Wetland and habitat dynamics in the evolving landscape of the Middle Yangtze River Basin

YANG Peng, SUN Kaiya, ZHU Yanchao, XIA Jun, HUANG Heqing, SONG Jingxia, SHI Xiaorui, LU Xixi

Journal of Geographical Sciences ›› 2025, Vol. 35 ›› Issue (1) : 88-111.

PDF(192397 KB)
Adv Search
EN中文
PDF(192397 KB)
Journal of Geographical Sciences ›› 2025, Vol. 35 ›› Issue (1) : 88-111. DOI: 10.1007/s11442-025-2314-7
Special Issue: Climate Change and Water Environment

Wetland and habitat dynamics in the evolving landscape of the Middle Yangtze River Basin

Author information +
History +

Abstract

Wetlands play a critical role in the global environment. The Middle Yangtze River Basin (MYRB), known for its abundant wetland resources, has experienced notable changes resulting from the complex interplay of environmental factors. Therefore, we investigated the spatiotemporal characteristics of wetland ecological quality in the MYRB from 2001 to 2020. Utilizing the random forest (RF) regression algorithm and patch-generated land-use simulation (PLUS) model, we forecasted variations in wetland habitat quality and their determinants under the Shared Socioeconomic Pathway-Representative Concentration Pathway (SSP- RCP) framework from 2035 to 2095. The main findings are as follows: (1) The RF algorithm was optimal for land-use and land-cover (LULC) classification in the MYRB from 2001 to 2020, when notable changes were observed in water bodies and buildings. However, the forested area exhibited an increase and decrease of 3.9% and 1.2% under the SSP1-2.6 and SSP5-8.5 scenarios, respectively, whereas farmland exhibited a diminishing trend. (2) Wetlands were primarily concentrated in the central and eastern MYRB, with counties in the southwest exhibiting superior ecological-environmental quality from 2001 to 2020. Notably, wetland coverage revealed significantly high level, significant changes, frequent but relatively minor changes under the SSP1-2.6, SSP2-4.5, and SSP 5-8.5 scenarios, respectively. (3) Regions with lower habitat quality were primarily concentrated in urbanized areas characterized by frequent human activities, indicating a clear degradation in habitat quality across different scenarios. In conclusion, we established a foundational framework for future investigations into the eco-hydrological processes and ecosystem quality of watersheds.

Key words

Middle Yangtze River Basin / wetland / environmental change / habitat quality / multiple scenarios

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations
YANG Peng, SUN Kaiya, ZHU Yanchao, XIA Jun, HUANG Heqing, SONG Jingxia, SHI Xiaorui, LU Xixi. Wetland and habitat dynamics in the evolving landscape of the Middle Yangtze River Basin[J]. Journal of Geographical Sciences, 2025, 35(1): 88-111 https://doi.org/10.1007/s11442-025-2314-7

1 Introduction

Ecosystems are indispensable resources serving as pivotal open spaces for human survival and development (Verburg et al., 2011; Deng et al., 2019; Qi et al., 2024; Shi et al., 2024). Wetlands have historically been recognized as critical ecosystems, often referred to as habitats for humans and diverse flora and fauna, as well as the “kidneys of the Earth” (Zhang et al., 2022b). Wetlands, covering roughly 4%-6% of the Earth’s surface, are a crucial land-use type and one of the most productive and complex ecosystems, and changes in their areas continuously impact the local ecological integrity and habitat quality (Wang et al., 2011).
However, significant changes have occurred in land use and land cover (LULC), resulting in severe wetland losses influenced by climate change and human activities (Yue et al., 2014; Liu et al., 2022; Yang et al., 2023). Moreover, alterations in land use indirectly influence broader ecosystem dynamics, posing substantial challenges to ecological security (Wang et al., 2012). For example, a series of human activities (i.e., urbanization, land occupation, and utilization of forests and grasslands) has led to a substantial deterioration in habitat quality (Bai et al., 2019; Song et al., 2020; Wei et al., 2024), which profoundly affects human living environments; a reduction in green living spaces poses threats to life and property security (Zhu et al., 2014). Since the 19th century, approximately 50% of Earth’s wetlands have been converted for industrial, agricultural, and residential applications (Sivakumar et al., 2016). Consequently, wetland security has emerged as a pressing issue in contemporary environmental sciences. Furthermore, although analyzing the historical evolutionary characteristics of wetlands can effectively reveal the temporal and spatial patterns of wetland changes, understanding the mechanisms, characteristics, and developmental trajectories of these changes necessitates the integration of future scenarios related to wetland habitats and alterations in ecological quality.
The Middle Yangtze River Basin (MYRB) features numerous lakes and wetlands surrounded by mountains and rivers, capturing the robust functions of water source conservation, biological propagation, and environmental purification (Wang et al., 2022; Zhang et al., 2022b; Cai et al., 2023). Recognized as a crucial gene bank and ecological barrier in China, the diverse geographical conditions of wetlands in the MYRB have amplified the significance of their health (Zhang et al., 2022b, 2023a). Therefore, comprehensive strategies for ecosystem protection in the MYRB have been proposed, considering the characteristics of wetland degradation under different scenarios and by analyzing the driving forces, processes, and mechanisms of such degradation (Zhang et al., 2022b). Rapid historical urbanization and dynamic changes in land types pose significant threats to habitat quality and the ecological environment of the Yangtze River Basin (YRB) (Chen et al., 2020a; Zhang et al., 2021b, 2022). However, research on the relationship between LULC and ecological environmental quality, as well as on how future urbanization and LULC changes will impact ecological environmental quality in the MYRB is still lacking. Therefore, considering the socioeconomic conditions and natural factors, studying wetland changes within the context of LULC changes in the MYRB is essential for China’s economic development and ecological research.
To understand the dynamics of LULC in the MYRB, investigate the future ecological health levels of wetlands under changing environmental conditions, establish a theoretical foundation, and offer decision-making support for the protection of the Yangtze River, we addressed the following key questions: (1) How does LULC in the MYRB change under the influence of human activities and climate change? (2) What are the historical and future scenarios for the ecological indices of wetlands in the MYRB under LULC change? (3) How does habitat quality in the MYRB change in response to environmental changes? This study not only contributes essential data support for broader Yangtze River conservation efforts but also provides valuable insights into the future habitat quality and ecological changes of small-to medium-scale wetlands.

