Variations in the Value and Trade-Offs/Synergies of Ecosystem Services on Topographic Gradients in Qinghai Province, China

Abstract

Qinghai Province is an important part of the Qinghai–Tibet Plateau. It is the birthplace of the Yangtze, Yellow, and Lancang (Mekong) Rivers, the recharge area for China’s freshwater resources. With different terrains, the temperature, daylight, and precipitation of Qinghai Province are widely variable. Consequently, the topography largely determines the spatial distribution of ecosystem services (ESs) and affects their interaction. Studying the impact of topography on the spatial-temporal evolution of ESs and their interaction is of great significance for land-use planning and the ecological civilization in Qinghai Province. To explore the spatial distribution and evolution characteristics of ESV and trade-offs/synergies among ESs in Qinghai Province, we considered topography (elevation, slope, RDLS, and terrain niche index) in 1980, 1990, 2000, 2010, and 2020 using the ArcGIS software and the equivalent factor method. The results were then corrected using various parameters. The results showed that the ESV of Qinghai Province decreased from 129,573.99 million USD in 1980 to 129,155.85 million USD in 2000, and then increased rapidly to 142,682.97 million USD in 2020. The spatial distribution of ESV is characterized by high in the south and east and low in the northwest. The geographical distribution and temporal variation of ESV and trade-offs/synergies of ES pairs show distinct vertical zonality, and the relationship between ESs showed different patterns on different topographic gradients. Hence, priority should be given to the ecological protection of high-altitude areas, and the implementation of ecological migration continued. The implementation of protection measures considering the ecological conditions under different topographic gradients can ensure more appropriate ecosystem management and more sensible decision-making.

1. Introduction

Ecosystem services (ESs) are the direct or indirect benefits provided by ecosystems to humans, thus serving as a link between human society and ecosystems [1,2]. With the rapid development of society, human beings have intensified various means of acquiring more benefits from ecosystems, such as excessive use of chemical fertilizers, deforestation, direct disposal of wastewater into rivers, and the uncontrolled exploitation of various resources. While changing the structure of land use, these activities also damage the structure and function of ESs, thus posing serious threats to human well-being. To ensure the sustainable provision of services by ecosystems, the international initiatives Millennium Ecosystem Assessment (MA) and the Economics of Ecosystems and Biodiversity (TEEB) divided ESs into four main categories: provisioning services (PS), regulation services (RS), habitat services (HS), and cultural and amenity services (CaS); they also quantified the benefits of these services to human beings in terms of economic value. The quantification of ecosystem services is an important research topic, it is helpful to provide important information reference for government managers in the process of planning and decision-making [3,4].

Land cover and land use constitute the two attributes of land. Land cover refers to the natural and biophysical properties of the earth’s surface, while land use is the way and condition in which humans use the land for their own purposes, based on the natural properties of the land. Land use/cover change (LUCC) is, therefore, a concentrated expression of the interaction between human activities and the natural environment [5,6]. As the most significant area of interaction between human society and the natural environment, the terrestrial ecosystem is strongly affected by land use and land cover change [7]. Evaluating ESV based on land use change is necessary. LUCC can alter the structure, distribution and processes of terrestrial ecosystems, thereby affecting their ability to provide benefits to human production and livelihoods [8,9]. With the rapid growth of the world’s population and socio-economic development, the excessive demand for ecosystem services has caused serious damage to ecosystems, resulting in the reduction and degradation of ecosystem services in many areas. Previous studies have shown that LUCC caused by human activities results in a 60% decline in the supply of ESs, which further leads to a series of environmental problems, such as soil erosion, salinization, and desertification [10,11]. Costanza et al. divided ESs into 17 types for the first time, proposed a table of ecosystem service value (ESV) coefficients for different land use types, and quantified ESs worldwide. Based on the research results of Costanza et al. and according to the actual situation of China, Xie et al. conducted a questionnaire survey and proposed that the equivalent value table of ecosystem services suitable for China [12]. Since then, more scholars have carried out quantitative research on ESs from the perspective of different disciplines, gradually propelling China to the frontier and hotspot of research [13].

Since the 21st century, many studies have investigated the impact of LUCC on ESV at the global [14,15], national [16,17,18,19], regional [20,21,22], and city scales [23,24,25]. Previous research mainly focused on four aspects. The first is the response of ESV to LUCC. For instance, Wei proposed an elastic model of the ESV response to LUCC and found that 1% of LUCC in China resulted in 0.15% and 0.1% average changes in ESV in 1988–2000 and 2000–2008, respectively. ESV was the most sensitive to LUCC in the Northeast Plain, the North China Plain, and the central part of Northwest China [16]. Jiang found that the expansion of water bodies and the reduction in bare land and glaciers in the Qinghai–Tibet Plateau from 1990 to 2015 led to an increase in the total ESV by 0.04 billion Yuan [26]. The second is the impact of LUCC on a single ecosystem or a single ecosystem service function, such as a farmland ecosystem, forest ecosystem [27], watershed ecosystem [28,29], biodiversity [30], and water production [31]. The third is to use relevant models to predict future LUCC under different scenarios in order to determine the future trend of ESV. For example, Xi used the FLUS model to predict the future land use structure in Zhoushan City and determined the future trend of ESV in Zhoushan City according to the prediction results [32]. The fourth is the study of ecosystem service trade-offs, including the identification of trade-offs/synergies among ESs, trends, differentiation characteristics, and driving mechanisms [33,34]. In addition, ESV and its trade-offs/synergies vary with spatial scales. In recent years, some research has explored the influence of topography on ESV and trade-offs/synergies among different topographic gradients. Wu et al. studied the effects of topography on ESs in the Qinghai–Tibet Plateau from four perspectives (elevation, slope, relief degree of land surface (RDLS), and terrain niche index (Tni)), and found that ESV decreased with increasing elevation gradient, and initially decreased but subsequently increased on the gradient of the slope, RDLS, and Tni [35]. Zhang et al. studied the trade-off/synergistic effects of a forest ecosystem in the Funiu Mountains from the perspective of slope and found that the synergistic effect of forest ESs was stronger on the south slope than on the north slope of the Funiu Mountains [36]. Li studied the role of ESs and mountain environmental factors in ES trade-offs/synergies and found that altitude had a stronger impact on synergies and that synergies between habitat quality and water yield decreased with increasing altitude [37]. Xu et al. found a geographical hierarchy in the spatial distribution patterns of ESs in the Bailongjiang Watershed [38]. The terrain is a significant factor affecting the gradient difference of regional ESs. However, only a few studies have investigated the impact of different topographic gradients on ESV and trade-offs/synergies comprehensively in provincial administrative units. Approximately 67% of China’s land area is mountainous, and about 56% of China’s population depends on mountain resources [39]. In order to optimize the land use structure, improve the utilization efficiency of land resources, develop appropriate environmental management strategies on a spatial scale, and maintain the stability and security of natural ecosystems, it is necessary to explore the distribution and variation of the value and relationship of ESs in mountainous areas from the perspective of different topographic gradients.

