The spatiotemporal changes and trade-off synergistic effects of ecosystem services in the Jianghan Plain of China under different scenarios

The Jianghan Plain of China is in the middle reaches of the Yangtze River, the middle and lower reaches of the Han River, and the central and southern parts of Hubei Province. It is an important part of the middle reaches of the Yangtze River Plain, one of the three major plains in China. It is between 29°26'∼31°37' north latitude and 111°14'∼114°36' east longitude, covering an area of more than 34,000 square kilometers. It has a subtropical monsoon climate, with an average annual sunshine hours of approximately 2,000 h and a total annual solar radiation of approximately 460–480 kilojoules/cm². The frost-free period is approximately 240–260 days, the duration above 10 °C is approximately 230–240 days, and the active accumulated temperature is 5100 °C–5300 °C. The average annual precipitation is 1100–1300 mm. China's Jianghan Plain mainly includes 8 counties and cities in Jingzhou City: Jingzhou District, Shashi District, Jiangling County, Gong'an County, Jianli City, Shishou City, Honghu City, and Songzi City, and 3 provinces, including Xiantao, Qianjiang, and Tianmen. It governs county-level cities and radiates to the surrounding five prefecture-level cities of Wuhan, Xiaogan, Jingmen, Yichang and Xiangyang: Caidian District, Hanchuan City, Yingcheng City, Shayang County, Jingshan City, Zhongxiang City, Zhijiang City, Yicheng City, etc, including 16 counties and cities (Ren et al 2022) see figure 2.

Zoom In Zoom Out Reset image size

Land use data from 2000 to 2020 were from the Resource and Environmental Science Data Center of the Chinese Academy of Sciences (http://www/resdc.cn/), with a spatial resolution of 30 m. Visual interpretation was generated using Landsat remote sensing images as the data source (http://glovis.gov/). This mostly included arable land, forest land, grass land, water land, build-up land and unused land. The data interpretation accuracy reached 95.7%, DEM elevation data came from the Geospatial Data Cloud (http://www.gscloud.cn/), with a spatial resolution of 30 m. The data extract slope, aspect, etc and meteorological data came from the National Meteorological Center (http://data.cma.cn/), mostly including annual average temperature, rainfall, etc, with a spatial resolution of 100 m, soil data came from the World Soil Database (http://www.fao.org/soils- portal/soil-survey/), mostly including soil texture, root depth, etc, with a spatial resolution of 100 m, grain output data from 2000 to 2020 came from the 'Hubei Statistical Yearbook' and 'Hubei Rural Statistical Yearbook', population density and GDP data from 2000 to 2020 came from the Resource and Environmental Science Data Center of the Chinese Academy of Sciences (http://www.resdc.cn/), with a spatial resolution of 1000 m, road vector data such as rivers, highways, national highways, railways and county roads came from the National Basic Geography Information Center (http//www.ngcc.cn/), and then ArcGIS10.8 was used to extract the distance to roads and rivers with a spatial resolution of 100 m. The permanent basic arable land and ecological protection red lines in the policy constraint data came from the Hubei Provincial Department of Natural Resources (https://zrzyt.hubei.gov.cn/), with a spatial resolution of 30 m. Finally, the coordinate system of each grid factor was WGS_1984_EASE_Grid_Global, and the spatial resolution of the resampled data was unified to 30 m. See table 1 for details.

In this study, we used the GeoSOS-FLUS model (hereinafter referred to as FLUS) to simulate future land use structure changes and spatial distribution characteristics in the Jianghan Plain of China. The specific steps were as follows:

The built-in Extract Land Expansion module of the PLUS model was used to overlay the land use type raster data from 2000 to 2020 to obtain five periods of land use type change data. With reference to existing research (Liang et al 2018b) and combined with the current situation of the Jianghan Plain in China, 12 study area elevations, slopes, aspects, average annual precipitation, average annual temperature, population density, GDP, and distances to rivers, highways, national highways, county roads and railways were selected as driving factors of land use change. The LEAS module was used to identify its driving forces. The default value of the random forest decision tree was set to 20, the sampling rate was 0.01, the number of training RF features was 12, and the suitability atlas of six land use types was obtained. We input the suitability atlas of each land use type and the land use image in 2000 into the CARS module of the model, and used the 2020 Markov chain prediction results to simulate the land use situation in the Jianghan Plain of China in 2020, which was the same as the actual situation in 2020. The land use images were compared and tested. The Kappa value of the predicted image in 2020 was 0.8801, and the overall accuracy was 91.56%, which met the accuracy test. The land use image in 2035 was then simulated. Referring to existing research (Liu et al 2017) and the actual situation of the study area, the neighborhood weights of each land type are shown in table 2 after debugging. The scenario simulation settings in this article were as follows:

