value of the simple correlation coefficient between water yield and soil conservation was 0.319. The first-order partial correlation results (excluding precipitation) and the second-order partial correlation results (excluding precipitation and land use intensity) show that the mean value of the correlation coefficient between water yield and soil conservation decreased significantly, and the reduction range was the same, indicating that the correlation between water yield and soil conservation was dominated by precipitation. When the land use intensity was excluded and only the influence of precipitation was considered, the mean value of the correlation coefficient between water yield and soil conservation increased significantly and the trade-off percentage decreased significantly. When the vegetation coverage, land use intensity, and both vegetation coverage and land use intensity were excluded, the correlation between water yield and carbon sequestration was significantly enhanced, and the mean value of the correlation coefficient increased from 0.189 to 0.528. In addition, the area percentage of the synergy between water yield and carbon sequestration increased significantly, indicating that the vegetation coverage and land use intensity inhibited the synergy between water yield and carbon sequestration. Since the elevation and slope did not change with time, the second- and third-order partial correlation coefficients between soil conservation and carbon sequestration could not be calculated. Only the first-order partial correlation coefficients between the three environmental factors were analyzed. The results show that when any two factors (elevation, slope, and land use intensity) were held constant, the trade-off between soil conservation and carbon sequestration was affected by the other factor, and the trade-off area decreases, indicating that rational planning of elevation, slope, and land use was conducive to the development of the spatial relationship between soil conservation and carbon sequestration towards the direction of synergistic optimization.
Table 5 Correlation coefficients between ESs and the proportion of the trade-off/synergy
Paired ESs Excluded factors Correlation coefficients Trade-off area percentage (%) Synergy area percentage (%)
Water yield-
Soil conservation		 — 0.319 32.21 67.79
Precipitation 0.070 46.26 53.74
Land use intensity 0.545 16.94 83.06
Precipitation and land use intensity 0.080 46.65 53.35
Water yield-
Carbon sequestration		 — 0.189 37.81 62.19
Vegetation coverage 0.286 32.26 67.74
Land use intensity 0.381 25.02 74.98
Vegetation coverage and land use intensity 0.528 19.69 80.31
Soil conservation-
Carbon sequestration		 — 0.246 37.06 62.94
Elevation 0.379 24.37 75.63
Slope 0.369 25.15 74.85
Land use intensity 0.413 21.68 78.32

Note: The level of significance (p value) is <0.05.

The distribution of trade-offs/synergies of the simple correlation between ESs and the net correlation between the two after removing the extremely important factors was compared (Figure 4). The results show that there was a transition from trade-off to synergy between water yield and soil conservation in Beijing’s ECRL area, which is distributed sporadically. In addition, 27.86% of the area of Beijing’s ECRL changed from synergy to trade-off. This change mainly occurred in the water retention ECRL area and the soil conservation ECRL area. The synergistic relationship between water yield and carbon sequestration was relatively stable. The areas where the synergy remained unchanged after excluding the extremely important factors accounted for 53.07% of the ECRL area. After removing the influences of the extremely important factors, the areas where the trade-off between water yield and carbon sequestration changed to synergy increased significantly, and these areas were mainly distributed in the water retention ECRL area and the biodiversity maintenance ECRL area, accounting for 27.24% of Beijing’s ECRL area.
Figure 4 Spatial distribution of the tradeoff and synergy under simple and net correlations in Beijing