Qinghai Province is an important part of the Qinghai–Tibet Plateau. It is a significant source of freshwater resources in China and the most concentrated area of plateau biodiversity. Qinghai Province is also a sensitive area and an important initiation area of climate change in Asia, the Northern Hemisphere, and even the world. With its unique geographical location, rich natural resources, and important ecological functions, Qinghai Province is an important ecological security barrier in China. The eco-environment of highly sensitive and fragile regions is usually essential to human well-being and sustainable development. However, the eco-environment of such regions is extremely difficult to recover if destroyed. After the reform and opening up of China, and relying on the advantages of natural resources, the social economy of Qinghai Province developed rapidly and town areas expanded rapidly. However, the eco-environment was also damaged. Nevertheless, Qinghai Province is a complete administrative division with a complex topography and clear spatial differences in ESs. This study investigated the spatial distribution and evolution characteristics of ESV and its trade-offs/synergies relationships by selecting Qinghai Province as the study area. The specific objectives were to (1) reveal LUCC in Qinghai Province from 1980 to 2020; (2) explore the spatiotemporal distribution characteristics of ESV and topographic gradients; (3) observe the distribution and variation characteristics of ESV and trade-offs/synergies under different topographic gradients. Grids of 5 km × 5 km were selected as basic units to explore the relationship between ESs and the topographic gradients, including elevation, slope, RDLS, and Tni. The results can provide a basis for coordinating the relationship between environmental protection and social-economic development in Qinghai Province, and provide a reference for decision-makers to coordinate economic development and environmental protection and optimize the land use structure.

2. Materials and Methods

2.1. Study Area

Qinghai Province is located in western China and the northeastern Qinghai–Tibet Plateau. It has two prefecture-level cities, six ethnic autonomous prefectures, five county-level cities, and forty county-level administrative units. The total area is approximately 70 million ha. It has a complex topography with various types of landforms, and the overall terrain is high in the west, low in the east, high in the north and south, and low in the center (Figure 1). The Yangtze River, Yellow River, and Lancang (Mekong) River all originate from this province. The ecological strategic position of Qinghai province is extremely important. Benefitting from national strategies such as the Great Western Development and the ‘Belt and Road’ initiative, the social economy of Qinghai Province has developed rapidly and is moving towards a sustainable development direction, in which the social economy is coordinated with the eco-environment. In addition, the eco-environment of this province is fragile; LUCC is a widespread phenomenon under the strong pressure of climate change and human activities. Therefore, this area is an ideal research area.

Land use data from Qinghai Province for 1980, 1990, 2000, 2010, and 2020, with a spatial resolution of 30 m × 30 m, were collected from the Resource and Environmental Science and Data Center of the Chinese Academy of Sciences (http://www.resdc.cn/, accessed on 2 November 2021). The land use type was divided into seven categories: farmland, forest, grassland, water, build-up land, barren land, and glacier and snow. DEM data with a spatial resolution of 90 m × 90 m derived from the Resource and Environmental Science and Data Center of the Chinese Academy of Sciences were used to measure the topographic features. Agricultural statistical data were obtained from the Statistical Yearbook of Qinghai Province, Statistical Bulletin of National Economic and Social Development, and Compilation of Cost-benefit Data of National Agricultural Products.

2.2. Methods

2.2.1. Calculation of Ecosystem Service Values

In this study, ESV, including supply, regulation, support, and cultural services, was assessed on the basis of the method proposed by Xie et al. [12]. Moreover, the unit area value equivalent factor method was used, in which the economic value of an ESV equivalent factor is approximately equal to 1/7 of the average grain market value of the year. The value of ecosystem service function per unit area of various types of ecosystems was also combined; this value refers to the basic equivalent of ecosystem service function value per unit area proposed by Xie et al. [40]. This method has been widely used in ESV evaluation for different regions. The annual planting area, unit yield, and average price of main crops (wheat, corn, highland barley, beans, and potato) in Qinghai Province from 1980 to 2020 were selected. The economic value of grain crops per unit area of farmland ecosystem in Qinghai Province was calculated to be 158.81 USD/hm² using Equation (1), and the ESV equivalent table per unit area of Qinghai Province was obtained (

Table 1. Ecosystem service equivalent value per unit area of Qinghai province.

Ecosystem Classification	Provisioning Services			Regulating Services				Habitat Services			Cultural and Amenity Services
	FP	MP	WP	AR	CR	WT	WR	EP	SM	HS	CaS
Farmland	134.99	63.52	3.18	106.4	57.17	15.88	42.88	163.58	19.06	20.65	9.53
Forest	30.17	68.29	34.94	223.93	671.78	203.28	532.02	273.16	20.65	249.34	109.58
Grassland	34.94	52.41	28.59	181.05	479.62	158.81	350.98	220.75	17.47	201.69	88.94
Water	127.05	36.53	1316.56	122.29	363.68	881.41	16,237.07	147.7	11.12	404.97	300.16
Barren land	1.59	4.76	3.18	17.47	15.88	49.23	33.35	20.65	1.59	19.06	7.94
Glacier and snow	0	0	343.04	28.59	85.76	25.41	1132.34	0	0	1.59	14.29

Note: FP (food production); MP (materials production); WP (water production); AR (Air quality regulation); CR (climate regulation); WT (waste treatment); WR (regulation of water flows); EP (erosion prevention); SM (maintenance of soil fertility); HS (habitat services); CaS (Cultural and amenity services).

).

$$VC=\frac{1}{7}{\displaystyle \sum _{i=1}^m\frac{o_i{p}_i{q}_i}{M}}$$

(1)

where VC represents the economic value of grain crops per unit area of farmland in Qinghai Province, o_i represents the planting area of i grain crops, p_i represents the yield of i grain crops, q_i represents the average price of i grain crops, and M represents the total planting area of three grain crops.

The ESV of the study area will be affected by local economic factors, and it is thus necessary to select appropriate economic parameters to correct the calculation results. Residents’ willingness and ability to pay for local ecosystems will strongly affect the ESV of the study area [41,42,43]. Therefore, residents’ ability and willingness to pay were integrated into the calculation of ESV (Equations (2)–(5)).

$$ESV={\displaystyle \sum _{i=1}^n\left(V_i\times A_i\times C\right)}$$

(2)

$$C=P\left(\mathit{Ability} \mathit{to} \mathit{pay} \mathit{for} \mathit{ESV}\right)\times W\left(\mathit{Willingness} \mathit{to} \mathit{pay} \mathit{for} \mathit{ESV}\right)$$

(3)

where V_i and A_i represent the ESV equivalent (USD/hm²) and area (hm²) of land type i, respectively, and C is the correction coefficient.

$$P=\frac{GD{P}_{\mathit{sp}}}{GD{P}_{\mathit{cp}}}$$

(4)

$$W=\frac{2}{1+e^{-(\frac{1}{E{n}_u\times U+E{n}_r\times R}-2.5)}}$$

(5)

The level of social and economic development can affect residents’ payment capacity for ecosystems. The ratio of the per capita GDP of Qinghai Province in 2020 to the national per capita GDP was used as one of the correction coefficients. In addition, the structure of consumer spending can influence the willingness to pay for the ES. Using Engel’s coefficient of urban and rural residents in the study area and referring to previous studies, residents’ willingness to pay for ESs was calculated [44]. Product C of the two variables was taken as the correction coefficient of ESV. GDP_sp and GDP_cp represent the per capita GDP of Qinghai Province and China, respectively. Enu and Enr represent Engel’s coefficient of urban and rural areas in Qinghai Province, respectively. U and R represent the proportion of urban and rural populations in the total population of Qinghai Province in 2020.