• (1)  
  Natural development scenario (Scenario 1): According to the current pattern of land use change and development model, this scenario was based on the land use transfer rate and natural and socioeconomic driving factors from 2000 to 2020, without considering policy limiting factors.
• (2)  
  Arable land protection scenario (Scenario 2): Basic arable land was the vehicle for grain security. This scenario was based on the natural development scenario, and permanent basic arable land protection areas were added as restricted conversion areas. The Markov transfer probability matrix was modified to reduce the probability of arable land transferring to build-up land by 60%, thereby strictly implementing arable land protection.
• (3)  
  Ecological protection scenario (Scenario 3): Accounting for the land structure of ecology, agriculture, and cities and incorporating restrictive factors such as ecological protection while reducing the probability of ecological land transfer to built-up land, the transfer of forest land and grass land to build-up land was reduced by 50%, the transfer of arable land to forest land and grass land was increased by 30%, and the transfer of water land and arable land to build-up land was reduced by 30%.
• (4)  
  Urban development scenario (Scenario 4): According to the needs of urban development and natural laws, the build-up land area was controlled within 1850 km² to reduce the probability of various types of land being converted to build-up land.

We used the InVEST model to quantitatively assess five ecosystem services, including carbon storage, crop production, habitat quality, water yield and soil conservation, in the Jianghan Plain of China in 2000, 2005, 2010, 2015 and 2020. ArcGIS 10.8 was then used to calculate the above ecosystem services to each county and city according to the average value, thereby realizing research on ecosystem services in the counties and cities of the Jianghan Plain in China from 2000 to 2020.

• ①  
  Ecosystem service assessment

See table 3 for details.

• ②  
  Comprehensive ecosystem service assessment

Table 3. Calculation methods of ecosystem services.

Ecosystem Services	Calculation methods	Main parameters and processing
Carbon Storage	InVEST model carbon storage module	The carbon storage module uses the average carbon density of four carbon pools of different land use types to multiply the area of different land use types to calculate the carbon storage of the ecosystem (Nelson and Daily 2010). The calculation formula is as follows:
		${C}_{{total}}={C}_{{above}}+{C}_{{below}}+{C}_{{soil}}+{C}_{{dead}}$ (1)
		Where (4), ${C}_{{total}}$ is the total carbon storage $\left(t/{{hm}}^{2}\right),{C}_{{above}}$ is the aboveground carbon storage, ${C}_{{below}}$ is the underground carbon storage, ${C}_{{soil}}$ is soil carbon storage, ${C}_{{dead}}$ is the carbon storage of dead organic matter. The carbon density value used in the module mainly refers to the relevant research of Chuai et al (2013) and determine the carbon pool table
Crop Production	statistical yearbooks	Obtain the grain (grain) output at the county level according to the regional statistical yearbooks
Habitat Quality	InVEST model habitat quality module	In this study, bulid-up land and arable land are set as threat sources (He et al., 2016). The calculation formula of habitat quality is as follows:
		${Q}_{{xj}}={H}_{j}\times \left[1-\frac{{D}_{{xj}}^{z}}{{D}_{{xj}}^{z}+{k}^{z}}\right]$ (2)
		Where (5), ${Q}_{{xj}}$ is the habitat quality of grid $x$ with land use type, ${H}_{j}$ is the habitat suitability of land use type $j,$ the range is 0–1, and 1 indicates the highest habitat suitability, $k$ is a semi saturation constant, half of the maximum habitat stress value, $Z$ is a normalized constant, usually set to 2.5. ${D}_{{xj}}$ is the habitat stress level of grid $x.$ Required Habitat threat table and sensitivity table refer to Xu et al (2018) and assign values according to the actual situation of the study area
Water Yield	InVEST model water production module	Based on the water balance method, the water yield is obtained by calculating the difference between the precipitation and the actual evapotranspiration of each grid (Wang et al 2022b) in the study area. The specific calculation formula is as follows:
		${Y}_{x}=\left(1-\frac{{{AET}}_{x}}{{P}_{x}}\right)\times {P}_{x}$ (3)
		Where (6), ${Y}_{x}$ is the annual water yield of grid $x$ $\left({mm}\right),$ ${{AET}}_{x}$ is the actual evapotranspiration of grid $x$ $\left({mm}\right),$ ${P}_{x}$ is the annual precipitation of grid $x$ $\left({mm}\right)$
Soil Conservation	InVEST model soil loss module	The modified universal soil loss equation (RUSLE) is used to estimate the amount of soil conservation (Sieber and Pons, 2015). The calculation formula is as follows:
		${SC}=R\times K\times {LS}\times \left(1-C\times P\right)$ (4)
		Where (7), ${SC}$ is soil conservation amount $\left({t}{{hm}}^{2}{a}^{-1}\right),$ $R$ is rainfall erosivity $\left({MJ\; mm}{{hm}}^{-2}{h}^{-1}{a}^{-1}\right),$ $K$ is soil erodibility $\left({t}{{hm}}^{2}{h}{{hm}}^{-2}{{MJ}}^{-1}{{mm}}^{-1}\right),$ $C$ is the vegetation coverage factor, $P$ is the water and soil conservation measure factor, ${LS}$ is the slope length and gradient factor, which are automatically calculated by the InVEST model. Watershed are extracted by SWAT model

