Multi-Scenario Simulation of Ecosystem Service Value in Xiangjiang River Basin, China, Based on the PLUS Model

Abstract

With rapid socio-economic development, excessive anthropogenic consumption and the exploitation of natural resources have impaired the self-healing, supply, and carrying capacities of ecosystems. The assessment and prediction of ecosystem service values (ESVs) are crucial for the coordinated development of ecology and economy. This research examines the Xiangjiang River Basin and combines land use data from 1995 to 2020, Landsat images, meteorological data, and socio-economic data. These data are incorporated into the PLUS model to simulate land use patterns in 2035 under the following five scenarios: natural development, economic development, farmland protection, ecological protection, and coordinated development. Additionally, this research analyzes the dynamics of land use and changes in ESVs in the Xiangjiang River Basin. The results show that between 1995 and 2020 in the Xiangjiang River Basin, urbanization accelerated, human activities intensified, and the construction land area expanded significantly, while the areas of forest, farmland, and grassland decreased continuously. Based on multi-scenario simulations, the ESV showed the largest and smallest declines under economic development and ecological protection scenarios, respectively. This results from the economic development scenario inducing a rapid expansion in construction land. In contrast, construction land expansion was restricted under the ecological protection scenario, because the ecological functions of forests and water bodies were prioritized. This research proposes land use strategies to coordinate ecological protection and economic development to provide a basis for sustainable development in the Xiangjiang River Basin and constructing a national ecological security barrier, as well as offer Chinese experience and local cases for global ecological environment governance.

1. Introduction

As global industrialization and urbanization continue to accelerate, the excessive exploitation of nature by humans has undermined the self-healing, supply, and carrying capacities of ecosystems [1,2]. The United Nations Sustainable Development Goals explicitly call for the protection, restoration, and promotion of sustainable ecosystem usage, as well as halting the loss of biodiversity [3,4]. The “Master Plan for Major Projects on the Protection and Restoration of China’s Key Ecosystems (2021–2035)” clearly states that by 2035, through a series of major ecological engineering projects, a fundamental improvement in the quality, functionality, stability, and virtuous cycle of natural ecosystems will be achieved, establishing a national ecological security barrier system. Ecosystems are closely related to human life, providing an essential foundation for the survival and development of human society and offering value and benefits to humanity as natural capital. The assessment of ecosystem service values (ESVs), which area critical measure used to evaluate the contribution of ecosystems to human well-being, has become an important basis for achieving a balance between ecological protection and economic development [5].

Since Costanza et al. proposed a method for calculating ESVs based on biomes [6], scholars from various countries have conducted extensive and in-depth research in various directions in terms of different ecosystems, including watershed, urban, and rice ecosystems [7,8,9], and at different research scales ranging from global and regional levels to specific ecosystem scopes [10,11]. A selection of studies have built on this foundation and adapted the ESV equivalent coefficient table, combining China’s unique natural geographical conditions that promote the application and development of ESV in China [12]. Subsequently, researchers have used this coefficient table to comprehensively evaluate ecosystems in different regions of China [13,14]. For example, by measuring the dynamic changes in ESVs in mountainous regions [15], researchers have revealed their intrinsic connection with land use, highlighting the importance of waters, grasslands, forests, and wetlands to ESVs. By assessing the response of land use type changes to ESVs and landscape patterns [16], the significant impact of land use changes on ecosystem services has been uncovered. Currently, quantitative ESV assessment methods in different regions have matured [17]; thus, many scholars have also employed geographical detectors, geographically weighted regression models, and other tools to elucidate driving factors and predict changes in ESVs [18,19]. Among these, the main models for spatial–temporal prediction include the Markov, Conversion of Land Use and its Effects at Small Region Extent (CLUE-S), System Dynamics (SD), and Future Land Use Simulation (FLUS) [20,21,22,23,24,25]. Compared with these models, the PLUS model integrates simulation and decision-making improvement algorithms [26], utilizing a coupled land expansion analysis strategy (LEAS) and a cellular automation model based on multi-type random patch seeds (CARS) to simulate the generation and mutual transformation of different land patches under various scenarios, resulting in more accurate simulation outcomes.

The Xiangjiang River Basin, as a crucial ecological region in China, plays a vital role in the ecological security and sustainable socio-economic development of the entire Yangtze River Basin and even the entire country. However, existing research in this area has primarily analyzed temporal changes in land use patterns and assessed ESVs [27,28,29], with a lack of emphasis on spatial–temporal predictions. Therefore, this study builds on previous studies, utilizes six periods of land use data and socio-economic data from 1995 to 2020, and employs the PLUS model to conduct multi-scenario simulations. It explores the spatial–temporal evolution characteristics of ESVs under different land use scenarios in the Xiangjiang River Basin by 2035. This study aims to provide a scientific basis and decision-making support for the coordinated development of land use planning and ecological economy in the Xiangjiang River Basin, closely aligning with the national strategic goals of ecological protection and restoration, thus offering Chinese experiences and local cases for global ecological environment governance.