2.2.2. Degree of Trade-Offs among Ecosystem Services

The interaction between the two ESs was analyzed. When people consume one or more ecosystem services, it has an impact on the provision of other ecosystem services, and the issue of trade-offs and synergies of ecosystem services arises [45]. A negative degree represents a trade-off in the ES pair, where the provision of one ES increases at the expense of the other ESs. A positive degree represents a synergy in the ES pair, where the provision of one ES enhances the servicing capacity of the other ES [46,47]. The calculation of the ES trade-off degree (ESTD) is as follows (Equation (6)):

$$EST{D}_{\mathit{ij}}=\frac{ES{V}_{\mathit{ib}}-ES{V}_{\mathit{ia}}}{ES{V}_{\mathit{jb}}-ES{V}_{\mathit{ja}}}$$

(6)

where ESTD_ij is the ESTD between ESs i and j; ESV_ib and ESV_ia is the ESV with i at time b and a, respectively and ESV_ja and ESV_ja are the ESVs with j at time b and a, respectively. The absolute ESTD represents the degree of change in ESs of type i compared with that of type j [48].

In order to visualize the trend of ESTD for different terrain gradients, this study used a cross-linked table of ESTD with increasing terrain gradient levels(Figure 2). Each horizontal row I–V represents five levels of topographic gradient and the color represents the ESTD value between a group of ESs, with the trade-off relationship in blue and the synergistic relationship in yellow, the darker the color representing the stronger the trade-off or synergistic relationship. When the terrain gradient level changes from IV to V, the color of the square below changes from yellow to blue, indicating a switch from synergy to trade-off between ESi and ESj, and vice versa; The same color filling on the vertical indicates the same relationship between ESi and ESj for the same terrain gradient category.

3. Results

3.1. LUCC Patterns of Qinghai Province

As shown in Figure 3 and

Table 2. Changes in LUCC during 1980–2020 (million ha).

Land Use Type	1980	1990	2000	2010	2020	Proportion	Change Rate 1980–2020	Change Rate 1980–2000	Change Rate 2000–2020
Farmland	0.79	0.8	0.82	0.86	0.86	1.18%	0.24%	0.25%	0.22%
Forest	2.87	2.87	2.87	2.86	2.85	4.11%	−0.02%	−0.0014%	−0.03%
Grassland	37.31	37.28	37.23	39.24	39.14	54.6%	0.12%	−0.01%	0.26%
Water	2.32	2.32	2.31	2.52	2.82	3.53%	0.53%	−0.03%	1.11%
Build-up land	0.09	0.09	0.1	0.16	0.16	0.18%	2.22%	0.91%	2.98%
Barren land	25.87	25.88	25.91	23.59	23.38	35.78%	−0.24%	0.01%	−0.49%
Glacier and snow	0.42	0.42	0.42	0.45	0.45	0.62%	0.19%	0.0001%	0.37%

, grassland and barren land appear to be the main land use types in Qinghai Province from 1980 to 2020. Grassland was widely distributed in most parts of Qinghai Province (54.6% of the total area), with an annual growth rate of 0.12% throughout the study period, and the area decreased from 37.31 million ha in 1980 to 37.23 million ha in 2000 (−0.01%) and then increased rapidly to 39.14 million ha in 2020 (0.26%). Significant growth of grassland cover in central and south-western Qinghai Province.

Barren land was mainly distributed in the Qaidam Basin (35.78% of the total area), with an annual growth rate of −0.24% throughout the study period, and the area increased from 25.78 million ha in 1980 to 25.91 million ha in 2000 (0.01%) and then decreased to 23.38 million ha in 2020 (−0.49%). Overall, the area of barren land tends to shrink towards the northwest.

Water and glacier areas exhibited a scattered distribution (3.53% and 0.62% of the total area), and an annual growth rate of 0.53% and 0.19% throughout the study period, respectively. Water areas rapidly increased at an annual rate of 1.11% in 2000–2020, with spatially significant growth in the western.

Farmland (1.18%) and build-up land (0.18%) were distributed in the eastern region of Qinghai Province, which is highly associated with low elevation and the concentration of population. Farmland showed steady growth with an annual growth rate of 0.24% from 1980 to 2020. Areas of build-up land showed the most rapid expansion in 1980–2020, it increased from 0.09 million ha in 1980 to 0.16 million ha in 2020, with an annual growth rate of 2.22%.

It should be noted that the area of forest land in Qinghai Province decreased during the study period. Forest was mainly located in the northeastern Qilian Mountains. Areas of forest decreased from 2.87 million ha in 1980 to 2.85 million ha in 2020, with an annual growth rate of −0.02%.

Clear vertical spatial heterogeneity was found regarding the proportion and change of each land use type from 1980 to 2020 at elevation intervals of 500 m (Figure 4). Farmland was mainly distributed in areas below 3500 m. During the study period, farmland below 3500 m continued to decrease, while farmland within 3500–4500 m increased. This is mainly due to the urban encroachment on farmland below 3500 m altitude during the study period and the growth of population in the 3500–4500 m altitude area resulted in farmland reclamation. Forest land was mainly distributed within 3000–5000 m, it increased within 4000–5000 m (it was associated with afforestation and ecological migration policies), but decreased below 2500 m (population growth increased the demand for the supply of raw materials from ecosystems). Grassland was widely distributed, but decreased within 5000–6000 m, it was associated with climate change and overgrazing. Water was mainly distributed within 3000–5500 m, and its area continued to increase. Build-up land was mainly distributed below 4500 m, and the growth rate was the fastest among all types from 1980 to 2020, with an average annual growth rate of 8% within 4000–4500 m, it had increased in the same elevation range as the previously mentioned farmland, population growth and the relocation of industry to the west both require large areas of farmland and build-up land. Barren land was mainly distributed within 2500–6000 m, and its area shrunk within 2500–5000 m but expanded below 2000 m and above 5500 m. Glacier and snow areas were mainly distributed above 4000 m, and they increased above 5000 m during the study period, However, the impact of global warming has led to a significant reduction in the glacier and snow cover at altitudes of 3500–4500 m.

3.2. Spatial Pattern of ESV in Qinghai Province

The ESVs of Qinghai Province in 1980, 1990, 2000, 2010, and 2020 were 129,573.99 million USD, 129,509.29 million USD, 129,155.85 million USD, 136,829.78 million USD, and 142,682.97 million USD, respectively, with an overall increase of 13,108.98 million USD and an average annual increase of 0.25%. Overall, ESV showed a trend of an initial decrease, followed by an increase. Grassland showed the largest contribution to ESV, with an average proportion of 53.7%, followed by water, with an average proportion of 39.6%. From 1980 to 2020, the contributions of grassland and water increased by 4.9% and 21.3% (the most rapid), respectively. The average contributions of forest and barren land to total ESV were 5.4% and 3.4%, respectively, their values decreased by 0.7% and 9.6%, respectively, during the study period. In contrast, the average contributions of farmland and glacier and snow cover to total ESV were less than 1%, but their values increased by 9.6% and 7.4%, respectively, during the study period.

The average ESV was found to decrease gradually from south and east to northwest (Figure 5). The spatial distribution of ESV in Qinghai Province was high in the east and south, and low in the northwest. The high ESV was mainly distributed around Qinghai Lake and the Sanjiangyuan Region, and the low ESV was mainly distributed in the Qaidam Basin and its surrounding regions. During the study period, the high ESV in Qinghai province increased significantly in the central part of the Qaidam Basin and the western part of the Sanjiangyuan Region, and the low-value area decreased significantly in the eastern and southern marginal areas of the Qaidam Basin.