2 Study area and data

2.1 Study area

The MYRB, spanning from Yichang to Hukou, encompasses the main course of the Yangtze River, Dongting Lake, Poyang Lake, Jianghan Lake, as well as their associated tributaries and floodplains (24°29ʹ-34°12ʹN, 106°5ʹ-118°36ʹE) (Zhao et al., 2014) (Figure 1). Characterized by consistent forest coverage exceeding 55%, this region is a major ecological conservation region in China (Zhang et al., 2021a). It accounts for 14.7% of the total length and 37.6% of the entire area of the YRB, which encompasses nine provinces (Gupta et al., 2012). The MYRB experiences a typical subtropical monsoon climate, featuring the coldest and warmest temperatures in January and July, respectively, and average annual precipitation ranging from 700 to 2000 mm (Wang et al., 2021).
Figure 1 Overview of the study area (Middle Yangtze River Basin)

Full size|PPT slide

However, extensive land reclamation and developmental activities during the mid-20th century led to reductions in the areas of major lakes in the Yangtze River Plain (Zhang et al., 2022a). These areas have diminished from 5190 and 4350 km2 to 2625 km2, respectively (Zhang et al., 2022a, 2023b). This phenomenon poses a significant threat to the ecological security of the MYRB, with discernible repercussions on the regional ecological environment attributed to human activity (Xu et al., 2020).

2.2 Data

2.2.1 Remote sensing data

To ensure both clarity and adherence to symbol conditions across different years, in this study, we utilized the Google Earth Engine (GEE) platform to acquire surface reflectance composite images for 2001, 2005, and 2009 from the Landsat 5 satellite, as well as for 2013, 2017, and the period between June and August 2020 from the Landsat 8 satellite (https://earthdata.nasa.gov/) (Table 1).
Table 1 Number of different satellite images used in this study
Sensor 2001 2005 2009 2013 2017 2020
Landsat5 245 215 312 269 - -
Landsat8 - - - - 362 317

2.2.2 Fundamental geographic data

The transformation of LULC is a confluence of multifactorial interactions, necessitating the consideration of various natural, social, and economic elements (Hong et al., 2021; Liang et al., 2021). In this study, 16 factors (precipitation, temperature, elevation, slope, soil type, gross domestic product (GDP), population, and distances to different land features based on Euclidean distance) were chosen as fundamental geographic data to act as predictive driving forces for LULC (Table 2). The spatial distribution dataset of GDP in China at a grid resolution of 1 km was generated based on county-level GDP statistical data across the country (Xu, 2017a). It incorporates spatial interaction patterns between the GDP and factors closely related to human activities, such as land-use type, nighttime light intensity, and population density, resulting in a 1 km×1 km grid (Xu, 2017a). The spatial distribution dataset of China’s population at this grid resolution was based on county-level population statistics (Xu, 2017b). It considers factors affecting population distribution, such as land-use type, night time light intensity, and population density, through spatial techniques to produce multi-year population distribution grids (Xu, 2017b).
Table 2 Basical geographic data
Dataset Year Spatial resolution Sources
Elevation 2013 30 m×30 m https://search.earthdata.nasa.gov/
Lake 2015 ‒ https://www.webmap.cn/
Road 2020 - https://www.openstreetmap.org/
Highway 2020 - https://www.openstreetmap.org/
Railway 2020 - https://www.openstreetmap.org/
Government position - - https://lbsyun.baidu.com/
GDP 2015 1 km×1 km https://www.resdc.cn/
Population 2015 1 km×1 km https://www.resdc.cn/
Soil - 1 km×1 km https://www.resdc.cn/
Precipitation 2015 1 km×1 km https://www.resdc.cn/
Temperature 2015 1 km×1 km https://www.resdc.cn/
Water bodies 2020 30 m×30 m https://global-surface-water.appspot.com/
Elevation data, obtained from the United States Geological Survey, and lake data, obtained from the National Geographical Information Resource Catalog Service System database, constitute the integral components of the primary geographic datasets. Data representing roads, railways, buildings, and water bodies were primarily sourced from the OpenStreetMap website and presented in the form of points, lines, and polygons. Critical socioeconomic indicators, such as GDP and population, along with environmental factors, such as soil type, precipitation, and temperature, were acquired from the Chinese Academy of Sciences (CAS) Resource and Environmental Science Data Center (RESDC) (https://www. resdc.cn/). Additionally, water body data were derived from the Global Surface Water dataset, which is recognized for its high resolution, contemporary temporal coverage, and extensive spatial reach, making it a staple in numerous studies.
Future projections of GDP and population under various Shared Socioeconomic Pathways (SSPs; i.e., SSP1, SSP2, and SSP5) were derived from a 0.5°×0.5° raster dataset provided by Jiang et al. (2022). Temperature and precipitation data for future scenarios were sourced from the Coupled Model Intercomparison Project Phase 5 (CMIP5) of the World Climate Research Program, especially the Representative Concentration Pathways (RCPs; i.e., RCP2.6, RCP4.5, and RCP8.5) available as a 0.5°×0.5° dataset (accessible at https://esgf-node.llnl.gov/search/cmip5/). To enhance the spatial resolution of such foundational data, we employed a growth rate of 0.5°×0.5° grid data from 2015, and spatial resolution population and GDP data from 2015 (1 km×1 km), obtained from the CAS RESDC (https://www.resdc.cn/). By employing a proportionate growth approach based on primary driving data from 2015, we conducted raster matrix operations to downscale the GDP and population for future scenarios in 2035, 2050, 2065, 2080, and 2095.