Studies have assessed carbon storage, crop production, habitat quality, water yield and soil conservation in different dimensions. To comprehensively analyze the ecosystem service functions of the Jianghan Plain in China in 2035, it was necessary to eliminate the dimensions between ecosystem services, standardize the five services with values ranging from 0 to 1, and then multiply them by their respective weights to obtain a comprehensive ecosystem service (Wu et al 2021).

$$\begin{eqnarray}&&{{ES}}_{x}=\frac{{E}_{x}-{E}_{\min }}{{E}_{\max }-{E}_{\min }}\end{eqnarray} \tag{ 5 }$$

In the formula, ${{ES}}_{x}$ is the standard value of the x coefficient, ${E}_{x}$ is the original value of the $x$ coefficient, ${E}_{\min }$ is the minimum value of all coefficients, and ${E}_{\max }$ is the maximum value of all coefficients.

Based on the special conditions of the study area and the results of previous studies (Wu et al 2019), we assigned values to five ecosystem services in the Jianghan Plain of China, as shown in table 4. The calculation formula for comprehensive ecosystem services is as follows:

$$\begin{eqnarray}&&{{TES}}_{j}=\displaystyle \sum _{i=1}^{n}{w}_{i}\times {S}_{{ij}}\end{eqnarray} \tag{ 6 }$$

In the formula, ${{TES}}_{j}$ is the comprehensive ecosystem service of scenario $j,{w}_{i}$ is the weight of ecosystem service ${i},$ and ${S}_{{ij}}$ is the standardized assessment of ecosystem service ${i}$ in scenario ${j}.$

3.2.1. Ecosystem services trade-off and synergy analysis

4.1.1. Analysis of land use characteristics on the jianghan plain of China from 2000 to 2020

The spatial distribution of land use in the Jianghan Plain of China from 2000 to 2020 is shown in figure 3. The land use types in the study area are mostly arable land and water land, followed by build-up land and forest land. Between 2000 and 2020, the arable land in the study area decreased by 1008.84 km², grass land and unused land showed a decreasing trend, and build-up land, water land and forest land showed an increasing trend. In the past 20 years, build-up land has increased by 542.53 km², and the growth rates of build-up land, water land and forest land were 23.43%, 12.77% and 5.71%, respectively. From 2000 to 2020, the area of arable land transferred out was 1367.82 km², with a transfer contribution rate of 62.11%, which was 4.9 times the transferred area, of which 44.45% was converted into water land areas. The area of forest land transferred in was 64.24 km², which was higher than the area transferred out of 41.23 km². The area mainly transferred from arable land was 23.01 km². Grass land and unused land showed weaker changes between 2000 and 2020. The transfer area of 857.80 km² was much larger than the transfer area of 322.84 km², with the transfer contribution rate reaching 39.76%. The built-up land shows a substantial growth trend, with the transferred area being 428.77 km² and the transferred area being as high as 971.30 km², of which the transferred area of arable land accounted for 32.57%.