3.3. Topographic Gradient Features of Qinghai Province

Figure 6 shows the spatial variations in the topography of Qinghai Province. The average elevation, slope, RDLS, and Tni of Qinghai Province were calculated according to 5 × 5 km grid scales (28,606 grids in total), and each topographic index was divided into five categories using the natural breaks method. The lowest and highest altitudes of Qinghai Province were found to be 1740 m and 6207 m, respectively. The altitude gradually increases from east to west. The lowest slope is 0°, and the highest is 37.44°. Areas with a gentle slope (≤7.94°) are the most abundant and mainly distributed in Qinghai Lake, the Qaidam Basin, and the Sanjiangyuan Region. Areas with steep slopes (≥17.4°) are mainly distributed in the Qilian Mountains, Kunlun Mountains, and Qinghai Plateau. The spatial distribution of RDLS is similar to that of a slope. Areas with high Tni (>1.45) are the most widely distributed in Qinghai Province. In general, the terrain characteristics of Qinghai province are widely variable. Overall, most of the regions have high altitudes with a large drop. The eastern, southern, and central regions feature a steep relief, and most of the western regions are relatively flat.

3.4. Topographic Gradient Effect on the ESV of Qinghai Province

3.4.1. ESV on Different Topographic Gradients

The natural breaks classification method was selected to divide the topographic data into five classes, and the changes in ESV for different topographic gradients were analyzed (Figure 7). In terms of elevation gradient, the ESV of classes I to IV increased, and that of IV to V gradually decreased in 1980, 1990, 2000, 2010, and 2020. From 1980 to 1990 (Figure 7a), the ESV of classes I, III, and IV increased by 183.16 million USD, 0.61 million USD and 9.91 million USD, respectively, whereas that of classes II and V decreased by 103.25 million USD, and 155.13 million USD, respectively. From 1990 to 2000 (Figure 7b), the ESV of class III increased by 15.79 million USD, whereas that of classes I to V, except III, decreased by 86.67 million USD, 63.27 million USD, 184.49 million USD, and 34.81 million USD, respectively. From 2000 to 2010 (Figure 7c), the ESV of classes I to V increased by 1550.63 million USD, 597.11 million USD, 1385.85 million USD, 3082.4 million USD, and 1057.94 million USD, respectively. Similar to 2000–2010, from 2010 to 2020 (Figure 7c), the ESV of classes I to V increased by 3478.01 million USD, 493.13 million USD, 185.33 million USD, 622.26 million USD, and 1074.45 million USD, respectively.

In terms of slope gradient, the ESV of classes I to V gradually decreased in 1980, 1990, 2000, 2010, and 2020. From 1980 to 1990, the ESV of classes II and III increased by 11.22 million USD and 23.52 million USD, respectively, whereas that of classes I, IV, and V decreased by 90.77 million USD, 4.77 million USD, and 3.9 million USD, respectively. From 1990 to 2000, the ESV of classes III and IV increased by 9.05 million USD and 19.56 million USD, whereas that of classes I, II, and V decreased by 235.95 million USD, 144.98 million USD, and 1.12 million USD, respectively. From 2000 to 2010, the ESV of classes I to V increased by 3648.86 million USD, 2169.15 million USD, 1067.01 million USD, 576.9 million USD, and 212.02 million USD, respectively. From 2010 to 2020, the ESV of classes I to V increased by 5123.62 million USD, 617.05 million USD, 70.8 million USD, 28.9 million USD, and 12.81 million USD, respectively.

In terms of RDLS, the ESV of classes I to V gradually decreased in 1980, 1990, 2000, 2010, and 2020, similar to that based on slope. From 1980 to 1990, the ESV of classes II and III increased by 41.87 million USD and 100.68 million USD, respectively, whereas that of classes I, IV, and V decreased by 191.64 million USD, 11.3 million USD, and 4.32 million USD, respectively. From 1990 to 2000, the ESV of classes III and IV increased by 38.56 million USD and 7.33 million USD, whereas that of classes I, II, and V decreased by 343.44 million USD, 54.87 million USD, and 1.03 million USD, respectively. From 2000 to 2010, the ESV of classes I to V increased by 3503.48 million USD, 2161.96 million USD, 1222.11 million USD, 562.46 million USD, and 223.93 million USD, respectively. From 2010 to 2020, the ESV of classes I to V increased by 5152.11 million USD, 383.37 million USD, 82.82 million USD, 20.72 million USD, and 14.17 million USD, respectively.

In terms of Tni, the ESV of classes I to III showed little change, and that of classes III to V gradually decreased in 1980, 1990, 2000, 2010, and 2020. From 1980 to 1990, the ESV of classes I and III increased by 3.45 million USD and 7.64 million USD, respectively, whereas that of classes II, IV, and V decreased by 65.72 million USD, 3.21 million USD, and 6.86 million USD, respectively. From 1990 to 2000, the ESV of class V increased by 0.44 million USD, whereas that of classes I to IV decreased by 182.34 million USD, 59.93 million USD, 98.95 million USD, and 12.67 million USD, respectively. From 2000 to 2010, the ESV of classes I to V increased by 1948.5 million USD, 2325.5 million USD, 1902.28 million USD, 1104.44 million USD, and 393.2 million USD, respectively. From 2010 to 2020, the ESV of classes I to V increased by 4069.29 million USD, 1330.33 million USD, 392.91 million USD, 51.64 million USD, and 9.03 million USD, respectively.

3.4.2. Structure of ESV on Different Topographic Gradients

Figure 8 shows changes in the proportions of PS, RS, HS, and CS in the average ESV and their structures under different topographic gradients from 1980 to 2020. In terms of elevation gradient, PS, RS, HS, and CS in classes I to IV gradually increased with increasing elevation, decreasing by 1452.03 million USD, 18,642.15 million USD, 2024.78 million USD, and 930.67 million USD, respectively. In classes IV and V, PS, RS, HS, and CS decreased by 310.88 million USD, 5044.58 million USD, 1153.97 million USD, and 481.38 million USD, respectively. In terms of slope gradient, PS, RS, HS and CS in classes I to V decreased with increasing slope gradient, decreasing by 2892.52 million USD, 34,543.54 million USD, 1577.18 million USD, and 884.49 million USD, respectively. In terms of RDLS, the variation characteristics were similar to those of slope, decreasing by 3155.69 million USD, 38,024.8 million USD, 2055.89 million USD, and 1095.24 million USD, respectively. In terms of Tni, PS and RS in classes I to V decreased with increasing Tni, decreasing by 940.93 million USD and 11,492.2 million USD, respectively. HS and CS in classes I to V increased with increasing Tni, increasing by 474.69 million USD and 89.25 million USD, respectively.