3 Methodology

We employed a combination of remote sensing imagery and machine learning techniques (classification and regression trees (CARTs), support vector machines (SVMs), gradient tree boosting (GTB), and random forest (RF)) to classify LULC. Subsequently, we simulated and predicted LULC data under future scenarios in the MYRB using geographical data and a patch-generated land-use simulation (PLUS) model. Subsequently, the ecological indices for historical scenarios in the MYRB were calculated based on the remote sensing ecological index (RSEI). Primary landscape indices, including wetland proportion, density-related indices (patch density, largest patch index, and number of patches), shape-related indices (mean fractal dimension index, mean shape index, mean perimeter-area ratio, and landscape shape index), and connectivity-related indices (patch contiguity index, cohesion index, landscape division index, aggregation index, circumscribing circle, and proportion of similar adjacencies) were computed. Principal component analysis (PCA) was employed to fit the density, shape, and connectivity indices. Considering the close relationship between ecological quality (i.e., RSEI) and landscape indices, the future RSEI in the MYRB was further simulated and predicted based on RF regression. Finally, we determined the quality of historical and future habitats and its variations in the MYRB based on the Integrated Valuation of Ecosystem Services and Trade-Offs (InVEST) model (Figure 2).
Figure 2 Technical roadmap for this research

Full size|PPT slide

3.1 Land classification

The process of land classification involved the computation of the modified normalized difference water index (MNDWI), normalized difference vegetation index (NDVI), and normalized difference built-up index (NDBI) data from remote sensing imagery. The NDVI and MNDWI are recognized for their effectiveness in vegetation and water body classification. The NDBI, designed to distinguish constructed land from other land types, was employed for its discriminative capabilities (Guha et al., 2021). These indices are expressed in Eqs. (1)-(3):
NDVI=ρNIR−ρRedρNIR+ρRed
(1)
where ρRed and ρNIR denote the reflectances of the red and near-infrared bands, respectively.
NDBI=ρSWIR1−ρNIRρSWIR1+ρNIR
(2)
where ρNIR and ρSWIR1 represent the reflectances of the near-infrared and shortwave infrared-1 bands, respectively.
MNDWI=ρGreen−ρMIRρGreen+ρMIR
(3)
where ρMIR and ρGreen denote the reflectances of the mid-infrared and green bands, respectively.
Additionally, PCA was employed to extract the first principal component (PC1) from 11 bands, including the bands of Landsat, NDVI, NDBI, MNDWI, slope, and elevation. PC1, which captured over 95% of the component variance, mitigated the correlation between land cover and topography without compromising the classification efficiency (Eq. 4), and is expressed as follows:
IMG=PC1(Bands,NDVI,NDBI,MNDWI,Slope,Elevation)
(4)
where Bands encompass the bands of the remote sensing image from Landsat, and Elevation and Slope denote the elevation and slope in the MYRB, respectively.
Land-cover types were classified into six categories (forest, grassland, wetland, farmland, buildings, and water bodies) based on the visual interpretation of true-colored images in 2017. Random sampling, ensuring both randomness and accuracy, was conducted considering the land-cover distribution and topographical features. Subsequently, the classification was employed to identify LULC in other periods.
Considering the significance of classification accuracy and substantial coverage of expansive vegetation and built-up areas in the MYRB, the acquired LULC data were categorized into six classes (i.e., forest, grassland, wetland, farmland, buildings, and water bodies) based on the standards of the CAS LULC dataset. The selection of training and testing samples followed a systematic approach, ensuring a representative distribution across diverse land-cover types. We allocated 80 and 20% of the samples for training and testing purposes, respectively, through machine learning methods and the GEE (Mccarty et al., 2020).