3.4 Comparison of driving forces between environmental variables and service variables

Table 6 compares the impacts of environmental variables and service variables on the changes in ESs. For environmental variables, the mean correlation coefficients between slope, precipitation and soil conservation were the highest, followed by elevation. After excluding the effects of precipitation and land use intensity, the mean value of the net correlation coefficient between water yield and soil conservation was less than 0.1, indicating that environmental variables had a stronger influence on soil conservation than water yield. After excluding the influence of land use intensity, the first-order partial correlation coefficient between soil conservation and carbon sequestration was slightly lower than that of slope and precipitation, indicating that the spatial variations in soil conservation were related to carbon sequestration but were still dominated by environmental variables. The spatial and temporal variations in water yield were jointly affected by environmental variables and service variables, and their correlation coefficients were similar. Precipitation, as an important supply of water yield in northern China, was the dominant environmental variable that affected the spatiotemporal distribution of water yield. The results of the correlation analysis revealed that the average spatial correlation coefficient between precipitation and water yield was 0.599, and the positive correlation area accounted for 90.99% of the entire study area. Table 6 shows that the mean values of the correlation coefficients between environmental variables and carbon sequestration in descending order were as follows: vegetation coverage > elevation > precipitation > slope > land use intensity. Compared with environmental variables, the spatiotemporal relationship between carbon sequestration and water yield was stronger. After excluding the extremely important factors, the net correlation coefficient between carbon sequestration and water yield was 0.528, and the first-order partial correlation coefficient (after excluding the land use intensity) between carbon sequestration and soil conservation was 0.413. This was significantly higher than the influence of environmental variables, indicating that the spatial variations in carbon sequestration were closely related to the other two service variables.
Table 6 Mean values of the correlation coefficients between ESs and impact variables
Environmental variables Service variables (net correlation)
Elevation Vegetation coverage Land use intensity Precipitation Slope Soil conservation Water yield Carbon sequestration
Soil conservation 0.257 0.006 0.095 0.435 0.436 — 0.080 0.413*
Water yield 0.014 0.048 0.129 0.599 0.090 0.080 — 0.528
Carbon sequestration 0.269 0.361 0.074 0.245 0.130 0.413* 0.528 —

Note: The level of significance (p value) is <0.05. Since the elevation and slope did not change with time, only the first-order partial correlation between soil conservation and carbon sequestration was calculated, and * denotes the first-order partial correlation coefficient after excluding the land use intensity.