3.5. Ecosystem Service Trade-Offs/Synergies Degree on Different Topographic Gradients

According to the previous calculation results, the ESV of Qinghai Province showed a downward trend from 1980 to 2000 and an increasing trend from 2000 to 2020. Therefore, changes in the trade-offs and synergies on different topographic gradients were analyzed by dividing the study period into two periods: 1980–2000 and 2000–2020. There were 100 groups of trade-offs/synergies between all ESs. Each group was divided into classes I–V according to the above elevation, slope, RDLS, and Tni, and changes in each group were analyzed according to different topographic gradients. Considering the necessity of analyzing the variation characteristics of the trade-offs/synergies on different topographic gradients individually, the variation characteristics of the trade-off/synergistic effects of ES pairs on different topographic gradients can be observed more intuitively in the form of a cross-column list, compared with the traditional transfer matrix. ESTD on different topographic gradients from 1980 to 2020 is listed in

Table 3. Ecosystem services trade-off degrees of different terrain gradients in Qinghai province from 1980 to 2000.

	Elevation
	FP					MP					WP					AR					CR					WT					WR					EP					SM					HS					CaS
FP
MP
WP
AR
CR
WT
WR
EP
SM
HS
CaS
	Slope
	FP					MP					WP					AR					CR					WT					WR					EP					SM					HS					CaS
FP
MP
WP
AR
CR
WT
WR
EP
SM
HS
CaS
	RDLS
	FP					MP					WP					AR					CR					WT					WR					EP					SM					HS					CaS
FP
MP
WP
AR
CR
WT
WR
EP
SM
HS
CaS
	Terrain niche index
	FP					MP					WP					AR					CR					WT					WR					EP					SM					HS					CaS
FP
MP
WP
AR
CR
WT
WR
EP
SM
HS
CaS
		Synergistic								10~50							trade-off								−10~−50
				0~10						>50									0~−10						<−50

Note: FP (food production); MP (materials production); WP (water production); AR (Air quality regulation); CR (climate regulation); WT (waste treatment); WR (regulation of water flows); EP (erosion prevention); SM (maintenance of soil fertility); HS (habitat services); CaS (Cultural and amenity services).

,

Table 4. ESTD variation between different terrain gradients from 1980 to 2000.

Type		Elevation	Slope	RDLS	Tni
Synergy–trade-off	I–II	6	0	45	0
	II–III	32	56	30	0
	III–IV	0	0	18	19
	IV–V	20	6	4	0
	Total	58	62	97	19
trade-off–synergy	I–II	42	20	0	20
	II–III	4	0	18	0
	III–IV	48	0	30	0
	IV–V	0	42	32	0
	Total	94	62	80	20

,

Table 5. Ecosystem services trade-off degrees of different terrain gradients in Qinghai province from 2000 to 2020.

	Elevation
	FP					MP					WP					AR					CR					WT					WR					EP					SM					HS					CaS
FP
MP
WP
AR
CR
WT
WR
EP
SM
HS
CaS
	Slope
	FP					MP					WP					AR					CR					WT					WR					EP					SM					HS					CaS
FP
MP
WP
AR
CR
WT
WR
EP
SM
HS
CaS
	RDLS
	FP					MP					WP					AR					CR					WT					WR					EP					SM					HS					CaS
FP
MP
WP
AR
CR
WT
WR
EP
SM
HS
CaS
	Terrain niche index
	FP					MP					WP					AR					CR					WT					WR					EP					SM					HS					CaS
FP
MP
WP
AR
CR
WT
WR
EP
SM
HS
CaS
		Synergistic								10~50							trade-off								−10~−50
				0~10						>50									0~−10						<−50

Note: FP(food production); MP(materials production); WP(water production); AR(Air quality regulation); CR(climate regulation); WT(waste treatment); WR(regulation of water flows); EP(erosion prevention); SM(maintenance of soil fertility); HS(habitat services); CaS(Cultural and amenity services).

and

Table 6. ESTD variation between different terrain gradients from 2000 to 2020.

			I–II	II–III	III–IV	IV–V				I–II	II–III	III–IV	IV–V
Elevation	➕	0–2	53	48	55	28	RDLS	➕	0–2	49	53	55	42
	➕	2–10	2	7	0	15		➕	2–10	6	0	0	9
	➕	>10	0	0	0	12		➕	>10	0	0	0	3
	➖	0–2	19	42	49	49		➖	0–2	26	51	53	54
	➖	2–10	18	6	5	6		➖	2–10	14	5	2	2
	➖	>10	18	7	1	0		➖	>10	15	1	0	0
			I–II	II–III	III–IV	IV–V				I–II	II–III	III–IV	IV–V
Slope	➕	0–2	44	44	48	47	Terrain niche index	➕	0–2	51	52	53	48
	➕	2–10	11	11	5	7		➕	2–10	4	3	1	5
	➕	>10	0	0	1	0		➕	>10	0	0	0	1
	➖	0–2	28	44	55	55		➖	0–2	22	36	48	53
	➖	2–10	13	9	1	1		➖	2–10	14	7	5	3
	➖	>10	14	2	0	0		➖	>10	19	12	3	0

Note: “➕” represents synergistic growth; “➖” represents synergistic reduction.

. In terms of elevation, slope, RDLS, and Tni, 550 groups of synergies were observed among ESs. The five grids in each group from left to right represent the ESTD in the topographic gradients of classes I, II, III, IV, and V. Cells shaded in yellow indicate positive ESTD values (synergies), whereas those in blue indicate negative ESTD values (trade-offs). The deeper the shade of the two colors, the greater the ESTD value, and the stronger the synergies/trade-offs.

From 1980 to 2000, in terms of elevation gradient, 406 groups were positive, 144 groups were negative, and classes I, II, III, IV, and V accounted for 13.3%, 22.2%, 15.3%, 27.1% and 22.2% of the synergies, respectively. The synergy degree between materials production and regulation of water flows in class II was the highest, and the trade-off degree between the maintenance of soil fertility and regulation of water flows in class V was the lowest. Classes I and III accounted for 72.2% of the trade-offs, and only the maintenance of soil fertility had the highest trade-off with other ESs in class V. Notably, trade-offs between food production and other ESs were concentrated in classes I, II, and III. In terms of elevation changes in ES synergies/trade-offs (Table 4), 32 groups of ESs in classes II–III and 20 groups in classes IV–V exhibited a translation from synergy to trade-off. In contrast, 42 groups in classes I–II and 48 groups in classes III–IV exhibited a transition of ESs from trade-offs to synergies. Among them, the synergies of water production with most of ES changed frequently, from synergies to trade-offs in classes II and III and from trade-offs to synergies in classes III and IV. All ESs and the maintenance of soil fertility shifted from synergies to trade-offs in classes IV–V.

In terms of slope gradient, 398 groups showed positive values, and 152 groups showed negative values. Classes I, II, III, IV, and V accounted for 22.6%, 27.6%, 13.6%, and 22.6% of the synergies, respectively, whereas classes III and IV accounted for 73.6% of the trade-offs. The synergy degree between the maintenance of soil fertility and the regulation of water flows in class II was the highest, and the trade-off degree between food production and the regulation of water flows in class I was the lowest. It is worth noting that the trade-offs between food production and other ESs were concentrated in classes I, III, IV, and V. From the perspective of a slope, 56 groups of ESs in classes II–III changed from synergies to trade-offs. In contrast, 20 and 42 groups of ESs in classes I–II and IV–V, respectively, changed from trade-offs to synergies. Among them, only food production showed synergies with other ESs in class II, and trade-offs with other ESs in classes I, III, IV, and V (except water production, waste treatment, and regulation of water flows).