3.2 PLUS modeling

The PLUS model combines a rule-mining framework rooted in land-spread analysis strategies with a cellular automaton (CA) featuring diverse random patch seeds (Ning et al., 2018; Liang et al., 2021). Such integration facilitates the exploration of driving factors and extends the suitability probabilities for various types of land use and land cover (LULC), as determined by the RF classification, as expressed in Eq. (5).
Pi,kd(x)=   ∑n=1M⁡I(hn(x)=d)M
(5)
where Pi,kd(x) represents the final increasing probability of k in cell i, I(∙) is the indicator function regarding the decision tree, hn(x) denotes the predicted type by the n-th decision tree for vector x, and M is the amount of decision trees.
The composition of the LULC is shaped by top-down constraints, in contrast to the CA models used to simulate the LULC framework. This approach helps maintain the dominance of competitive adaptive inertia in future LULC simulation models, facilitated by an innovative threshold and multitype random patch-seeding mechanism (Liang et al., 2021), as expressed in Eq. (6).
OPi,kd=1,t={Pi,kd=1×(r×μk)×Dkt if Ω i,kt=0 and r<Pi,kd=1Pi,kd=1× Ω i,kt×Dkt  all others 
(6)
where Pi,kd=1 is the increasing probability, r is a random value between 0 and 1, and μk represents the threshold for creating new land-use patches of k.
Initially, we utilized the demand prediction function within the PLUS model by employing a Markov chain approach to forecast changes in the LULC features in the image. Subsequently, land classification in the MYRB for 2020 was simulated by leveraging the development probability derived from the land classification image of 2005 to evaluate the accuracy of simulated data. Finally, the simulated LULC data for 2020 were compared with land classification products released by the CAS for the same period (https://www.resdc.cn/).
To better analyze the spatiotemporal dynamics of LULC under future scenarios, we focused on five periods with 15-year intervals from 2035 to 2095. To ensure the efficient determination of LULC changes under future scenarios, we utilized the RF algorithm to calculate the contributions of various driving factors (e.g., GDP, population, temperature, and precipitation). The transition rules and weights between different land classes under different scenarios were determined based on previous experience and expert assessments (Fu et al., 2018).

3.3 Ecological quality index

In this study, the RSEI was employed to quantify the ecological conditions in the region, whereas the NDVI was employed to quantify the greenness of the area, which is closely associated with plant biomass, leaf area, and vegetation coverage, serving as a representative indicator of greenness.
The normalized difference band soil index (NDBSI) is a valuable tool for detecting and analyzing the effects of human activities on land surfaces, particularly in relation to bare soil exposure and its environmental consequences. This is a negative index (Guha et al., 2021).
BSI=(ρSWIR1+ρRed)−(ρNIR+ρBlue)(ρSWIR1+ρRed)+(ρNIR+ρBlue)
(7)
BI=2ρSWIR1ρSWIR1+ρNIR−(ρNIRρNIR+ρRed+ρGreenρGreen+ρRSWIR1)2ρSWIR1ρSWIR1+ρNIR+(ρNIRρNIR+ρRed+ρGreenρGreen+ρRSWIR1)
(8)
NDBSI=BSI+BI2
(9)
where ρBlue,ρGreen,ρRed,ρNIR, and ρSWIR1 are the reflectances of the blue, green, red, near- infrared, and shortwave infrared bands, respectively.
The land surface temperature (LST) index indicates the heat condition of the area (Guha et al., 2021).
Lλ=[εB(TS)+(1−ε)L↓]τ+L↑
(10)
B(Ts)=[Lλ−L↑−τ(1−ε)L↓]/τε
(11)
TS=K2/ln(K1/B(TS)+1)
(12)
where TS, ε, τ, B(TS), L ↑, Lλ, and L ↓ represent the actual surface temperature, land surface emissivity, atmospheric thermal infrared transmittance, black body radiance, upward atmospheric radiance, thermal infrared radiance, and downward atmospheric radiance, respectively.
The NDBSI, NDVI, LST, and MNDWI indices represent exposure, greenness, heat, and wetness, respectively. PCA was employed to extract the principal components of the four bands, with the maximum principal component chosen as the RSEI value. The equation for the RSEI is expressed as follows:
RSEI=PCA(NDVI,NDBI,MNDWI,LST)
(13)
To investigate landscape fragmentation and connectivity in the MYRB, the wetland proportion was calculated using the FRAGSTATS 4.2 software. PCA was employed to fit the density shape and wetland landscape connectivity indices.