1.1. An Overview of the Study Area and Data Sources

1.1.1. An Overview of the Study Area

The Xiangjiang River Basin refers to the catchment area surrounded by the Xiangjiang River watershed, the geographical area ranges between 24°30′~28°40′ N and 110°30′~114°30′ E (Figure 1). The Luoxiao Mountains are located to the east, and the basin is adjacent to Guangdong Province and Guangxi Province in the south. The Hengshan Mountain range is located to the east, and Dongting Lake is located to the north. It encompasses a total area of 108,097.4 km² including Changsha, Zhuzhou, Xiangtan, Chenzhou, Yongzhou, Hengyang, Loudi, and Yueyang, which in combination occupy 51.04% of the total area in Hunan Province.

1.1.2. Data Sources

Data in this study include land coverage, meteorological, and climatic data, as well as socio-economic data, etc. According to a human–computer interactive interpretation of Landsat images, six periods of land coverage data from 1995 to 2020 (five years per period) were obtained. After identification and classification, the land use types were divided into nineteen subcategories that were reclassified into the following seven categories: forest, farmland, grassland, water, wetland, construction land, and unused land. The comprehensive evaluation of first-class land use reached 85%. The specific data source can be found in

Table 1. Data resources.

Data	Resource
Landsat TM/EMT	Data Center for Resources and Environmental Sciences, Chinese Academy of Sciences
Meteorological data: Elevation, Slope, Temperature, Precipitation, Soil types, Distance to waters.	Hunan Water Resources Bulletin
Socio-economic data: Population, GDP, Distance to a city; Distance to first-level road; Distance to second-level road; Distance to third-level road.	China Statistical Yearbook Hunan Statistical Yearbook Statistical Yearbook of prefecture-level cities in Hunan Province Data Compilation of National Agricultural Products Cost and Income Geo-spatial data cloud

.

2. Study Methods

2.1. PLUS Model Prediction

Based on land use raster data, a patch-generating land use change simulation model is introduced, integrating the LEAS with CARS (Figure 2) [30]. This model is designed to simulate the formation and dynamic evolution of land use patches across various scenarios. After extracting and sampling segments of land expansion between two periods, the random forest algorithm is applied to analyze the expansion patterns and driving factors for each land use type. This process yields the development probability for each land use type and quantifies the contributions of each driving factor to land expansion over a specified period. The LEAS strategy combines the advantages of traditional TAS (translation analysis strategy)and PAS(pattern analysis strategy)models, eliminating the necessity to analyze an exponentially increasing number of transformation types with an increase in the number of categories, while preserving the model’s capacity to analyze land use change characteristics within a defined time frame [31].

The changes in land use are mainly affected by natural, socio-economic, and other factors. According to previous studies and based on the characteristics of the study area and difficulty associated with obtaining data, this study selected 12 indicators, including population, GDP, distance to cities and counties, distance to first-level roads, distance to second-level roads, distance to third-level roads, elevation, slope, annual average temperature, annual average precipitation, and distance to water and soil types, as the driving factors for the analysis of land use (Figure 3).

2.2. Evaluation of ESVs

2.2.1. Ecosystem Service Equivalent Table

Due to the difference between the land use classification standard used in this study and the ecosystem classification standard of Xie [32], the existing land use types were reclassified (

Table 2. Ecosystem service equivalent value coefficient per unit area.

Type of Ecosystem		Provisioning			Regulating				Supporting			Cultural
First-Level Classification	Second-Level Classification	Food	Materials	Water	Air	Climate	Environmental Purification	Hydrological Regulation	Soil Conservation	Nutrient Cycling	Biodiversity	Aesthetic Value
Farmland	Paddy field	1.36	0.09	−2.63	1.11	0.57	0.17	2.72	0.01	0.19	0.21	0.09
	Dry land	0.85	0.4	0.02	0.67	0.36	0.1	0.27	1.03	0.12	0.13	0.06
Forest	Coniferous	0.22	0.52	0.27	1.7	5.07	1.49	3.34	2.06	0.16	1.88	0.82
	Mixed forest	0.31	0.71	0.37	2.35	7.03	1.99	3.51	2.86	0.22	2.6	1.14
	Broadleaf	0.29	0.66	0.34	2.17	6.5	1.93	4.74	2.65	0.2	2.41	1.06
Grassland	Shrub forest	0.19	0.43	0.22	1.41	4.23	1.28	3.35	1.72	0.13	1.57	0.69
	Shrubland	0.38	0.56	0.31	1.97	5.21	1.72	3.82	2.4	0.18	2.18	0.96
	Meadow	0.22	0.33	0.18	1.14	3.02	1	2.21	1.39	0.11	1.27	0.56
	Grassland	0.1	0.14	0.08	0.51	1.34	0.44	0.98	0.62	0.05	0.56	0.25
Wetland	Swamp	0.21	0.25	1.3	0.95	1.8	1.8	12.12	1.2	0.09	3.94	2.37
	Intertidal zone	0.3	0.25	1.29	0.95	1.8	1.8	12.11	1.11	0.09	3.93	2.36
Desert	Bare land	0.01	0.03	0.02	0.11	0.1	0.31	0.21	0.13	0.01	0.12	0.05
	Other	0	0	0	0.02	0	0.1	0.03	0.02	0	0.02	0.01
Waters	Waterway	0.4	0.12	8.36	0.72	2.12	4.57	87.5	0.41	0.04	2.05	1.58
	Lake	0.8	0.23	8.29	0.77	2.29	5.55	102.24	0.93	0.07	2.55	1.89
	Reservoir	0.4	0.11	2.09	0.23	0.71	1.14	21.87	0.52	0.03	0.51	0.4
Building/ Construction Land	Urban land	0	0	0	0	−1.29	0	−1.51	0.55	0	0	0.69
	Rural settlement	0	0	0	0	−1.29	0	−1.5	0.55	0	0	0.7
	Other	0	0	0	0	−1.29	0	−1.51	0.55	0	0	0.69