4 Discussion

As the bottom line of ecological environmental security, the delineation of the ECRL is a macro-policy implemented to enhance the positive mutual feedback between urban development and ecological protection, and it is an important measure for achieving the construction of an ecological civilization (Gao, 2019). The status of land use and land use planning needs to be considered in delineating the ECRL (CCICED, 2014). Similarly, the management of ESs in the ECRL area also needs to pay attention to the type and degree of land use (Hu et al., 2020). The results of this study show that the land use intensity plays a dominant role in the spatial differentiation of soil conservation, water yield, and carbon sequestration, and it is also an extremely important factor in the spatial relationships between these three services. This result is consistent with that of Gao et al. (2020): that is, land use is the dominant factor affecting the spatial distribution of water yield in Beijing’s ECRL area. The results of the risk detector in GeoDetector revealed that the supplies of the three ESs were significantly higher in the forest land than in the other land use types, which confirms the importance of the One Million-Mu (666 km2) Plain Afforestation Project in Beijing. In addition, the mean soil conservation value of the cultivated land was the lowest (39.16 t ha−1 a−1), and was only 1/12 of the mean value of the forest land, indicating that Beijing’s ECRL area should pay more attention to optimization of soil conservation in the cultivated land.
As an important recharge factor for runoff and an erosive factor of soil, precipitation significantly affects the correlation between water yield and soil conservation (Peng et al., 2012). The partial correlation results show that precipitation was the most important factor controlling the spatial and temporal variations in water yield and soil conservation. Under the influence of precipitation, the correlation between these two services increased significantly and the proportion of the trade-off area changed significantly. Balthazar et al. (2015) found that an increase in the forest coverage in high-elevation mountainous areas will reduce the supply of water yield and increase the cumulative carbon sequestration. This conclusion confirms the results of this study. That is, compared with the net correlation between water yield and carbon sequestration, the trade-off between water yield and carbon sequestration under the influence of vegetation coverage (after removing the land use intensity) increased by 5.33% (Figure 4), indicating that vegetation coverage can inhibit the synergistic relationship between these two services.
The spatial and temporal changes in ESs were complex and diverse and were affected by multiple variables (Naidoo et al., 2008). Bennett et al. (2009) summarized them into six basic forms: environmental variables drive a single service, and there are three situations in which there is no/single role/interaction among services and three cases where services are affected by common drivers and there is no/single role/interaction among the services. Previous studies have shown that the trade-off relationships between soil conservation, water yield, and carbon sequestration in mountainous areas are sensitive to gradational changes in precipitation, vegetation coverage, and elevation (Liu et al., 2019). The spatial trade-off/ synergy relationships between the three services differed significantly among land use types such as forest land, wetland, and cultivated land (Tian et al., 2016). The above conclusions are consistent with the extremely important factors identified in this study, which confirms that these three ESs interacted with each other and were affected by common drivers.
By quantitatively separating and comparing the driving forces, it was found that different ESs responded differently to environmental variables and service variables. Soil conservation was more strongly driven by environmental variables, and the areas where elevation and slope were positively correlated with soil conservation accounted for more than 64% of the entire study area. These results are similar to previous research results, that is, the soil conservation capacity of the forest ecosystem in Beijing increases with increasing elevation and slope (Zhang et al., 2009). Owing to the decisive influence of precipitation on water yield and the significant influence of water and heat resources on the spatial distribution of vegetation (Sun et al., 2018), the effects of precipitation and carbon sequestration on temporal and spatial changes in water yield were similar. Previous research has shown that there is a significant positive correlation between vegetation coverage and carbon sequestration (Teng et al., 2022). The results of the correlation analysis conducted in this study also show that the mean value of the spatial correlation coefficient between vegetation coverage and carbon sequestration was the largest, and the positively correlated areas accounted for 68.8% of Beijing’s ECRL area. Compared with environmental variables, the correlations between the two service variables and carbon sequestration were stronger, mainly because soil conservation and water yield can provide decisive controlling conditions such as nutrients and water for vegetation growth and carbon sequestration (Gao and Zuo, 2021).
In this study, three typical ESs in Beijing’s ECRL area were considered, partial correlation analysis was conducted to eliminate the cross-influence of environmental variables on the trade-offs/synergies, and the impacts of environmental variables and service variables on ESs were quantitatively compared. The premise of partial correlation analysis is that when two variables are both correlated with a third variable, the influence of the third variable is eliminated, and thus, the net correlation between the other two variables can be analyzed (Kadić et al., 2018). This study was limited to this premise and only the influences of the extremely important factors on the trade-offs/synergies were separated. The transmission effect of the generally important factors of one service on another needs to be further studied. In addition, there is a situation in which one service acts as an intermediary variable and affects the other services, and the chain effect is also worthy of further exploration.

5 Conclusions

From the perspective of the spatial heterogeneity and spatial correlation of geographic data, in this study, we analyzed the spatial variations in the net trade-offs/synergies after removing the cross-influence of environmental variables. Then, the net trade-offs/synergies were compared to the extent to which environmental variables contributed to the changes in ESs. The results of this study demonstrate that the spatial variations in ESs were jointly affected by service variables and environmental variables, and the interaction between the two types of variables was stronger than the explanatory power of a single variable. The spatial and temporal changes in soil conservation were mainly driven by environmental variables, i.e., slope and precipitation. Precipitation and carbon sequestration had the strongest impacts on water yield, and these two variables had similar impacts. The changes in carbon sequestration were closely related to the other two services, and the effects of environmental variables were relatively small.
In addition to influencing the spatial distributions of ESs, environmental variables also affected the spatial variations in the trade-offs/synergies. Precipitation played a dominant role in the temporal and spatial changes in soil conservation and water yield. When the influence of precipitation was excluded, the proportion of the area of the trade-off relationship increased. The net correlation coefficient between water yield and carbon sequestration was nearly three times the two simple correlation coefficients, and affected by the vegetation coverage and land use, the trade-offs between water yield and carbon sequestration increased significantly. Controlling the elevation, slope, and land use intensity inhibits the trade-offs between soil conservation and carbon sequestration.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
Arnoldus H M J, 1980. An approximation of the rainfall factor in the universal soil loss equation. In: De Boodt M, Gabriels D. Assessment of Erosion. Chichester UK: Wiley, 127-132.