3.4 Calculation of habitat quality using the InVEST 3.12 model

The InVEST model can quantify habitat quality as a continuous variable and incorporate the effects of LULC conditions and patterns. The model was used to compute habitat degradation in the MYRB considering both linear (Eq. (14)) and exponential decay (Eq. (15)) before calculating the habitat quality. Therefore, the calculation method for the degree of habitat degradation of pixel (x) in land-use type l is presented in Eqs. (14)-(16):
Dxl=∑Rr=1∑Yry=1(wr∑Rr=1wr)ryirxyβxSlr
(14)
irxy=1−(dxydrmax)
(15)
irxy=exp(−(2.99drmax)dxy)
(16)
where Dxl represents the degree of habitat degradation; R is the number of threat factors; wr is the impact weight of threat factor r; Yr is the total number of pixels for threat factor r in the remote sensing image of LULC; ry is the number of threat factors on a single pixel in the remote sensing image of LULC; irxy denotes the impact of threat factor r in pixel y on pixel x; dxy is the distance between image pixel x and threat factor pixel y;drmax is the impact range of threat factor r; βx represents the level of protection under local policies, and is set to 1 as land-use changes are influenced by local policies; Slr is the sensitivity of land class l to threat factor r.
Based on the habitat degradation quality Dxlz, the habitat quality can be expressed below:
Qxl=Hl(1−(DxlzDxlz+kz))
(17)
where Qxl is the habitat quality of pixel x in the LULC image l, Hl is the habitat attribute of LULC image l. k is a saturation constant usually set to half the maximum value of habitat degradation. The default parameter z in the InVEST 3.12 model is typically set to 2.5. The resulting habitat quality index values ranged from 0 to 1, with values closer to 1 indicating higher habitat quality.

3.5 Hotspot analysis

In this study, we employed hotspot detection to conduct a global spatial autocorrelation test on the MYRB, calculating the Local Moran’s I (G) based on remote sensing image pixels to indicate spatial autocorrelation (Sun et al., 2022). The calculation method is expressed in Eq. (18):
Gi∗=∑nj=1wij(xj−x¯)S[∑nj=1wij2−(∑nj=1wij)2]n−1
(18)
where Gi∗ is the local Moran’s I for region i, n is the number of pixels, xi and xj are the log-transformed values of the comprehensive dynamic attitudes of blocks i and j, respectively. wij is the spatial weight matrix, S is the standard deviation, and x¯ is the average of xi andxj. The larger the absolute value of Gi∗, the more credible is the result, indicating a non-random outcome.Gi∗>0 signifies a high degree of clustering in the area, making it a hotspot. Conversely,Gi∗<0 indicates a hotspot area with negative clustering.Gi∗=0 implies a random result without statistical significance.

4 Results

4.1 Remote sensing classification and simulation of land use in the MYRB

The computation yielded training and testing accuracies of 98.06 and 87.66%, respectively. The RF model demonstrated exceptional classification performance, achieving an accuracy of 87.66% and a kappa coefficient of 0.84, outperforming GTB, SVMs, and CARTs. Therefore, we successfully classified the land cover in the MYRB from 2001 to 2020 using the RF model (Figure 3). We found that approximately 60% of the forested area is concentrated in the northwestern and southwestern MYRB. Although farmland constituted a substantial proportion, accounting for approximately 27%, it trailed slightly behind forested land. Farmland is intricately interspersed around urban structures, primarily in the northwestern and southwestern regions. From 2001 to 2020, construction-designated land accounted for approximately 25% of the total area. Notably, construction land tended to be predominantly situated near water bodies and displayed distinct expansion patterns. The number of closely distributed wetlands slightly surpassed that of water bodies, primarily clustering around these water bodies in the central and eastern regions. Grasslands were primarily distributed near urban areas, accounting for a relatively small proportion of the total land cover. Additionally, the analysis of land transfer matrices indicated a significant reduction of 14.3% in construction land between 2001 and 2005, primarily owing to its conversion to farmland. Nevertheless, significant increases were observed in construction land (23%) and water bodies (14.5%) between 2013 and 2017, primarily resulting from farmland transformation. Forestland and farmland exhibited marginal increases of 0.5 and 1.6% from 2017 to 2020, respectively, with the majority originating from the conversion of construction land. In summary, from 2001 to 2020, water bodies increased by 46.2%, whereas forestland and wetlands showed modest increases of 0.4% and 1.6%, respectively.
Figure 3 Land use classification in the Middle Yangtze River Basin from 2001 to 2020

Full size|PPT slide

Figure 4 shows the projected land-cover outcomes under future scenarios in the MYRB. The findings indicated a consistently high percentage of forested areas, in contrast to relatively lower proportions of wetlands, water bodies, and grasslands. Under SSP1-2.6, from 2035 to 2095, an annual increase of 3.9% in forested areas was expected, along with a 9.2% reduction in farmland and 2.2% increase in built-up areas. Despite forests remaining the predominant land-use type, a discernible pattern emerged, characterized by a gradual increase followed by a sudden decline under the SSP2-4.5 scenario. Notably, forests showed an increase of 3%, followed by a subsequent decrease of 4.4% between 2035‒2080 and 2080‒2095, respectively, whereas water bodies initially remained stable before undergoing a significant reduction of 16.1% between 2080 and 2095. Except for an increase of 13.4% in farmland from 2080 to 2095, minimal changes were observed in other land types. Under SSP5-8.5, forests accounted for a substantial proportion from 2035 to 2095, which was markedly lower than that under SSP1-2.6 and SSP2-4.5.
Figure 4 LULC simulation in the Middle Yangtze River Basin under different scenarios