Zoom In Zoom Out Reset image size

4.1.2. Land use in the jianghan plain of China under different scenarios in 2035

The spatial pattern of land use in the Jianghan Plain of China under different scenarios in 2035 is shown in figure 4. Under the natural development scenario, arable land was the dominant land type in the Jianghan Plain of China, with an area of 20,133.69 km², accounting for 61.60% of the entire study area, followed by water land with an area of 5928.16 km², accounting for 18.14% of the total area, and built-up land area accounting for 13.71%. Compared with the land use in 2020, arable land, grass land and unused land showed a decreasing trend, with decreases of 900.14km², 16.23 km² and 31.11 km², respectively. Water land, forest land and build-up land areas showed increasing trends, increasing by 1205.37 km², 373.03 km² and 1622.03 km², respectively. Under the arable land protection scenario, the arable land area accounts for 67.19% of the entire regional land area, compared to 2020, the arable land increased by 1217.08 km², and the built-up land increased by 866.34 km², which was lower than the natural development scenario. The ecological protection scenario was an ideal scenario for future land use changes. Under this scenario, the natural landscape generally developed optimally. Ecological lands such as grass land, forest land and water land have been protected. Compared with 2020, grass land, forest land and water land increased by 15.66 km², 64.32 km² and 273.32 km², respectively, while arable land decreased by 872.58 km². Build-up land showed an increasing trend, but the increase rate was also much lower than that of 2020 under natural development scenarios. Under the urban development scenario, the increase in build-up land area was higher than that under the other three scenarios, with an area of 5004.38 km². Forest land and grass land were the major contributors to the increase in build-up land area. Compared with the natural development scenario, the two land types had a greater contribution to the increase in build-up land area. The proportion of volume contribution increased to 10.09% and the amount of water transfer accounted for 6.36% of the increase in built-up land area.

Zoom In Zoom Out Reset image size

4.2.1. Temporal and spatial changes in ecosystem services in the Jianghan Plain of China from 2000 to 2020

From 2000 to 2020, the ecosystem services in the Jianghan Plain of China generally showed fluctuating change characteristics (table 5). Carbon storage decreased from 2.34*10⁸ t in 2000 to 2.31*10⁸ t in 2020, a decrease of 0.05%, mainly due to the continuous expansion of build-up land, which led to a decrease in land use types with higher carbon density, such as arable land and grass land. Crop production showed a trend of first decreasing and then increasing. From 2000 to 2005, it dropped from 1.17*10⁸ kg to 1.00*10⁸ kg, and then increased by 1.41*10⁸ kg from 2010 to 2020. The main reason was that sudden climate changes and heavy rainstorms affected grain output in 2005 (Hao et al 2019). From 2000 to 2020, the habitat quality of the study area showed a downward trend, and its average score dropped from 0.43 to 0.41. The study area was greatly affected by human activities. Rapid urbanization has led to increased fragmentation and ecological sensitivity of habitat patches and a decline in habitat quality. Soil conservation represented by water and soil processes increased from 5.62*10¹⁰ kg in 2000 to 0.86*10¹⁰ kg in 2015, and then increased to 2.47*10¹⁰ kg in 2020, an increase of 2.9 times, indicating that the Jianghan Plain of China implemented the Yangtze River protection policy in 2016, and the small watershed, through the implementation of comprehensive management, afforestation and other ecological projects, water and soil conservation work, has achieved remarkable results (Zheng et al 2020). Water yield changed greatly between 2000 and 2020. In 2000, the annual water yield service in the study area was 1.34*10¹⁰ t. In 2020, it increased significantly by 49.89% compared with 2000, mainly due to changes in the surface impermeable layer and rainfall.

Table 5. Statistical of Total Ecosystem Services in the Jianghan Plain of China.