). Based on the classification results obtained by the research, construction land is added to the first-level classification. The secondary classification includes reservoirs and ponds; this can also be because their ESVs are similar to water ESVs and, thus, are classified as water in the ecosystem classification. The ecosystem value of intertidal zones and marshlands is similar to that of wetlands; thus, they are classified as wetlands in the ecosystem classification. Unused land, similar to bare land and other types of land, shares similar ESVs with deserts; thus, it is classified as desert in ecosystem classification. Construction land is subdivided into urban land, rural residential land, and other construction land. The secondary classification of farmland, forest, and grassland remains unchanged.

2.2.2. Correction of Coefficient

(1)

Food coefficient

Due to the different environmental, climatic, and economic features in different study areas, the ESVs differ significantly and change continuously. To improve the accuracy of evaluation, this study corrects the food coefficient, social development coefficient, and regional difference coefficient for the ESV of the Xiangjiang River Basin based on Xie’s equivalent table and previous research results. The formula for calculating the food adjustment coefficient is as follows:

$$P_t=C_{t^{\prime }}/C_t$$

In this formula, $P_t$ represents the food adjustment coefficient for year t; $C_{t^{\prime }}$ represents the average grain yield in the Xiangjiang River Basin (kg/hm²); and $C_t$ represents the national average grain yield (kg/hm²).

(2)

Social development coefficient

The social development coefficient refers to the relative level of an individual’s willingness to pay for ecological value at different socio-economic levels. The calculation formulas are as follows:

$${\mathrm{D}}_{\mathrm{t}}={\mathrm{l}}_{{\mathrm{t}}^{\prime }}/{\mathrm{l}}_{\mathrm{t}}$$

$$l^{\prime }=l_1{M}_1+l_2{M}_2$$

In these formulas, ${\mathrm{D}}_{\mathrm{t}}$ represents the social development adjustment coefficient for year t; ${\mathrm{l}}_{{\mathrm{t}}^{\prime }}$ represents the social development coefficient of the Xiangjiang River Basin; ${\mathrm{l}}_{\mathrm{t}}$ represents the national social development coefficient; $l^{\prime }$ represents the social development coefficient related to the actual willingness to pay; $l_1$ represents the urban social development coefficient; $M_1$ represents the proportion of the urban population; $l_2$ represents the rural social development coefficient; and $M_2$ represents the proportion of the rural population.

(3)

Regional difference coefficient

The regional difference coefficient reflects the difference caused by the mutual effects of varied factors in different regions. The ESVs of different regions were modified mainly through the difference in the number of creatures in different regions, where net primary potential productivity is used to replace the number of creatures. The calculation formulas are as follows:

$$Q_t=NP{P}_{t^{\prime }}/NP{P}_t$$

$$\mathrm{N}\mathrm{P}\mathrm{P}=3000[1-{\mathrm{e}}^{-0.000969(\mathrm{Z}-20)}]$$

$$\mathrm{Z}={\displaystyle \frac{1.05\mathrm{R}}{\sqrt{{1+1.05\mathrm{R}/\mathrm{H})}^2}}}$$

$$\mathrm{H}=3000+25\mathrm{t}+0.5{\mathrm{t}}^3$$

In these formulas, $Q_t$ represents the coefficient of variation between regions in year t; $NP{P}_{t^{\prime }}$ represents the net primary production potential of natural vegetation t·a⁻¹·hm⁻²; $NP{P}_t$ represents the average net primary productivity; Z is the actual evapotranspiration in the Xiangjiang River Basin over a year (mm); H is the average evapotranspiration in the Xiangjiang River Basin over a year (mm); t is the average temperature in the Xiangjiang River Basin over a year (°C); and R is the total precipitation in the Xiangjiang River Basin over a year (mm).