[2]
Bai Y, Jiang B, Wang M et al., 2016. New ecological redline policy (ERP) to secure ecosystem services in China. Land Use Policy, 55: 348-351. doi: 10.1016/j.landusepol.2015.09.002

DOI

[3]
Balthazar V, Vanacker V, Molina A et al., 2015. Impacts of forest cover change on ecosystem services in high Andean mountains. Ecological Indicators, 48: 63-75. doi: 10.1016/j.ecolind.2014.07.043

DOI

[4]
Bennett E M, Cramer W, Begossi A et al., 2015. Linking biodiversity, ecosystem services, and human well-being: Three challenges for designing research for sustainability. Current Opinion in Environmental Sustainability, 14: 76-85. doi: 10.1016/j.cosust.2015.03.007.

DOI

[5]
Bennett E M, Peterson G D, Gordon L J, 2009. Understanding relationships among multiple ecosystem services. Ecology Letters, 12: 1394-1404. doi: 10.1111/j.1461-0248.2009.01387.x.

DOI PMID

[6]
Briner S, Huber R, Bebi P et al., 2013. Trade-offs between ecosystem services in a mountain region. Ecology and Society, 18(3): 35. doi: 10.5751/ES-05576-180335.

DOI

[7]
Cai C F, Ding S W, Shi Z H et al., 2000. Study of applying USLE and geographical information system IDRISI to predict soil erosion in small watershed. Journal of Soil Water Conservation, 14(2): 19-24. (in Chinese)

[8]
Carpenter S R, Mooney H A, Agard J et al., 2009. Science for managing ecosystem services: Beyond the millennium ecosystem assessment. Proceedings of the National Academy of Sciences, 106(5): 1305-1312. doi: 10.1073/pnas.0808772106.

DOI

[9]
Chen T Q, Feng Z, Zhao H F et al., 2020. Identification of ecosystem service bundles and driving factors in Beijing and its surrounding areas. Science of the Total Environment, 711: 134687. doi: 10.1016/j.scitotenv.2019.134687.

DOI

[10]
China Council for International Cooperation on Environment and Development, Institutional Innovation of Eco-Environmental Redlining (CCICED), 2014. http://english.mee.gov.cn/Events/Special_Topics/AGM_1/AGM2014/download/201605/P020160524201838208313.pdf.

[11]
Chinese Ministry of Environmental Protection (CMEP), 2015. Technical Guidelines for the Delimitation of Ecological Conservation Redlines. Beijing, China. https://www.mee.gov.cn/gkml/hbb/bgt/201707/t20170728_418679.htm in Chinese)

[12]
Cord A F, Bartkowski B, Beckmann M et al., 2017. Towards systematic analyses of ecosystem service trade-offs and synergies: Main concepts, methods and the road ahead. Ecosystem Services, 28(3): 264-272. doi: 10.1016/j.ecoser.2017.07.012.

DOI

[13]
Costanza R, De Groot R, Braat L et al., 2017. Twenty years of ecosystem services: How far have we come and how far do we still need to go? Ecosystem Services, 28: 1-16. doi: 10.1016/j.ecoser.2017.09.008.

DOI

[14]
Dade M C, Mitchell M G E, Mcalpine C A et al., 2019. Assessing ecosystem service trade-offs and synergies: The need for a more mechanistic approach. AMBIO, 48: 1116-1128. doi: 10.1007/s13280-018-1127-7.

DOI

[15]
Feng Q, Zhao W W, Fu B J et al., 2017. Ecosystem service trade-offs and their influencing factors: A case study in the Loess Plateau of China. Science of the Total Environment, 607/608: 1250-1263. doi: 10.1016/j.scitotenv.2017.07.079.