In terms of RDLS, 374 groups showed positive values and 176 groups showed negative values. Classes I, II, III, IV, and V accounted for 29.4%, 16.6%, 13.4%, 16.6%, and 24.1% of the synergies, respectively, and classes II, III, and IV accounted for 27.3%, 34.1%, and 27.3% of the trade-offs, respectively. The synergy degree between the maintenance of soil fertility and regulation of water flows in class I was the highest, and the trade-off degree between the maintenance of soil fertility and regulation of water flows in class III was the lowest. In addition, trade-offs between food production and other ESs were concentrated in classes II, III, IV, and V. With increases in RDLS, the number of ES pairs transforming from synergies to trade-offs decreased, and the number of ES pairs transforming from trade-offs to synergies gradually increased.

In terms of Tni, 490 groups showed positive values and 60 groups showed negative values. it can be found that Classes I, II, III, IV, and V accounted for 18.4%, 22.4%, 22.4%, 22.4%, and 18.4% of the synergies, respectively. In contrast, classes I, IV, and V equally contributed (33.3%) to the trade-offs. The synergy degree between the maintenance of soil fertility and regulation of water flows in class III was the highest, and the trade-off degree between food production and regulation of water flows in class IV was the lowest. It is worth noting that the trade-offs between food production and other ESs were concentrated in classes I, IV, and V. From the perspective of Tni, 19 groups of ES pairs in classes III and IV shifted from synergies to trade-offs, and 20 groups in classes I and II shifted from trade-offs to synergies.

From 2000 to 2020, all ES pairs in Qinghai Province exhibited synergies. Among them, the maintenance of soil fertility showed the highest degree of synergy and regulation of water flows showed the lowest degree of synergy. Compared with 1980–2000, the synergy degree of ESs increased significantly, and the absolute value of ESTD increased. Irrespective of elevation, slope, RDLS, or Tni, the synergies between maintenance of soil fertility and regulation of water flow in class I was the largest. Changes in ESs synergies were further analyzed (Table 6). First, the degree of synergy of ESs pairs under different topographic gradients was relatively stable, compared with 1980–2000, and no ES pairs shifted from synergies to trade-offs. Second, the degree of synergy among ESs pairs increased and decreased at the same time from 2000 to 2020. The increments were generally 0–2, and the decrements were greater than 2 in classes I and II under all topographic gradients, indicating that the degree of the synergy of ESs pairs in these classes will decrease significantly under different topographic gradients. Third, the maintenance of soil fertility consistently showed the highest degree of synergy with other ESs, and food production, materials production, water production, maintenance of soil fertility, and cultural and amenity services showed the highest synergies with the regulation of water flows. At the same time, the degree of synergies decreased with increases in topographic gradient in classes I–IV. However, the synergies of food production, materials production, maintenance of soil fertility, and regulation of water flow in class V rebounded.

4. Discussion

4.1. Impact of Topographic Gradient on ESV

The ESV of Qinghai Province decreased from 129,573.99 million USD in 1980 to 129,155.85 million USD in 2000 and then increased rapidly to 142,682.97 million USD in 2020. The trend of ESV change is similar to the results of Han et al. and Wu et al. [37,49]. The research data selected for this paper have been used many times in previous studies and have a high degree of credibility and accuracy [35,50,51]. Overall, the average value of Provisioning services, regulating services, habitat Services, and cultural and amenity services in Qinghai Province gradually increased with increasing elevation gradient but decreased in areas above 4758 m. This phenomenon can be explained by the following: First, the spatial distribution of vegetation in Qinghai Province has obvious vertical zoning characteristics. In recent years, influenced by climate change and the policy of returning farmland to forest and grassland, vegetation in most areas has been restored [52,53,54]. The area of grassland continues to grow in regions of 4000–5000 m altitude with the fastest growth in grassland areas, especially at around 4500 m altitude (Figure 4). Second, under the impact of global warming, glacier melting and precipitation have increased, resulting in the expansion of water areas. The area of water is growing faster in areas below 5000 m altitude above sea level. Among them, the water area under class I increased by 114.63% in 2020, compared with 1980, which may be the main reason for the growth of ESV [55,56,57,58,59]. Third, the third-line construction ended in 1980. Affected by the market policy, a large number of local industrial enterprises in Qinghai Province shifted from decentralized distribution to integration and agglomeration distribution. Therefore, some industrial land in counties was abandoned and converted into grassland and barren land. For example, in the eastern part of the Datong Tuzu Hui Autonomous County and Xining city, many industrial enterprises distributed there in the early stage moved away after the Reform and Opening up, and the abandoned factory area was gradually covered by vegetation after decades. The area of build-up land increased by 198.54% in elevation class III, indicating that human activities in this area were very strong during the study period. Nevertheless, the final ESV growth in this area may be related to the increase in grassland and water area, and the ESV in this area increased the least among all elevation gradients. Although glacial and snow land decreased by 72.36% in elevation class III, the area of glacial snow cover in this region was small, and therefore, had little effect on ESV. Grassland decreased above 5500 m altitude and barren land began to increase, which is the main reason ESV decreases in elevation class V. It was mainly related to the degradation of grassland caused by overgrazing and the melting of glaciers under global warming, giving rise to barren land.

The distribution of the average ESV is similar for the slope and RDLS gradients (Figure 5). Regarding the average ESV for the slope and RDLS gradients in classes I to V during 1980–2020, the greater the slope and RDLS, the smaller the ESV. The ESV of Qinghai Province was mainly from class I in terms of slope and RDLS, including ecological reserves around Qinghai Lake, the Gonghe Basin, the Qaidam Basin, and the Sanjiangyuan Region. Therefore, class I has the largest ESV. During the first two decades of the study period (1980–2000), the ESV of class I in terms of slope and RDLS decreased the most (Figure 7). During this period, due to the increase in average temperature, drought, and soil erosion in the Qinghai–Tibet Plateau, large-scale grassland degradation occurred in the Sanjiangyuan Region of Qinghai Province and the eastern Qaidam Basin (Figure 2a–c), and the decrease in forest land area may be the cause of the decrease in ESV in class I. From 2000 to 2020, the grassland of class I in terms of slope and RDLS gradually recovered (Figure 2d,e). In particular, the grassland area in the southern Qinghai Plateau and the eastern Qaidam Basin recovered significantly, resulting in a significant ESV increase.

Under the Tni gradient, the average ESV slightly varied between classes I, II, and III, and decreased from class III to V. ESV showed a wave-like growth trend with the increase in Tni from 1980 to 2000 and gradually decreased with the increase in Tni from 2000 to 2020. Grassland and forest land, with high regulating services, in class I decreased significantly from 1980 to 2020, farmland and water, with high provisioning and regulating services, in classes I and II increased significantly (

Table 7. Changes in land use type area on topographic gradient during 1980–2020.