2.2.3. Calculation of ESVs

Based on the valuation method for ecosystem services, 1/7 of the economic value of grain production in 1 hm² of farmland is regarded as the ESV equivalent of a standard equivalent factor [33]. The major type of farmland in the study area is paddy fields, and the rice yield accounts for greater than 90% of the total output of the three major grains; therefore, the economic value of rice is substituted for the economic value of grain production. The average price of rice was 2.59 CNY/kg in 2020, and the rice yield was 6608 kg per hm²; therefore, an equivalent factor of the ESV in the study area was 2445.21 CNY/kg. The calculation formulas of ESV in the Xiangjiang River Basin are as follows:

$$ESV=\sum (A_k\times {VC}_k\times P_t\times D_t\times Q_t)$$

$${ESV}_f=\sum (A_k\times {VC}_{fk}\times P_t\times D_t\times Q_t)$$

$${VC}_k=\sum {VC}_{fk}$$

In these formulas, ESV is the total ecosystem service value of the Xiangjiang River Basin (unit: CNY); $A_k$ is the area of land use type k (unit: hm²); ${VC}_k$ is the ecosystem service value coefficient per unit area of land use type k (unit: CNY/hm²); $P_t$ is the food correction coefficient for year t; $D_t$ is the social development correction coefficient for year t; $Q_t$ is the regional difference coefficient for year t; ${ESV}_f$ is the value of the ecosystem service function of kind f (unit: CNY); and ${VC}_{fk}$ is the value coefficient of the service function of kind f for land use type k (unit: CNY/hm²).

2.3. Land Use Dynamics

Land use dynamics are a quantitative indicator for assessing the rate of change in land use that includes single and comprehensive land use dynamics [34]. The calculation formulas are as follows:

$$K={\displaystyle \frac{U_b-U_a}{U_a}}\times {\displaystyle \frac{1}{T}}\times 100\%$$

$$C={\displaystyle \frac{\sum _{i=1}^n|U_{ai}-U_{bi}|}{2\times \sum _{i=1}^n{U}_{ai}}}\times {\displaystyle \frac{1}{T}}\times 100\%$$

In these formulas, K is the single land use dynamic degree of a specific land type during the study period; $U_a$ and $U_b$ represent the area of a specific land type at the beginning and end of the study period, respectively (unit: hm²); T is the duration of the study period (unit: year); C is the comprehensive land use dynamic degree during the study period; and $|U_{ai}-U_{bi}|$ represents the area change value of land type i converted to other land types within the study area (unit: hm²).

3. Results and Analysis

3.1. Changes in Land Use 1995–2020

Land use data for the Xiangjiang River Basin from the six periods from 1995 to 2020 were sourced from the Data Center for Resources and Environmental Sciences of the Chinese Academy of Sciences. With reference to land use classifications in China and the research outcomes of relevant experts and scholars, combined with the current characteristics of land use in the Xiangjiang River Basin, the land use types in the study area were categorized into the following seven major types: farmland, forest, grassland, water, construction land, wetland, and unused land [35]. The land use status maps of the Xiangjiang River Basin during the six periods from 1995 to 2020 were obtained (Figure 4), and the classification system is presented (

Table 3. Land use reclassification system in the Xiangjiang River Basin.

First-Level Classification	Second-Level Classification
Cropland	Paddy fields and dry land
Forest	Coniferous forest, mixed forest, broadleaf, and shrub forest
Grass	Scrubland, meadow, and grassland
Waters	Waterways, lakes, and reservoirs
Building	Urban land, rural settlement, and other
Wetland	Swamps and intertidal zones
Unused land	Bare land and other

).

The main land use type in the area from 1995 to 2020 is forest, accounting for more than 60% of the total area in the Xiangjiang River Basin; farmland is the second largest land use type, accounting for 30% of the total area, and the area of each land use type is successively forest > farmland > water > grassland > wetland > construction land > unused land. Based on the dynamic assessment of land use changes (

Table 4. Area and dynamic index of land use changes in the Xiangjiang River Basin from 1995 to 2020.

Land Use Types	1995–2000		2000–2005		2005–2010		2010–2015		2015–2020
	Change	Dynamics	Change	Dynamics	Change	Dynamics	Change	Dynamics	Change	Dynamics
Grassland	7.51	0.56	−0.51	−0.04	−25.27	−1.83	−1.82	−0.15	−4.39	−0.35
Farmland	1.52	0.01	−19.02	−0.12	−35.04	−0.21	−23.21	−0.14	−36.61	−0.23
Construction land	4.38	0.51	25.44	2.87	83.24	8.21	41.86	2.93	62.54	3.81
Forest	−9.85	−0.03	−7.74	−0.02	−21.08	−0.06	−17.82	−0.05	−22.72	−0.07
Wetland	6.66	1.88	2.18	0.56	−5.36	−1.34	27.88	7.68	−1.29	−0.25
Waters	−10.3	−0.62	−0.18	−0.01	2.47	0.15	−26.89	−1.67	2.91	0.20
Unused land	0.01	0.15	−0.11	−1.51	1.06	17.10	0.01	0.11	−0.44	−3.88
Comprehensive land use dynamics	0.18		0.12		0.33		0.31		0.55

), C < 0.2% in the Xiangjiang River Basin from 1995 to 2020, the land use in this period developed steadily; C > 0.3% in the Xiangjiang River Basin from 1995 to 2020, because more forest land was converted to construction land and farmland after 2005 and land use changed rapidly. In general, the dynamic change inland use in the Xiangjiang River Basin in the past 25 years varied slightly, and the land use activity was not high.