Full size|PPT slide

4.2 RSEI analysis of wetlands in the MYRB

To explore the distribution of wetlands, we aggregated water bodies and adjacent grasslands at the county level. The wetlands are primarily concentrated in the central and eastern regions along the main stream of the Yangtze River. Furthermore, counties in the northern and southwestern regions exhibited higher proportions of wetlands, whereas those in the northwestern and eastern areas consistently exhibited lower levels. Simultaneously, we investigated and analyzed the RSEI of each county in the MYRB from 2001 to 2020 (Figure 5). In this study, we revealed that a relatively high RSEI was mainly concentrated in the southern and southwestern parts of the MYRB, which differed from the distribution of wetland proportions and indices. In 2001, the northwestern region exhibited lower RSEI values, with higher values concentrated in the southern region. By 2005, the northwestern and southwestern regions experienced significant increases in RSEI, with an average growth of 43.8%, whereas the southern and central regions experienced an average decrease of 12.1%. In 2017, the RSEI in the northwestern counties declined by an average of 27.9% compared with that in 2013. However, these counties showed improvements in 2020, with an average increase of 20.9%. Overall, the RSEI variations in the MYRB from 2001 to 2020 were dynamic, with improvements in the northwestern and southwestern counties, whereas the central and southern regions experienced decreases, which could be attributed to the impact of human activities or policies.
Figure 5 RSEI in the Middle Yangtze River Basin during the period 2001-2020

Full size|PPT slide

By leveraging the robust predictive performance of RF, which achieved training and testing accuracies of 98.57 and 92%, respectively, we simulated future scenarios for the RSEI in the MYRB. The simulations were based on prominent landscape indices, including the average shape index, patch density, landscape shape index, and wetland proportion patch count (Figure 6). The results indicated that under the SSP1-2.6 scenario, from 2035 to 2050, the mean RSEI showed an increasing trend of 8.3% for counties located in the northwestern and southern regions. Conversely, by 2095, many northwestern counties would experience a decrease in the RSEI, dropping below the levels recorded in 2035. Under SSP2-4.5, from 2035 to 2080, we observed a notable 13.3% reduction in RSEI levels across several northwestern counties, whereas specific southern counties exhibited only marginal increases. By 2095, several northwestern counties experienced a significant increase in the RSEI, whereas occasional decreases were observed in certain central and southwestern counties. Under SSP5-8.5, from 2035 to 2095, notable transformations would occur in the northwestern counties. However, by 2095, the RSEI showed a remarkable decrease of 2.2% in numerous northwestern counties. In summary, counties with suboptimal ecological environments were mainly concentrated in the central region adjacent to the main stream of the Yangtze River, which was a phenomenon inherently linked to the rapid economic development of the MYRB.
Figure 6 RSEI simulations in the Middle Yangtze River Basin under different scenarios

Full size|PPT slide

4.3 Predicting changes in wetland habitats in the MYRB

Figure 7 illustrates the habitat quality and degradation levels in the MYRB in 2020. We identified regions characterized by lower habitat quality as primary areas with dense construction and frequent human activities. In contrast, the northwestern and southwestern regions, characterized by lower human activity and high forest coverage, exhibited remarkable habitat quality. Furthermore, the central, northern, and southern regions exhibited moderate habitat quality, whereas high habitat quality was observed near the main stream of the Yangtze River. Regions that experienced heightened levels of habitat degradation were predominantly concentrated in the eastern part of the central region. These areas, characterized by rapid economic development and intense human activities, had undergone habitat quality degradation, concurrently affecting ecological quality.
Figure 7 Habitat quality levels and degradation levels in the Middle Yangtze River Basin in 2020

Full size|PPT slide

Figure 8 shows the habitat quality under future scenarios in the MYRB. This indicates that the distribution of habitat quality throughout the basin closely corresponds to the overall pattern of the ecological quality index. Under the various scenarios, areas with lower habitat quality were primarily located in the eastern part of the central region, reflecting the distribution pattern of ecological quality, whereas the northwestern and southwestern regions showed higher habitat quality. Under the SSP1-2.6 scenario, the southern region exhibited favorable ecological quality, although the habitat quality levels remained moderate. Ecological quality in the northwestern region experienced frequent fluctuations, whereas the habitat quality levels exhibited less pronounced disparities. Marginal areas in the northwestern region displayed lower ecological quality but better habitat quality, primarily in regions with dense construction, indicating that the influence of different land types on habitat quality surpassed their impact on ecological quality. Under the SSP2-4.5 scenario, changes in habitat quality were not prominent, which was consistent with the variations in ecological quality. However, habitat quality displayed insignificant spatiotemporal characteristics, whereas ecological quality showed noticeable changes in the northwestern region under the SSP5-8.5 scenario.
Figure 8 Habitat quality in the Middle Yangtze River Basin under different future scenarios

Full size|PPT slide

Figure 9 illustrates the degradation of habitat quality in the MYRB under future scenarios. We found that the spatial distribution of habitat degradation was aligned with the spatial distribution of habitat quality; areas with high degradation exhibited lower habitat quality, whereas those with low degradation displayed relatively better habitat quality. In general, variations in habitat quality degradation under the different scenarios were not pronounced. However, they exhibited better clustering, with regions characterized by high habitat degradation concentrated in economically developed areas in the eastern parts; this corresponded to the Wuhan metropolitan area, characterized by rapid economic development, significant urbanization, and population concentration. This pattern closely mirrored the distribution of ecological quality, emphasizing the substantial impact of human activities and urban development on habitat quality and degradation.
Figure 9 Habitat degradation levels in the Middle Yangtze River Basin under different future scenarios