DOI

[16]
Gao J B, Jiang Y, Wang H, et al., 2020. Identification of dominant factors affecting soil erosion and water yield within ecological red line areas. Remote Sensing, 12: 399. doi: 10.3390/rs12030399.

DOI

[17]
Gao J B, Zuo L Y, 2021. Revealing ecosystem services relationships and their driving factors for five basins of Beijing. Journal of Geographical Sciences, 31(1): 111-129. doi: 10.1007/s11442-021-1835-y.

DOI

[18]
Gao J B, Zuo L Y, Liu W L, 2021. Environmental determinants impacting the spatial heterogeneity of karst ecosystem services in Southwest China. Land Degradation and Development, 32: 1718-1731. doi: 10.1002/ldr.3815.

DOI

[19]
Gao J X, 2019. How China will protect one-quarter of its land. Nature, 569(7757): 457. doi: 10.1038/d41586-019-01563-2.

DOI

[20]
Gong M H, Fan Z Y, Wang J Y et al., 2017. Delineating the ecological conservation redline based on the persistence of key species: Giant pandas (Ailuropoda melanoleuca) inhabiting the Qinling Mountains. Ecological Modelling, 345: 56-62. doi: 10.1016/j.ecolmodel.2016.11.011.

DOI

[21]
Grêt-Regamey A, Weibel B, 2020. Global assessment of mountain ecosystem services using earth observation data. Ecosystem Services, 46: 101213. doi: 10.1016/j.ecoser.2020.101213.

DOI

[22]
Guerry A D, Polasky S, Lubchenco L et al., 2015. Natural capital and ecosystem services informing decisions: From promise to practice. Proceedings of the National Academy of Sciences, 112(24): 7348-7355. doi: 10.1073/PNAS.1503751112.

DOI

[23]
Howe C, Suich H, Van Gardingen P et al., 2013. Elucidating the pathways between climate change, ecosystem services and poverty alleviation. Current Opinion in Environmental Sustainability, 5(1): 102-107. doi: 10.1016/j.cosust.2013.02.004.

DOI

[24]
Hu T, Peng J, Liu Y X et al., 2020. Evidence of green space sparing to ecosystem function improvement in urban regions: A case study of China’s Ecological Red Line policy. Journal of Cleaner Production, 251: 119678. doi: 10.1016/j.jclepro.2019.119678.

DOI

[25]
Jiang B, Bai Y, Wong C P et al., 2019. China’s ecological civilization program: Implementing ecological redline policy. Land Use Policy, 81: 111-114. doi: 10.1016/j.landusepol.2018.10.031.

DOI

[26]
Kadić A, Denić-Jukić V, Jukić D, 2018. Revealing hydrological relations of adjacent karst springs by partial correlation analysis. Nordic Hydrology, 49(3): 616-633. doi: 10.2166/nh.2017.064.

DOI

[27]
Li Z H, Deng X Z, Jin G et al., 2020. Tradeoffs between agricultural production and ecosystem services: A case study in Zhangye, Northwest China. Science of The Total Environment, 707: 136032. doi: 10.1016/j.scitotenv.2019.136032.

DOI

[28]
Liu L B, Wang Z, Wang Y et al., 2019. Trade-off analyses of multiple mountain ecosystem services along elevation, vegetation cover and precipitation gradients: A case study in the Taihang Mountains. Ecological Indicators, 103: 94-104. doi: 10.1016/j.ecolind.2019.03.034.

DOI

[29]
Millennium Ecosystem Assessment (MEA), 2005. Ecosystems and Human Well-Being:Synthesis. Washington, DC: Island Press.

[30]
Mina M, Bugmann H, Cordonnier T et al., 2017. Future ecosystem services from European mountain forests under climate change. Journal of Applied Ecology, 54(2): 389-401. doi: 10.1111/1365-2664.12772.

DOI

[31]
Monteith J L, Moss C J, 1977. Climate and the efficiency of crop production in Britain. Philosophical Transactions of the Royal Society of London, B281(980): 277-294. doi: 10.1098/rstb.1977.0140.