From 1995 to 2020, the single dynamic degree of construction land utilization presented a positive change, while those of grassland, farmland, and forest generally exhibited negative changes. The urbanization rate in the Xiangjiang River Basin was 24.77% in 1995 and reached 60.65% in 2020. Socio-economic factors significantly influenced the changes in land use. With the continuous improvement in urban areas, the growth rate of construction land accelerated constantly in the 25 years, with an increase of 2.17 × 10⁵ hm². In addition, the urban development boundary has expanded continuously, mainly because of the spreading of suburbs of cities and counties. The wetland area has shown a slow upward trend over the 25 years, with nearly no change before 2010 and a rapid increase from 2010 to 2015, with a single land use dynamic degree as high as 7.68%. This occurred mainly because water areas were converted to wetlands during this period. Before 2005, unused land remained largely unchanged; however, from 2005 to 2010, the single land use dynamic degree of unused land reached a maximum value of 17.10%, with the area increasing from 1.23 × 10³ hm² to 2.29 × 10³ hm², an increase of 1.06 × 10³ hm². Due to the degradation of forest and grassland, 1186 hm² of grassland and 146 hm² of forest were converted into unused land, resulting in soil erosion and accelerated land desertification. Overall, the land use structure in the Xiangjiang River Basin is relatively stable, mainly forest and farmland. The majority of areas in the Xiangjiang River Basin are mainly engaged in agricultural production; therefore, agriculture and aquaculture are relatively well-developed. With the acceleration in urbanization, the construction land type has shown a gradual trend of outward expansion from the centers of cities and counties. As the use of construction land increased continuously, the ecological environment was damaged to a certain degree.

3.2. Prediction of Multiple Land Use Scenarios

According to some research, the FLUS model must consider various constraints when predicting trends inland use changes [36]. Scenario setting plays a significant role in forecasting land use development. Further integrating the scenario settings from some researchers [37,38] led to the following five development scenarios being established, including the natural development scenario, economic development scenario, farmland protection scenario, ecological protection scenario, and coordinated development scenario (

Table 5. Descriptions of land use change scenarios in different models.

Pattern	Description
Natural development scenario	Simulate land use change patterns from 1995 to 2020.
Economic development scenario	Primarily expand construction land while appropriately increasing the conversion probability of other land uses to construction.
Farmland security development scenario	Prohibit the conversion of arable land to other uses while appropriately increasing the probability of converting other land uses to arable land.
Ecological protection scenario	Prohibit the conversion of water bodies and forest land to other uses while increasing the probability of converting other land uses to water bodies and forest land.
Coordinated development scenario	Increase the probability of converting other land uses to arable and forest land while decreasing the probability of conversion to construction land.

). Under these conditions, using the 2020 land use data for the Xiangjiang River Basin, the PLUS model was employed to predict the land use situation in the Xiangjiang River Basin in the year 2035.

Natural development scenario: Under this scenario, the historical change trend of all land use types should be maintained, and no restricted area should be set for any land use type during the conversion. The change should follow the natural development mode.

Economic development scenario: This scenario highlights the importance of economic development. The expansion rate of construction land increased on the basis of the natural development scenario, manifesting mainly as an increase in construction land area and decreases in farmland, forest, grassland, water, and wetland.

Farmland protection scenario: Under this scenario, farmland is prohibited from being converted to other land types and appropriately increases the probability of converting other land types into farmland. The red line of farmland protection is firmly abided by, and any attempts to use farmland for a purpose other than agriculture, specifically grain production, are prohibited. Farmland should be primarily used for grain production, and the quality of farmland should be recovered on the premise of ecological security.

Ecological protection scenario: This scenario emphasizes the importance of ecological protection and strictly forbids the conversion of water and forest into other land types, as well as increasing the probability of converting other land types into water or forest.

Coordinated development scenario: This scenario emphasizes the improvement of economic development, fully integrating the concept of ecological civilization into the entire process of socio-economic development, and achieves the comprehensive conversion of resource utilization and environmental protection. The main parameters considering the need for socio-economic development and adjusting the land use conversion to increase the probability of other land use types being converted into farmland and forest, while reducing the probability of their conversion into built-up land.

The neighborhood weights for each land use type were determined based on the ratio of the expansion area of each land use type to the total land expansion.

Using the PLUS model, different land use weights (Figure 5) are set for different prediction scenarios to achieve the purpose of multi-scenario prediction. Under the five scenarios, the PLUS model predicts that substantial changes will occur for each land use type in 2035 compared with 2020 (Figure 6). Under the natural development scenario, the areas of grassland, farmland, forest, water, and unused land will decrease, while the areas of wetland and construction land will increase. The expansion in construction land area will accelerate continuously compared with 1995–2020. From the perspective of space, construction land in the central urban area has a trend of further expansion, mainly by converting farmland and forest.

Under the economic development scenario, prioritizing economic development, construction land will expand rapidly, and the areas of grassland, farmland, forest, water, and unused land will decrease. From the perspective of quantity, construction land will increase by approximately 1.9 × 10⁵ hm², while the areas of grassland, farmland, and forest will decrease by 1.42 × 10³ hm², 32.67 × 10³ hm², and 131.78 × 10³ hm², respectively. From a spatial perspective, the construction land area in various regions is constantly expanding.