Full size|PPT slide

Figure 10 shows the spatial distribution of habitat quality hotspots in the MYRB from 2035 to 2095 under future scenarios. We found analogous patterns in the distribution of hotspots for both the SSP1-2.6 and SSP2-4.5 scenarios, with hotspots predominantly concentrated in the eastern half. Notably, hot and cold spots intersected, with cold spots scattered around hotspots in the southern and northern regions. Under the SSP1-2.6 scenario, spatial distribution changes were concentrated in the central region, characterized by prevalent hotspots, whereas few hotspots emerged in the northwest region. The continuous increase in hotspot areas from 2035 to 2095 suggests that improvements in habitat quality will become more concentrated with ongoing ecological conservation. The distribution was similar but the concentration of hotspots in the central region showed an increasing trend from 2035 to 2065 and subsequently decreased under the SSP2-4.5 scenario. Under the SSP5-8.5 scenario, hotspots are mainly concentrated in the eastern region, with a cross-distribution of hot and cold spots in the western region. Owing to the overall lower habitat quality under the SSP5-8.5 scenario, changes were primarily concentrated in areas with prevalent hotspots, particularly in the central and northern regions. The northern region exhibited an initial increase, followed by a decline in habitat quality, whereas hotspot areas in the central region experienced a continuous decrease in habitat quality.
Figure 10 Spatial distribution of habitat quality hotspots in the Middle Yangtze River Basin from 2035 to 2095 (ESC: extremely significant cold point; SC: significant cold point; CP: cold point; NS: no significant; HP: hot points; SH: significant hot; ESH: extremely significant point)

Full size|PPT slide

Figure 11 shows the spatial distribution of habitat degradation hotspots in the MYRB from 2035 to 2095 under the future scenarios. We found similar spatial distributions under the SSP1-2.6 and SSP2-4.5 scenarios, characterized by a large area of significant cold spots. This indicated that most regions were clustered in areas with minimal degradation, interspersed with scattered cold and significantly cold spots. In contrast, although cold spots covered large areas, they were concentrated in urban areas under the SSP5-8.5 scenario. This suggested that the degradation was high and concentrated in specific areas, aligned with the ecological conditions and habitat quality distribution.
Figure 11 Spatial distribution of habitat degradation level hotspots in the Middle Yangtze River Basin from 2035 to 2095 (ESC: extremely significant cold point; SC: significant cold point; CP: cold point; NS: no significant; HP: hot points; SH: significant hot; ESH: extremely significant point)

Full size|PPT slide

Figure 12 shows the spatial distribution of habitat quality within the MYRB wetlands. We identified significant variations in wetland habitat quality across different regions under different scenarios, with the northern, southern, and eastern-central regions exhibiting lower habitat quality. In particular, regions with lower wetland habitat quality were concentrated in the north and scattered across the central and southern areas under the SSP1-2.6 scenario. From 2035 to 2050, a noticeable decline was observed in wetland habitat quality, with an increased distribution of areas with poor habitat quality, primarily near the western and tributary areas of the MYRB. Frequent changes in habitat quality were observed, mainly in the northern wetland aggregation areas and near the southern tributaries during 2050 and 2080. By 2095, there will be a significant improvement in habitat quality in the northern and southern regions, with a slight decrease in the central region. Under the SSP2-4.5 and SSP1-2.6 scenarios, changes in wetland habitat quality were mainly concentrated in the northern and southern regions. Between 2035 and 2080, a persistent decrease in wetland habitat quality is observed in both the northern and southern regions, culminating in a notable improvement in poor wetland habitat quality by 2095. This reduction may be attributed to the conversion of poorly performing areas into other land types, resulting in a significant decrease in the wetland area. In comparison, the SSP5-8.5 scenario exhibited fewer areas with low wetland habitat quality, mainly concentrated in the human-activity-intensive eastern-central region. Notably, areas with poor wetland habitat quality in the SSP1-2.6 scenario disappeared under the SSP5-8.5 scenario, where wetland areas had significantly decreased, and regions with low habitat quality were prone to transformation into other land types.
Figure 12 Spatial distribution of wetland habitat quality in the Middle Yangtze River Basin under different future scenarios

Full size|PPT slide

Figure 13 shows the projected levels of wetland habitat degradation in the MYRB under future scenarios. We found that the distribution of wetland degradation was similar, with high degradation levels observed in the eastern-central region, concentrated urban- development areas, and scattered regions near the southern tributaries. Notably, the degradation level coverage under the SSP5-8.5 scenario was lower than that under the SSP1-2.6 scenario, indicating more significant wetland losses and severe habitat quality deterioration. The SSP2-4.5 scenario exhibited moderate degradation levels with lower coverage than the conservation-oriented scenario. However, a slight increase was observed in the levels of habitat quality degradation in the northern wetland aggregation areas. These consistent distribution patterns emphasized the vulnerability of wetlands, particularly in areas with intense human activity and urbanization. These findings underscore the importance of effective conservation and management strategies to mitigate future wetland habitat degradation.
Figure 13 Spatial distribution of wetland habitat degradation levels in the Middle Yangtze River Basin under different future scenarios