Terrain Gradient		Farmland	Forest	Grassland	Water	Build-Up Land	Barren Land	Glacier and Snow
Elevation	I	−0.35%	−0.95%	−1.05%	114.63%	82.53%	−3.82%	0
	II	36.11%	−0.64%	2.06%	6.84%	76.77%	−5.3%	0
	III	11.68%	−0.54%	6.44%	12.83%	198.54%	−12.54%	−72.36%
	IV	7.38%	−0.8%	9.11%	13.68%	87.97%	−20.11%	46.24%
	V	0	0.16%	1.15%	12.36%	9119.23%	−4.49%	6.55%
Slope	I	32.92%	−2.23%	0.81%	26.05%	167.64%	−6.3%	3.09%
	II	5.71%	−3.03%	7.46%	12.25%	72.17%	−13.07%	−0.12%
	III	−0.3%	−1.33%	7.21%	7.45%	32.86%	−14.71%	2.02%
	IV	0.23%	−0.39%	3.96%	16.6%	31.85%	−8.33%	5%
	V	1.44%	0.02%	2.79%	6.01%	15.78%	−5.25%	26.36%
RDLS	I	37.06%	−2.22%	0.98%	25.57%	173.43%	−5.78%	4.19%
	II	3.76%	−3.12%	8.04%	11.03%	68.62%	−15.95%	−3.53%
	III	1.69%	−0.8%	5.84%	15.4%	51.61%	−12.81%	−2.15%
	IV	2.13%	−0.1%	3.77%	11.68%	21.25%	7.54%	5.21%
	V	−0.08%	−0.24%	3.63%	17.97%	14.07%	−5.81%	25.67%
Terrain niche index	I	26.84%	−2.27%	−4.10%	28.35%	136.12%	−4.18%	0
	II	6%	−2.92%	4.90%	23.03%	68.56%	−9.04%	−18.5%
	III	0.49%	−1.91%	8.05%	9.2%	41.50%	−16.78%	0.98%
	IV	1.75%	−0.35%	6.32%	6.25%	57.76%	−13.87%	2.83%
	V	5.28%	−0.14%	2.49%	8.11%	24.60%	−4.99%	8.94%

). This may be the main reason for the decrease in ESV in class I from 1980 to 2000 and the significant increase from 2000 to 2020.

It is worth noting that, under the elevation, slope, RDLS, and Tni gradients, the expansion rate of the area of build-up land in classes I, II, III, IV, and V from 1980 to 2020 was the highest. Specifically, in terms of elevation, the area of build-up land in class V increased by 9119.23% in 2020, compared with that in 1980. Qinghai Province implemented the relocation of most villages in accordance with ecological protection policies after 2000, and a large number of herdsmen entered and settled in the county. The urbanization of the Yushu Tibetan Autonomous Prefecture developed rapidly and this prefecture has become an important node city for high-quality urbanization in the Qinghai–Tibet Plateau [60]. Similarly, the area of build-up land in class I in terms of slope, RDLS, and Tni increased the fastest, and the ESV of class I also increased significantly during the study period. This suggests that the ecological relocation in Qinghai Province not only effectively controls the destruction of grassland caused by overgrazing due to population growth, but also contributes to the restoration and protection of the local eco-environment and reservation of sufficient space for the restoration of the eco-environment.

4.2. Impact of Topographic Gradient on ES Trade-Offs and Synergies

The spatial patterns of synergies and trade-offs among ESs in Qinghai Province in terms of various topographic gradients were found to be heterogeneous and varied depending on ES pairs. Many researchers have studied the trade-off and synergy effects of ESs at the national, city, and provincial scales [61,62,63,64]. The results of this study support that previous studies have found differences in trade-offs and synergies between the same pair of ESs in different spaces [65,66,67], and changes with spatial scale [68,69,70]. Identifying the changes in ESs at different spatial scales and their interrelationships has enormous benefits for realizing regional ecological civilization and environmental management.

The trade-offs/synergies of ES pairs in terms of different topographic gradients in Qinghai Province were found to significantly vary, with distinct vertical differentiation characteristics. This variation is primarily because ES interactions in different topographic gradients are influenced by a combination of drivers including climate, vegetation types, land use, and biodiversity. The findings are similar to Nepal and the Eastern Margin, which are also located on the Qinghai-Tibet Plateau. The interrelationship of ESs is related to the LUCC in a given area [28,34]. All trade-offs between ESs occurred from 1980 to 2000, during which the ESV of Qinghai Province showed a decreasing trend. The shift from synergies to trade-offs between ES pairs mostly occurred under the elevation, slope, and RDLS gradients in classes II and III, followed by classes IV and V (Table 3). Each ES pair was synergistic from 2000 to 2020, during which the ESV of Qinghai Province showed an increasing trend. The degree of synergy of each ES pair significantly varied among topographic gradients, and the degree of synergy of most ES pairs showed the largest change in classes I and II (Table 5). This difference is likely due to the impact of complex terrain conditions on the spatial pattern of land use and ESs in Qinghai Province. In terms of natural conditions, the spatial distribution of vegetation types on the Qinghai–Tibet Plateau exhibits a distinct vertical zonation [71]. In addition, precipitation is also one of the dominant factors affecting the relationship between ES pairs [72]. The vertical distribution characteristics and changes in vegetation and water have the most prominent influence on the distribution pattern of ESs in Qinghai province. They provide a series of ESs, including food production, water supply, material production, regulation of water flows, and soil fertility maintenance, with high ES supply consistently being associated with beneficial natural conditions [73]. In terms of social development, human activities also have significant vertical distribution characteristics in Qinghai Province. In 2020, 64.6% of the population and 62.8% of the GDP in Qinghai Province were distributed in Xining City and Haidong City, which lie at lower altitudes in the east and account for approximately 3% of Qinghai Province. Moreover, 7.9% of the population and 20.6% of the GDP were distributed in the Qaidam Basin. Most high-altitude areas were sparsely populated, the remaining 71.1% of the territory accounted for only 27.5% of the population and 16.6 % of GDP [74,75]. The extremely uneven distribution of economic development and population is also one of the reasons for the variability of the trade-offs/synergies of ESs pairs along topographic gradients.

4.3. Implications for ES Management via the Tradeoffs and Synergies on Different Topographic Gradients

The unique geographical environment and climatic characteristics of Qinghai Province have created a unique ecosystem. ESs of Qinghai Province are remarkable and play an important role in the national ecological security, but the eco-environment is fragile and would be difficult to recover if destroyed [57]. In Qinghai Province, 27 counties covering 58.51% of the province have been identified as Ecological Functional Conservation Areas, delineated to sustain ESs for the entire nation. Thus, the eco-environment of Qinghai Province is essential for human well-being and sustainable development in China. In this study, the total ESV of Qinghai Province showed a downward trend from 1980 to 2000. The Chinese government has implemented a number of national policies for land and space planning and comprehensive management of the eco-environment, clarified the fundamental task of Qinghai Province to build a model province for ecological civilization, and emphasized that the unique eco-environment is an asset, responsibility and potential for Qinghai Province. According to these national policies, the Qinghai provincial government takes the eco-environment as the foundation for achieving the goal of sustainable development. In order to promote ecological restoration and environmental governance, the government actively implements the return of farmland and limited grazing land to forest and grassland, the ecological migration project, and the comprehensive prohibition of the development of the Qilian Mountain National Park, Sanjiangyuan National Park, and Qinghai Lake Bird Island. Previous ecological protection measures are mostly based on the overall ecological status of Qinghai Province, without formulating protection measures according to the vertical distribution characteristics of the natural ecosystem in Qinghai Province. Therefore, understanding the topographic gradient distribution characteristics and changes of ESV and formulating protection measures according to the ecological conditions of different topographic gradients can facilitate the design of ecosystem management measures more suitable for Qinghai Province and provide more information to decision-makers [14,76].