Under the farmland protection scenario, farmland and forest areas will increase by 227.3 × 10³ hm² and 62.15 × 10³ hm², respectively, compared with the natural development scenario, and the wetland area will be reduced by 248.92 × 10³ hm² and 41.69 × 10³ hm. The farmland area mainly results from the conversion of construction land and wetlands. Compared with 2020, the area of farmland decreased by 63.01 × 10³ hm², while the rate of decrease has declined continuously. In the future, the phenomenon of farmland conversion will continue to decrease, and the area of farmland will increase and remain stable.

Under the ecological protection scenario, the areas of wetland and water increased by 398.39 × 10³ hm² and 23.04 × 10³ hm², respectively, compared with the natural development scenario. Meanwhile, the areas of farmland, construction land, and wetland decreased by 51.54 × 10³ hm², 322.62 × 10³ hm², and 48.21 × 10³ hm², respectively, while the areas of grassland and unused land changed minimally. When prioritizing ecological development, an increase in forest and water mainly results from farmland, construction land, and wetlands; in this scenario, the importance of ecological protection is highlighted, and the conversion of ecological land into other land types is prohibited.

Under the coordinated development scenario, the areas of construction land, wetland, and unused land decreased by 243.96 × 10³ hm², 41.39 × 10³ hm², and 0.4 × 10³ hm², respectively, compared with the natural development scenario. Meanwhile, the areas of grassland, farmland, forest, and water increased by 3.52 × 10³ hm², 8.93 × 10³ hm², 272.3 × 10³ hm², and 0.99 × 10³ hm². The main reason is that the expansion of construction land for human activities was controlled, because some construction land was converted to ecological land use types, such as forest and water, which is beneficial to the sustainable ecological development of the Xiangjiang River Basin.

The land use dynamic degree was calculated using the land use data predicted by the PLUS model for 2021–2035, combined with the formula in Section 2.3. The land use dynamics of grassland, farmland, waters, and unused land present a negative change from 2020 to 2035. In the coordinated development scenario, the land use dynamics of grassland is −1.39% (

Table 6. Dynamic index of land use changes in the Xiangjiang River Basin in 2035.

Land Use Types	1995–2020 Land Use Dynamics	2020–2035 Land Use Dynamics (%)
		Natural Development Scenario	Economic Development Scenario	Farmland Protection Development Scenario	Ecological Protection Scenario	Coordinated Development Scenario
Grassland	−0.36	−1.42	−1.46	−1.41	−1.39	−1.32
Farmland	−0.14	−0.40	−0.47	0.08	−0.51	−0.38
Construction	5.03	4.42	7.68	0.17	−1.09	0.25
Forest	−0.05	−0.03	−0.16	0.03	0.38	0.25
Wetland	1.69	2.54	0.92	−0.21	−0.64	−0.19
Waters	−0.38	−0.62	−0.61	−0.60	−0.11	−0.60
Unused land	1.59	−3.66	−4.80	−2.94	−3.87	−5.09

), which indicates that the grassland has the lowest rate of decrease under this scenario comparing with other scenario; however, it still decreased compared with the grassland area in 2020. Farmland has the lowest rate of decrease under the farmland protection scenario followed by the coordinated development scenario; while construction land has the highest rate of the increase under the economic development, it presents a negative growth trend under the ecological protection scenario. This type of land use reflects human activity and economic development and generally presents a positive growth trend. Therefore, the development of construction land under the ecological protection scenario contradicts reality; whereas, forest has the largest area under the ecological protection scenario, followed by the coordinated development scenario.

The changes in the area of different types of land use in the Xiangjiang River Basin from 1995 to 2020, as well as the differences in land use predictions for 2035 under various scenarios, are presented below (Figure 7a,b).

Grassland has the largest area under the coordinated development scenario; farmland has the largest area under the farmland protection development scenario, followed by the coordinated development scenario; waters have the largest area under the ecological protection scenario, followed by the coordinated development scenario; unused land has the largest area under the farmland protection development scenario and the smallest area under the coordinated development scenario; and construction land has the largest area under the economic development scenario and the smallest area under the ecological protection scenario.

3.3. ESV Estimations Based on Land Use Predictions

3.3.1. Data Prediction and Its Change

(1)

Prediction of equivalent coefficient

Food correction: The economic value of different crops was quantified to adjust the ecosystem service value (ESV) results, making them more realistic.

NPP correction: The regional difference coefficient reflects distinctions formed by interactions between different factors across regions. It primarily adjusts the ESV for different areas based on variations in biomass.

Social development correction: As the economy and society develop, people’s willingness to pay for ecosystem functions and services gradually increases. The social development stage coefficient reflects the relative level of willingness to pay under different socio-economic conditions and living standards.

Based on the correction coefficient from 1995 to 2020, these data were analyzed through the prediction worksheet in Excel. The food correction coefficient, NPP correction coefficient, and social development correction coefficient in 2035 were predicted. Then, the total correction coefficient was calculated to estimate the ESV, as shown in Figure 8.

Based on the land use data under the five different scenarios in the Xiangjiang River Basin in 2035, five different ESVs were calculated by combining the ESV evaluation system. Consider the 2020 as the actual scenario, this study compares ESVs under the five development models.

(2)

Quantitative change characteristics

From the perspective of ESV contributed by different scenarios (

Table 7. ESVs of different land use types in the Xiangjiang River Basin under different scenarios.