Year	Carbon storage	Crop production	Habitat quality	Water yield	Soil conservation
2000	2.34*108	1.17*108	0.43	1.34*1010	5.62*1010
2005	2.37*108	1.00*108	0.42	1.38*1010	5.76*1010
2010	2.33*108	1.10*108	0.42	1.57*1010	6.45*1010
2015	2.30*108	1.32*108	0.47	1.77*1010	6.48*1010
2020	2.31*108	1.41*108	0.41	1.14*1011	8.95*1010
2000–2020	−5.09%	20.34%	−4.91%	749.89%	59.11%

From 2000 to 2020, the spatial pattern of ecosystem services in the Jianghan Plain of China was basically stable, but spatial changes showed differential characteristics (figure 5). The spatial distribution of carbon storage was high in the west and low in the east. The high-value areas of carbon storage were concentrated in the western part of the study area. The main reason was that the climate in this area was humid, the land use types were basically forest land and grass land, and the vegetation coverage and soil organic matter were high, the highest average carbon storage from 2000 to 2020 was in Dangyang city (average 378.163 t/hm²) and Songzi city (average 353.217 t/hm²) in Hubei Province. The vegetation coverage rate in the eastern part of the study area was low, human activities were frequent, the proportion of build-up land in land use was high, and carbon storage was relatively low, with the lowest in Honghu city from 2000 to 2020 (average 152.334 t/hm²). Crop production showed a spatial distribution pattern of high in the north and low in the west. High-value areas were mainly distributed in the northern plains of the study area, generally higher than 15 t/hm². Low-value areas were mainly concentrated in the western mountainous and hilly areas, generally lower than 5 t/hm². This was mainly because the northern plain area had flat terrain and a large arable land area, while in the western mountainous areas, the terrain was more undulating, the forest land coverage rate was high, and the arable land area was smaller. The distribution of habitat quality and carbon storage was relatively consistent and was spatially characterized by a gradient of higher values in the west and lower values in the east. Areas with higher habitat quality were distributed in Dangyang city, Songzi city and Gongan County in the central part (the average value is above 0.50), mainly due to the high vegetation coverage in this area, which was less disturbed by human activities. The areas with the lowest habitat quality were mainly concentrated in Tianmen city, Rongjiang city and Xiantao city. The area has experienced rapid social and economic development in recent years, rapid expansion of build-up land, and a high proportion of arable land. Areas with high soil conservation values are mainly concentrated in the western plain area of the study area (average soil conservation value was 5500 t/hm²). The land types in these areas were mostly forest and grassland, and the soil conservation capacity was strong. The southern urban area and surrounding counties and cities had flat terrain and low soil retention capacity. From 2000 to 2020, the spatial differences in water yield were relatively obvious, but they all showed the characteristics of high in the north and low in the south. Compared with 2000, the water yield in all counties and cities in the study area increased in 2020, and the number of high-value areas increased significantly. The main reasons for changes in water yield were affected by land use, precipitation and topography (Krois and Schulte 2014).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

4.2.2. Changes in ecosystem services in the Jianghan Plain of China under different scenarios in 2035

Ecosystem services under the four scenarios simulated in the Jianghan Plain of China in 2035 all had different changing trends (figure 6). Among them, carbon storage and water yield showed a downward trend in the natural development, arable land protection, ecological protection and urban development scenarios. Carbon storage was in the range of 0–300 t/hm², and the average total carbon storage was 1.87 *10⁸t. Compared with 2020, the four scenarios decreased by 7.30%, 3.01%, 0.76% and 1.45%, respectively, the water production was between 0–1938 mm, and the average total water yield was 1.07*10¹¹ m³. Compared with 2020, the four scenarios decreased by 6.10%, 1.69%, 1.32% and 1.50%, respectively. The carbon storage and water yield under the natural development scenario were the smallest, while the carbon storage and water yield under the ecological protection scenario were the largest. Crop production was in the range of 0–22 t/hm², and the average crop production was 1.01*10⁸ t. Compared with 2020, the natural development, ecological protection and urban development scenarios decreased by 0.37%, 0.61% and 0.59%, respectively, while in arable land protection, the growth rate was 0.83% under the scenario. Soil conservation was in the range of 0–8900 t/hm², and the average soil conservation volume was 5.03*10¹⁰ t. Compared with 2020, the growth rates of the soil conservation arable land protection and ecological protection scenarios were 0.22% and 0.54%, respectively, and the natural development scenario decreased by 0.83%, while it remained basically unchanged under the urban development scenario, with a decrease of less than 0.01%. Habitat quality was in the range of 0–1, and the average habitat quality score was 0.41. Compared with 2020, the natural development, arable land protection and urban development scenarios decreased by 6.38%, 2.19%, and 1.80%, respectively, while under the ecological protection scenario, it grew slowly at 0.52%.