DOI

[32]
Naidoo R, Balmford A, Costanza R et al., 2008. Global mapping of ecosystem services and conservation priorities. Proceedings of the National Academy of Sciences, 105(28): 9495-9500. doi: 10.1073/pnas.0707823105.

DOI

[33]
Naipal V, Reick C, Pongratz J et al., 2015. Improving the global applicability of the RUSLE model: Adjustment of the topographical and rainfall erosivity factors. Geoscientific Model Development, 8(9): 2893-2913. doi: 10.5194/gmd-8-2893-2015.

DOI

[34]
Ouyang Z Y, Zheng H W, Xiao Y et al., 2016. Improvements in ecosystem services from investments in natural capital. Science, 352(6292): 1455-1459. doi: 10.1126/science.aaf2295.

DOI PMID

[35]
Peng T, Wang S J, 2012. Effects of land use, land cover and rainfall regimes on the surface runoff and soil loss on karst slopes in southwest China. CATENA, 90: 53-62. doi: 10.1016/j.catena.2011.11.001.

DOI

[36]
Potter C S, Randerson J T, Field C B et al., 1993. Terrestrial ecosystem production: A process model based on global satellite and surface data. Global Biogeochemical Cycles, 7(4): 811-841. doi: 10.1029/93gb02725.

DOI

[37]
Qiu J, Turner M G, 2013. Spatial interactions among ecosystem services in an urbanizing agricultural watershed. Proceedings of the National Academy of Sciences, 110(29): 12149-12154. doi: 10.1073/pnas.131053911.

DOI

[38]
Renard K G, Foster G R, Weesies G A et al., 1997. Predicting soil erosion by water: A guide to conservation planning with the revised universal soil loss equation (RUSLE). Agriculture Handbook, 703: 27-28.

[39]
Rodriguez J P, Beard T D, Bennett E M et al., 2006. Trade-offs across space, time, and ecosystem services. Ecology and Society, 11(1): 28. doi: 10.5751/ES-01667-110128.

DOI

[40]
Sharp R, Tallis H T, Ricketts T et al., 2016. InVEST + VERSION + User’s Guide. The Natural Capital Project, Stanford University, University of Minnesota, The Nature Conservancy, and World Wildlife Fund. Stanford (CA): Stanford University.

[41]
State Council of China (SCC), 2018. Beijing’s Ecological Protection Redlines. Beijing, China. http://www.gov.cn/xinwen/2018-07/13/content_5306150.htm. in Chinese)

[42]
Sun X, Lu Z M, Li F et al., 2018. Analyzing spatio-temporal changes and trade-offs to support the supply of multiple ecosystem services in Beijing, China. Ecological Indicators, 94: 117-129. doi: 10.1016/j.ecolind.2018.06.049.

DOI

[43]
Teng Y M, Zhan J Y, Liu W et al., 2022. Spatial heterogeneity of ecosystem services trade-offs among ecosystem service bundles in an alpine mountainous region: A case-study in the Qilian Mountains, Northwest China. Land Degradation and Development, 33(11): 1846-1861. doi: 10.1002/ldr.4266.

DOI

[44]
Tian Y C, Wang S J, Bai X Y et al., 2016. Trade-offs among ecosystem services in a typical karst watershed, SW China. Science of the Total Environment, 566/567: 1297-1308. doi: 10.1016/j.scitotenv.2016.05.190.

DOI

[45]
Tomscha S A, Gergel S E, 2016. Ecosystem service trade-offs and synergies misunderstood without landscape history. Ecology and Society, 21(1): 43. doi: 10.5751/ES-08345-210143.

DOI

[46]
Wang H Y, Qin F, Xu C D et al., 2021. Evaluating the suitability of urban development land with a Geodetector. Ecological Indicators, 123:107339. doi: 10.1016/j.ecolind.2021.107339.