Land Use Types	Ecosystem Service Value/Hundred Million CNY
	2020	Natural Development Scenario	Economic Development Scenario	Cropland Security Development Scenario	Ecological Protection Scenario	Coordinated Development Scenario
Grassland	175.99	75.46	74.90	75.60	75.96	76.84
Cropland	488.45	385.27	381.06	414.51	378.63	386.41
Construction land	−23.75	−32.85	−42.52	−20.25	−16.52	−20.50
Forest	4837.13	4203.19	4118.64	4243.07	4458.82	4377.91
Wetland	102.89	118.35	97.50	83.02	77.50	83.28
Water	1072.26	775.25	776.44	777.43	840.44	778.06
Unused land	0.07	0.02	0.01	0.02	0.02	0.01
Total	6653.03	5524.68	5406.03	5573.41	5814.84	5682.01

), the total ESV under all scenarios presented a declining trend compared with 2020. The economic development scenario showed the largest decline, from 665.303 billion CNY in 2020 to 540.603 billion CNY, a decrease of 124.7 billion CNY. In the ecological protection scenario, the decrease was the smallest, with a decrease of 83.810 billion CNY, followed by the coordinated development scenario, farmland protection scenario, and natural development scenario, with decreases of 97.102 billion CNY, 107.962 billion CNY, and 112.835 billion CNY, respectively. Under the five scenarios, in addition to the ecological protection scenario, the ESV under the coordinated development scenario is the highest, with the highest values of supply service, adjustment service, supply service, and cultural service at 32.865 billion CNY, 393.59 billion CNY,117.98 billion CNY, and 24.548 billion CNY, respectively.

3.3.2. Space Prediction and Its Change

To better visualize the spatial distribution of ecosystem service values (ESV) in the Xiangjiang River Basin for 2035, a grid-based spatial analysis was applied across five scenarios, with results shown in Figure 9.

According to the 2035 spatial distribution map of ESVs, under different scenarios, the overall spatial distribution of high-value, moderate–high-value, medium-value, moderate–low-value, and low-value ecosystem services in the Xiangjiang River Basin is basically identical, descending from north to the middle and from west to east with the medium value services distributed surrounding the Xiangjiang River Basin. The proportional area of each grade differs; high-value areas are mainly distributed near Dongting Lake, west of Yueyang City. The ESV is largest under the ecological protection scenario, followed by the coordinated development scenario; low-value areas are distributed in Changsha, and the central urban areas of other cities, and the economic development scenario amounts to a large proportion.

4. Discussion

4.1. Land Use Strategies

4.1.1. Strengthen Farmland Protection and Optimize Land Use Patterns

The Xiangjiang River Basin is mainly dominated by rice cultivation, and from 1995 to 2020, there has been an unregulated expansion in construction land area, resulting in a continuous decrease in farmland. Local governments should strictly adhere to the red line of protecting farmland and ensure that the farmland area is maintained. The towns that experienced a significant reduction in farmland should implement policies that redesign farmland properly, ensuring that farmland is replenished in a balanced manner. Meanwhile, firmly ceasing any attempts to use farmland for any purpose other than agriculture, specifically grain production, is necessary, thus safeguarding national food security. Besides, strengthening the protection of water areas and enhancing river flow and water transport capacity is also important. Under the natural development scenario, construction land area retains the state of unregulated expansion, and it should be developed efficiently and economically to reduce its expansion rate while rationally arranging rural residential areas and actively guiding the development and utilization of unused land to improve land use efficiency. By adjusting the land use mode and structure and optimizing land use patterns, under the coordinated development scenario, construction land for human activities is restricted, and land use types mainly transform to ecological land use types, such as forest and water, which are beneficial to sustainable development in the Xiangjiang River Basin.

4.1.2. Strengthening the Construction of Ecological Origins and Improving ESVs

(1)

Protect existing forests and grasslands and increase their area.

Forests represent the main land use type in the Xiangjiang River Basin, accounting for greater than 50% of total land use. Forests primarily serve regulatory functions, providing essential material resources for human beings and, more importantly, improving the natural environment and regulating climate. To address the expansion of construction land, establishing ecological origin areas as ecological barriers in high-value areas such as the northern part of Yueyang City and the southern part of Chenzhou City while gradually improving the surrounding environment is recommended. Ecological protection and restoration work should be conducted in low-value areas by implementing programs such as returning farmland to forest or grassland to enhance ESV. Promoting natural regeneration measures for forest and grassland, conducting land greening actions, and returning farmland to forest or grassland in key eco-functional zones in order to improve ecological quality and stability.

(2)

Protect water and wetlands to maximize their ecological functions.

Water and wetlands have high ESVs; thus, they are key ecological protection areas that mainly regulate hydrology, supply water, regulate climate, and protect biodiversity. In future developments, establishing water resource protection zones along with strengthening water body conservation efforts through promoting water-saving agriculture practices is necessary, while also enhancing the storage capacity of rivers by dredging channels. Improving the overall connectivity between river basins’ water systems and important terrestrial ecosystems by constructing multi-scale ecological corridors will assist in building a network for biodiversity conservation.