Zoom In Zoom Out Reset image size

In 2035, the comprehensive ecosystem services of the Jianghan Plain in China showed a downward trend (figure 7). The comprehensive ecosystem services of the four scenarios of natural development, arable land protection, ecological protection and urban development decreased from 0.186 in 2020 to 0.174, 0.181, 0.183 and 0.180, respectively. The reason was that in terms of water yield, under natural scenarios, water yield showed a spatial pattern of high in the northwest and low in the southeast. The low-value area of ​​water yield in Wuhu city in the southeast has expanded greatly. Under the arable land protection, ecological protection and urban development scenarios, the decline in water yield was smaller than that under the natural development scenario, indicating that the implementation of urban planning and environmental protection policies in the study area partially reduced the decline in water yield services. In terms of carbon storage, the carbon storage space under the four scenarios changed little, mostly in the northwest and in the southeast. The carbon storage capacity of the arable land protection and ecological protection scenarios was significantly better than that of the natural development and urban development scenarios. In terms of crop production, crop production decreased in the natural development, ecological protection and urban development scenarios, while it increased in the arable land protection scenario. This was mainly due to the expansion of build-up land in the study area and the occupation of a large amount of arable land that could provide crop production services. In terms of soil conservation, the spatial pattern under the four scenarios showed the characteristics of high in the northwest and low in the southeast. The order of soil conservation capacity under different scenarios was ecological protection > arable land protection > natural development > urban development. The total soil conservation capacities under the four scenarios in 2035 were 5.15*10¹⁰ t, 5.12*10¹⁰ t, 5.01*10¹⁰ t and 4.84*10¹⁰ t. Compared with 2020, there were downward trends to varying degrees. This showed that under the natural development and urban development scenarios

Zoom In Zoom Out Reset image size

in 2035, soil conservation in the study area continued to decrease, and soil loss gradually increased. Under the arable land protection and ecological protection scenarios, soil conservation services showed a slight upward trend. Planning constraints and ecological protection policies were conducive to enhancing the regional soil holding capacity. In terms of habitat quality, the ecological protection scenario showed an upward trend in habitat quality, while the natural development, arable land protection and urban development scenarios declined. From the spatial pattern, the low-value areas were mainly distributed in Qianjiang city and Xiantao city in the central Jianghan Plain of China, and the northern water land in Yunmeng County had fewer areas. The high-value areas Dangyang city and Songzi city had excellent habitat quality and high biodiversity. It could be seen from this that the proportion of woodland and grassland greatly impacted habitat quality. Rapid social development has led to build-up land gradually encroaching on habitat-friendly forest land and grass land.