The spatial distribution and change of ESV in Qinghai Province were found to exhibit distinct vertical zonal characteristics, and the changes were found to vary with topographic gradients. Trade-offs/synergies of the same ES pair exhibited diverse patterns on different topographic gradients. According to the above research results and the local situation in Qinghai Province, the corresponding ecological system protection measures are put forward. (1) Implementation of grassland restoration projects should be continued in the eastern Qaidam Basin, Qinghai Lake Basin, and Gonghe Basin, where the elevation is less than 3718 m, the slope is less than 17.39°, RDLS is greater than 1003 m, and Tni greater is than 1.66. It is noteworthy that overgrazing is limited in areas above 3719 m. In addition, limiting overgrazing is not only to limit the number of livestock and grazing areas but also to optimize the structure of livestock according to different types of vegetation in different regions. With the increase in altitude, the temperature of the Qinghai–Tibet Plateau will change significantly, resulting in variations in vegetation growth rate at different altitude gradients [77,78]. In a recent field survey, we found that herders raising yaks and sheep simultaneously in their grazing land were more likely to face a shortage of pasture. This phenomenon may be related to the separation of pastures for yaks and sheep. Thus it is important to coordinate the relationship between the number and structure of livestock and the carrying capacity of regional resources and the environment.

(2) According to the planning of the Sanjiangyuan National Park in Qinghai Province (2018), the Sanjiangyuan Region is divided into the core conservation zone, ecological conservation and restoration zone, and traditional utilization zone. Closure against grazing should be continued in the core conservation zone of the Sanjiangyuan Region with elevation exceeding 4000 m, slope less than 12.31°, and RDLS greater than 712 m. Ecological migration is of great significance in the protection of the eco-environment. However, with the gradual increase in population and livestock in the destination of migration, the appropriate release of ecological conservation and restoration zones to allow grazing activities can alleviate the environmental carrying pressure of the destination. It is worth noting that for future ecological migration projects, the Quma River township can be considered, and herders from high-altitude regions, which are unsuitable for human settlement, can be relocated to the better-equipped Golmud city.

(3) Desertification control in the eastern part of the Qaidam Basin should be strengthened and afforestation should be continued on the basis of the Three-North Shelterbelt to prevent soil erosion, which would also help improve climate regulation and air quality regulation. For desertification caused by overgrazing and other human factors in the Gong-he Basin, grazing should be limited, and local PEH (Photovoltaic, Ecology, Husbandry) development path, as well as the restoration of the ecosystem, should be promoted [79]. Policies for reforestation and ecological migration should be implemented on the southern slope of the Qilian Mountains with an elevation of less than 4285 m and a slope of more than 17.4°.

(4) Considering the high degree of trade-offs between regulation of water flows and food production, materials production, water production, and maintenance of soil fertility in Qinghai Province from 1980 to 2000, more attention should be paid to the protection and management planning of farmland and grassland. During 1980–2000, the food production showed different degrees of trade-offs with other ESs in areas lying at elevations less than 4285 m, with slope less than 7.94° and greater than 12.32°, RDLS greater than 247 m, and Tni less than 1.02 and greater than 1.45. Therefore, more attention should be paid to limiting overgrazing and returning farmland to forest land and grassland. The maintenance of soil fertility showed a greater degree of trade-offs with other ESs in areas lying at an elevation above 4500 m. Grasslands should be restored and wetlands should be protected as much as possible to promote eco-environment optimization, such as for soil and water conservation, biodiversity protection, and aesthetics.

(5) Afforestation should be carried out to build a wetland park based on the Huangshui River to improve the living environment and promote the use of clean energy in the Huangshui River Basin, with elevation less than 3091 m, slope greater than 22.79°, and RDLS greater than 1000 m. Social and economic development cannot be ignored. The eco-environment of Qinghai Province can be protected and ecological civilization can be achieved only when the social economy of Qinghai Province is developed. A circular economic zone should be developed in the Qaidam Basin and its industrial layout should be optimized. As the social and economic center of Qinghai Province, the trend of population agglomeration in the Huangshui River Basin will not change in a short time, or even gradually increase. Nevertheless, permanent basic cropland protection, strict adherence to the ecological red line in urban construction, improvement of land use efficiency, and construction of a community with a common goal can be implemented in the Huangshui River Basin. Moreover, attention should be paid to high-rise residential construction, and ecological security should be ensured.

4.4. Limitations and Prospects

In this study, Qinghai Province was taken as the research area, and the table of equivalent factors of ESV in China formulated by Xie et al. (2017) was directly adopted. Residents’ willingness and ability to pay for local ecosystems will have a strong impact on the ESV of the study area. Therefore, the ESV was revised by combining the local residents’ willingness and ability to pay as parameters. Other natural factors such as desertification, permafrost melting, precipitation, and drought, and human activities such as grazing, urban construction, and coal mining also affect ESV and trade-offs/synergies of ES pairs [72]. However, the influence of these factors was not considered in the revised method of ESV in this study. Therefore, the revised method should be further studied in the future considering the above-mentioned factors. In the coming years, Qinghai Province is anticipated to be the most popular tourist destination in China, and its natural environment plays an important role in providing ESs, especially cultural and amenity services. However, the value of ecotourism in Qinghai Province was not systematically evaluated in this study. In future research, regional biomass differences, NPP, tourism resources, and tourist preferences can be added as parameters to correct ESV. This study selected four types of topographic gradients and studied the spatiotemporal heterogeneity under different topographic gradients of ESV changes and trade-offs/synergies of ES pairs, extending the understanding that ESV changes and trade-offs/synergies of ES pairs depend on different topographic gradients in Qinghai Province. However, the change mechanism of ESV and trade-offs/synergies of ES pairs under different topographic gradients were not analyzed in depth. Future research should comprehensively select various factors that may affect the change of ESV and trade-offs/synergies ES pairs under different topographic gradients.

5. Conclusions

The ESV of Qinghai Province decreased from 129,573.99 million USD in 1980 to 129,155.85 million USD in 2000 and then increased rapidly to 142,682.97 million USD in 2020. Regarding the spatial distribution of ESV in Qinghai Province, ESV was high in the east and south and low in the northwest. Areas with high ESV were mainly distributed around Qinghai Lake and the Sanjiangyuan Region, and those with low ESV were mainly distributed in the Qaidam Basin and its surrounding regions. Most of the regions in Qinghai Province have high altitudes with sharp relief. The eastern, southern, and central regions have a steep relief, whereas most of the western regions are relatively flat.

The average values of provisioning services, regulating services, habitat Services, and cultural and amenity services in Qinghai Province gradually increased with the increase in elevation gradient but decreased in areas above 4758 m. In terms of slope and RDLS, ESV decreased with the increase in the two topographic gradients. The distribution and variation of ESV and trade-offs/synergies of ES pairs in Qinghai Province exhibited distinct vertical zonality, with different spatial patterns on different topographic gradients. All the trade-offs occurred in 1980–2000, during which the total ESV in Qinghai Province showed a downward trend. The ESV of Qinghai province showed an upward trend from 2000 to 2020, and all ESs pairs exhibited synergies during this time. The same ES pairs showed differences in trade-offs/synergies on the elevation, slope, RDLS and Tni gradients, with varying degrees. The results and policy recommendations of this study will provide help for ecological protection and realizing ecological civilization in Qinghai Province.