4.1.3. Establishing a Mechanism for Achieving the Value of Ecological Products and Optimizing Regional Economic Structures

Although the ecological protection scenario can significantly reduce the loss of ESV, it may also exert a certain inhibitory effect on the growth rate of regional GDP. This trade-off indicates that relying solely on ecological protection to meet the demands of economic development is insufficient. Therefore, transforming ESV into economic benefits through a mechanism for obtaining the value associated with ecological products is necessary. Specifically, a cross-watershed compensation mechanism can be promoted, utilizing fiscal transfer payments or ecological compensation funds to provide economic compensation to regions engaged in ecological protection, thus mitigating the negative impact of ecological protection on local economies. Concurrently, pilot projects for carbon trading can be initiated, converting the carbon sequestration functions of forests, wetlands, and other ecosystems into economic gains and encouraging enterprises and local governments to participate in ecological protection projects. Furthermore, exploring market-oriented pathways for ecological products, such as eco-tourism and eco-agriculture, can directly transform ecosystem services into economic benefits, fostering the synergistic development of regional economies and ecological systems.

4.1.4. The Need for Refined Spatial Governance

The distribution of ESVs exhibits significant spatial heterogeneity, with high-ESV areas primarily concentrated in regions of critical ecological importance, such as wetlands and forests. In contrast, low-ESV areas are predominantly urban centers. This spatial variability requires the adoption of refined spatial governance strategies to achieve the coordinated development of regional ecologies and economies. Consequently, delineating ecological redlines in high-ESV areas and strictly prohibiting industrial development and large-scale urbanization to preserve the integrity and stability of ecosystems is imperative. Simultaneously, pilot projects for ecological migration should be implemented within these redline areas to minimize human interference in ecosystems. In low-ESV urban areas, the promotion of “sponge city” initiatives is essential, employing measures such as rain gardens and permeable pavements to enhance a city’s capacity for rainwater absorption and utilization. Additionally, the adoption of three-dimensional afforests techniques, including rooftop and vertical greening, can increase urban green space and elevate the ESV per unit area. These refined governance measures can render it possible to protect high-ESV ecological regions, while simultaneously enhancing the ESV of urban areas, thus facilitating the harmonious development of regional ecologies and economies.

4.2. Problems and Shortcomings

In terms of model limitations, the PLUS model, while effective in capturing dynamic changes in land use patches, has certain limitations in selecting driving factors. For instance, the model does not fully take into account some socio-economic factors such as the intensity of policy interventions and dynamic changes in ecological protection policies that may affect the accuracy of long-term predictions.

Regarding the ESV assessment method, this study employs a static equivalent coefficient method. Future research could consider dynamic changes in climate, technological advancements, and the value of ecosystem services. For example, future changes in food production and net primary productivity (NPP) may significantly impact ESV; however, the current predictive method using correction coefficients does not adequately reflect these dynamic changes. Additionally, the ESV assessment does not fully account for the interactions and trade-offs among different ecosystem services. For instance, forest conservation may have positive effects on water purification and biodiversity protection, but its impact on land use efficiency has not been sufficiently quantified.

5. Conclusions

(1)

Dynamic Characteristics of Land Use: From 1995 to 2020, the construction land in the Xiangjiang River Basin experienced rapid expansion, with an average annual growth rate exceeding 0.5%, primarily encroaching upon farmland and forests. The wetland area exhibited periodic fluctuations because of the transformation of waters, while unused land significantly increased because of ecological degradation, reflecting the substantial impact of human activities on ecosystem structures.

(2)

Differences in Multi-Scenario Predictions: By 2035, under the natural development scenario, construction land continues to expand, increasing to 649.16 × 10³ hectares, while the ecological protection scenario restricts the conversion of ecological land, leading to increases in forest and water areas by 5.6% and 9.8% respectively, with the smallest decline in ESV, a 12.6% reduction compared with 2020. The coordinated development scenario balances ecological and land use needs, resulting in a 3.7% increase in forest area and an ESV higher than other non-ecological priority scenarios. Under the economic development scenario, construction land expansion grows by 115.2% compared with 2020, with an ESV decline of 18.8%, indicating the significant negative effects of extensive growth on ecosystems. In the ecological protection scenario, by limiting the conversion of ecological land, forests and water areas increase to 6949.09 × 10³ hectares and 296.00 × 10³ hectares, respectively, increases of 6.1% and 8.4%, compared with the natural scenario, with the smallest ESV decline (12.6%), validating the effectiveness of the ecological priority strategy. Under the coordinated development scenario, the growth rate of construction land decreases to 3.8%, a 38.2% reduction compared with the natural scenario; the forest area increases by 4.2%, and the ESV of 568.201 billion CNY is higher than the farmland protection and natural development scenarios, demonstrating the feasibility of the “growth–protection” synergistic pathway.

(3)

Spatial Heterogeneity of ESV: The distribution of ESVs across the basin exhibits heterogeneity. High-value areas are predominantly concentrated in the western and northern regions that are characterized by extensive forests and water areas. In contrast, low-value areas are primarily located in urban centers, reflecting the negative impact of urbanization on ecosystem services.