There were differences in the synergistic effects weighed under different scenarios from 2020 to 2035 (figure 8). From 2020 to 2035, there was a trade-off relationship between crop production and carbon storage, soil conservation, habitat quality, and water yield in the Jianghan Plain of China. Moreover, between carbon storage and soil conservation, habitat quality, and water yield, crop production, and soil conservation, there was a collaborative relationship. Under the natural development scenario, the land use change pattern continued to trend from 2000 to 2020. Build-up land expanded, grass land and arable land areas decreased, the relative benefits of carbon storage and crop production weakened, and their trade-off effects were reduced. The trade-off effects between other ecosystem services were also reduced and enhanced to varying degrees. Under the arable land protection scenario, the increase in arable land area caused the relative returns of crop production to be the best, moreover, the increase in soil erosion and water and soil loss indicated that the increase in the intensity of arable land use weakened its services. The relative returns of the two ecosystem services weakened, crop production significantly enhanced the trade-off between soil conservation and habitat quality, the decline rate of carbon storage was lower than that of the natural development scenario, and the trade-off effect between the two was higher than that of the natural development scenario. Under the ecological protection scenario, crop production decreases, and the relative benefits of carbon storage, soil conservation, and habitat quality increased significantly, which strengthened the trade-off effect between crop production and carbon storage, but the trade-off effect was lower than that of the arable land protection scenario. Crop production, soil conservation and carbon storage and the trade-off effect between water yield, carbon storage and habitat quality were weakened. Under the urban development scenario, except that the synergistic relationship between water yield and soil conservation gradually increased, the synergistic relationship between the other three ecosystem services decreased. In summary, land use change importantly impacted the trade-off relationship between ecosystem services. Arable land, forest land and build-up land were important factors affecting the trade-off effect between ecosystem services in the Jianghan Plain of China. The trade-off effect between ecosystem services under the arable land protection scenario was enhanced, and soil conservation and habitat quality were significantly degraded, posing a threat to the security of ecosystem services in the study area. Although the trade-off effect between ecosystem services was weakened under the ecological protection scenario and the relative benefits of carbon storage, soil conservation and habitat quality increased, and the transfer of arable land and arable land to ecological land such as forest land and water land caused a decline in water yield services. Thus, sufficient attention should be given to grain and water shortage risks under ecological protection scenarios.

From 2000 to 2020, the ecosystem structure of the Jianghan Plain in China underwent significant changes (figure 3). Compared with 2000, the area of arable land in the study area decreased by 10.89%, the area of grass land decreased by 0.024%, the area of unused land decreased by 0.13%, and the area of forest land decreased by 10.89% in 2000. The area increased by 0.23%, the water land area increased by 5.35%, and the build-up land area increased by 5.43%. The main reason for this is that since the late 1990s, the number of cities and population size in the Jianghan Plain of China have grown rapidly, and build-up land has expanded on a large scale, resulting in arable land, grass land and unused land. Since 2006, the country has focused on the protection of the Yangtze River Basin and promoted policies related to returning grain for green and lakes. The area of forest land and water land has increased. For this reason, it can be concluded that changes in the ecosystem structure can lead to increased production of ecosystem services (Xiao et al 2022b). Therefore, to ensure ecosystem services and the sustainable use of land resources, it is recommended to strictly control the occupation of arable land for construction, ensure that high-quality arable land is not occupied, prohibit uncontrolled land reclamation, and avoid intensification of water and soil erosion (Chen et al 2021). This can be achieved by reasonably planning land resources, revitalizing unused land, and selecting appropriate areas for transformation and utilization (Lu et al 2022). High-value areas of ecosystem services should be established as ecological reserves to enhance the ecological functions of the area (Zhang et al 2018, Yuan et al 2023).

In the past 20 years, the rapid social and economic development in the Jianghan Plain of China has intensified the transformation of land use types, which is mainly characterized by urban expansion leading to a significant increase in build-up land and a substantial decrease in arable land area (figure 3). Land use changes affect ecosystem services (figure 5). For example, when arable land is converted into build-up land, build-up land can prevent soil from being exposed on the surface, reduce soil erosion, and improve soil conservation. However, it will also lead to decreases in carbon storage, habitat quality, and water yield (Wu et al 2019, Wang et al 2022a). After arable land is converted into forest land, forest vegetation with well-developed root systems can effectively increase carbon storage, habitat quality, soil conservation, and water yield. These results are consistent with the research results of Xue et al (2023) and Jose (2009). Under the four scenarios from 2020 to 2035, there was a very significant synergistic relationship between soil conservation and carbon storage, habitat quality, and water yield, but there was no trade-off relationship. This is consistent with Chen et al (2022) and Yang et al (2018), who found consistent results in the conversion of grain for green. In this study, arable land and build-up land together accounted for more than 65% of the city and county area of Jianghan Plain of China. However, due to the poor vegetation coverage and soil resistance to rainfall erosion of these two land types, as well as the high degree of human interference, Jianghan Plain of China's ability to provide carbon sequestration, water yield, soil conservation and habitat is relatively weak. This is why there is an obvious synergistic relationship between the five services of carbon storage, crop production, habitat quality, soil conservation and water yield in the Jianghan Plain of China without trade-offs. However, this result is different from the study of Xu et al (2018) in the Yangtze River Economic Zone, which is mainly due to the spatial heterogeneity of the ecosystem and the dominant role of forest land in regulating and supporting ecological functions due to the area of forest land in different study areas.