DOI

[47]
Wang J F, Li X H, Christakos G et al., 2010. Geographical detectors-based health risk assessment and its application in the neural tube defects study of the Heshun region, China. International Journal of Geographical Information Science, 24(1): 107-127. doi: 10.1080/13658810802443457.

DOI

[48]
Wang J F, Xu C D, 2017. Geodetector: Principle and prospective. Acta Geographica Sinica, 72(1): 116-134. doi: 10.11821/dlxb201701010. (in Chinese)

DOI

[49]
Wang J F, Zhang T L, Fu B J, 2016. A measure of spatial stratified heterogeneity. Ecological Indicator, 67: 250-256. doi: 10.1016/j.ecolind.2016.02.052.

DOI

[50]
Wang P T, Zhang L W, Li Y J et al., 2017. Spatio-temporal characteristics of the trade-off and synergy relationships among multiple ecosystem services in the upper reaches of Hanjiang River Basin. Acta Geographica Sinica, 72(11): 2064-2078. doi: 10.11821/dlxb201711011. (in Chinese)

DOI

[51]
Williams J R, Jones C A, Kiniry J R et al., 1989. The EPIC crop growth-model. Transactions of the ASAE, 1989, 32: 497-511. doi: 10.13031/2013.31032.

DOI

[52]
Wiśniewski P, Märker M, 2019. The role of soil-protecting forests in reducing soil erosion in young glacial landscapes of Northern-Central Poland. Geoderma, 337: 1227-1235. doi: 10.1016/j.geoderma.2018.11.035.

DOI

[53]
Xu X B, Tan Y, Yang G S et al., 2018. China’s ambitious ecological red lines. Land Use Policy, 79: 447-451. doi: 10.1016/j.landusepol.2018.08.037.

DOI

[54]
Xu Y, Tang H P, Wang B J et al., 2016. Effects of land-use intensity on ecosystem services and human well-being: A case study in Huailai County, China. Environment Earth Science, 75: 415-425. doi: 10.1007/s12665-015-5103-2.

DOI

[55]
Xu Y J, Yao Z H, Zhao D B, 2012. Estimating soil erosion in North China Plain based on RS/GIS and RUSLE. Bulletin of Soil and Water Conservation, 32(6): 217-220.

[56]
Yang J T, Song C, Yang Y et al., 2019. New method for landslide susceptibility mapping supported by spatial logistic regression and GeoDetector: A case study of Duwen Highway Basin, Sichuan Province, China. Geomorphology, 324: 62-71. doi: 10.1016/j.geomorph.2018.09.019.

DOI

[57]
Yin K, Tian Y C, Yuan C et al., 2015. NPP spatial and temporal pattern of vegetation in Beijing and its factor explanation based on CASA model. Remote Sensing for Land and Resources, 27(1): 133-139. doi: 10.6046/gtzyyg.2015.01.21. (in Chinese)

DOI

[58]
Yu Y Y, Li J, Zhou Z X et al., 2021. Response of multiple mountain ecosystem services on environmental gradients: How to respond, and where should be priority conservation? Journal of Cleaner Production, 278: 123264. doi: 10.1016/j.jclepro.2020.123264.

DOI

[59]
Zhang B, Li W H, Xie G D et al., 2009. Comprehensive assessment of soil conservation capacity of forest ecosystems in Beijing. Research of Soil and Water Conservation, 16(1): 240-244. (in Chinese)

[60]
Zhang Z Y, Liu Y F, Wang Y H et al., 2020. What factors affect the synergy and tradeoff between ecosystem services, and how, from a geospatial perspective? Journal of Cleaner Production, 257: 120454. doi: 10.1016/j.jclepro.2020.120454.

DOI

[61]
Zuo L Y, Gao J B, 2021. Investigating the compounding effects of environmental factors on ecosystem services relationships for the ecological conservation red line areas. Land Degradation and Development, 32: 4609-4623. doi: 10.22541/au.161844981.12324108/v1.

DOI

Options
Outlines

/

〈 〉