The results showed that due to the impact of human activities and the implementation of a series of policies, the spatial structure of the ecosystem in the Jianghan Plain of China has undergone significant changes in the past 20 years, which is consistent with the results of Ren et al (2022). Land use management needs to comprehensively consider the possible outcomes of all envisioned scenarios and incorporate ecosystem service assessment into land use planning and management (Martínez-Sastre et al 2017, Wu et al 2019). Based on the potential changes in land use and previous research results (Liu et al 2017, Fang et al 2021, Zhang et al 2023), we propose 4 land use scenarios to provide a better understanding of land use in the Jianghan Plain of China and provide a framework for the future development of management.

Under the natural development scenario, the land use changes from 2020 to 2035 will continue, the continued expansion of cities will lead to a significant reduction in comprehensive ecosystem services, and the synergistic relationship between ecosystem services will be significantly weakened. Under the arable land protection scenario, the conversion probability of arable land to other land uses is limited, the grain supply capacity is increased, and the soil conservation and habitat quality capabilities are sharply reduced, indicating that stabilizing arable land to increase crop production may damage ecosystem regulating services and enhance ecosystem service trade-off effects. Under the urban development scenario, the scale of build-up land tends to be stable, but due to the reduction in forest land and grass land, the comprehensive ecosystem services are reduced, and the synergistic relationship between ecosystem services mainly shows a weakening trend. Compared with the urban development scenario, urban expansion is not effectively controlled under the ecological protection scenario, but the protection of forest land, grass land, and water land maximizes comprehensive ecosystem services, and the trade-offs and synergies between ecosystem services show a relatively healthy development trend. This solution is more conducive to the healthy development of urban ecosystems and the solution of ecological security issues. Therefore, to achieve sustainable land use and improve ecosystem services in the Jianghan Plain of China, we propose the following suggestions based on previous studies (Zheng et al 2019, He et al 2023) and relevant policies such as the 'Three Zones and Three Lines' in the study area: (1) Type of build-up land: Control the speed of urban expansion and the total amount of construction land limited to 3000 km². Green space construction should be increased in the urban areas of Jingzhou District and Shashi District to alleviate the negative impact on ecosystem services, strictly control various construction activities outside the urban development boundary, develop and utilize the spatial resources of Hanchuan city in the central urban area and surrounding Yunmeng County, strengthen the comprehensive utilization of space resources, and consider the urban development model as the leading model in four counties and cities. (2) Type of arable land: The arable land area is not less than 5200 km². Improve arable land quality to enhance arable land biodiversity and other ecosystem services. The arable land in Tianmen city, Qianjiang city and Yingcheng city in the north is recommended to grain for green. The arable land in Jiangling County, Jianli County and Gong'an County in the central part of the country is included in the arable land red line as the main grain-producing area. Additionally, six counties and cities are considered guided by the arable land protection and development model. (3) Ecological land types: The ecological red line range should not be less than 2,500 km², of which the area of forest land and grass land should not be less than 600 km², and the water land should not be less than 1,800 km². Dangyang city and Zhijiang city in the northwest and Songzi city in the southwest are areas with high comprehensive ecosystem services. We recommend adjusting the existing ecological red lines and natural protection areas to improve protection efficiency and strengthen the cultivation of water conservation forests in the eastern part of the department, control development activities around water source protection areas, and consider the ecological protection scenario development model of the three cities as the leading model. (4) Types of unused land: The area of unused land is limited to 50 km², and the transformation and utilization of unused land such as abandoned factories and bare land in Xiantao city and abandoned open-pit mines in Shishou city and Honghu city in the south are strengthened. In future land spatial planning and management, we should adhere to the ecological priority strategy, properly coordinate the delineation of arable land protection red lines, ecological red lines and urban expansion boundaries, optimize the allocation of land resources, reduce the trade-off effect between ecosystem services and improve their synergistic effects. Coordinated and parallel economic development and ecological protection should be achieved.