Full size|PPT slide

5 Discussion

5.1 Land-use classification and prediction in the MYRB

In the MYRB, the precision of land-use classification was greater when utilizing RF and GBT than when utilizing CARTs and SVMs. This finding is consistent with the results reported by Orieschnig et al. (2021), and emphasizes the effectiveness of RF in tackling the classification challenges posed by mixed pixels, as noted by Floreano and de Moraes (2021). The findings revealed frequent mutual transitions between land types, particularly between for ests, farmlands, and construction land, from 2001 to 2020. From 2000 to 2018, the upper Yangtze River region experienced an increase in forest and aquatic areas, whereas farmlands decreased, which is consistent with the findings of Li et al. (2021). Except for a significant increase of 13.4% in farmland between 2080 and 2095, other land types exhibited minimal changes. Despite the anticipated resource pressures, a global trend may lead to an inevitable increase in agricultural areas, accompanied by a decrease in forested regions (Hurtt et al., 2011; Chen et al., 2020b). Since 2000, China’s “Returning Farmland to Forest” policies have significantly mitigated the historical decline in forest cover (Liu et al., 2003). This may be related to factors, such as resettlement associated with the construction of reservoirs in the Han River Basin and the policy of returning farmland to lakes in the southern Dongting Lake Basin (Chen et al., 2020c). Furthermore, the effectiveness of the policy was reflected in the expanded forested area within the MYRB.
Increased economic development has led to the concentration of human activities in regions with rapid economic growth (Chen et al., 2020c). In this study, under the SSP5-8.5 scenario, forests retained substantial coverage from 2035 to 2095, although they experienced a noticeable reduction compared with the forest coverage under the SSP1-2.6 and SSP2-4.5 scenarios. Overall, prioritizing economic development has resulted in notable ecological degradation, with intensified LULC dynamics characterized by a reduction in forest and grassland areas and an increase in built-up areas. Thus, the acceleration of economic development has led to ecological degradation and conversion of forested areas into urban environments (Liao et al., 2020). The ecological prioritization strategy in the MYRB involves the conversion of farmland and grassland into urban and forested areas, leading to a significant expansion of forested regions. Despite projections suggesting increasing resource pressures in the future, global trends may inevitably lead to the expansion of agricultural areas, accompanied by a decline in forested regions (Hurtt et al., 2011).
In this study, by employing weights derived from RF analysis, we identified diverse determinants influencing various land-cover types across distinct temporal spans. Notably, terrain features (i.e., forests, grasslands, wetlands, farmlands, and urban areas) in the MYRB were significantly influenced by regional elevational variations, whereas water bodies were primarily responsive to temperature and precipitation, which is consistent with the findings of Shao et al. (2020). Moreover, the MYRB, characterized by a dense concentration of urban agglomerations and serving as a pivotal locus in the YRB as China’s second largest urban cluster, is particularly susceptible to anthropogenic interventions that dictate alterations in LULC (Dale, 1997). Moreover, the distinctive monsoon climate highlights the significant impact of temperature and precipitation on land-use determinants and variations within the MYRB (Wang et al., 2021; Cai et al., 2023).

5.2 Wetland ecological quality analysis in the MYRB

The landscape indices of the wetlands in the MYRB exhibited significant dynamics under future scenarios. Under the SSP1-2.6 scenario, wetland areas are projected to increase by 10.2% from 2035 to 2095; this is consistent with the observations of Zhang et al. (2022b), who noted denser and less fragmented wetlands in the MYRB, attributed to enhanced awareness and conservation efforts (Yin et al., 2007; Zhai et al., 2021). Moreover, transitions from other LULC types have significantly contributed to the expansion of wetland areas (Nelson et al., 2009). Conversely, under the SSP2-4.5 scenario, wetland fragmentation increases with greater spatial discontinuity, primarily owing to the transformation of wetlands into other land types, particularly arable land (Zheng et al., 2016). Under the SSP5-8.5 scenario, wetland integrity is compromised, resulting in a decline in ecological quality driven by rapid economic development and urban expansion (Mind’je et al., 2021; Rehmani et al., 2021).
In summary, rapid urbanization and agricultural expansion have significantly altered habitats by converting wetlands into arable land or urban areas. This results in habitat fragmentation that adversely affects the ecological integrity of these regions and disrupts ecosystems, while also reducing biodiversity (Chen et al., 2020a). Changes in precipitation patterns and temperature variations affect wetland hydrology, potentially leading to reduced water availability, which exacerbates stress on wetland ecosystems and contributes to habitat degradation (Liu et al., 2022). Moreover, irrational human activities, such as pollution from agricultural runoff and industrial discharge, have deteriorated water quality (Mind’je et al., 2021). These factors disrupt the delicate balance within wetland ecosystems and threaten the biodiversity in the MYRB. Conservation policies have demonstrated positive effects on wetland restoration; however, their effectiveness varies by region and requires ongoing evaluation to ensure sustainable outcomes. Addressing these challenges through targeted policies and community engagement is essential to preserve critical ecosystems in the MYRB for future generations.

5.3 Prediction of habitat quality in the MYRB under future scenarios

A noticeable decline was observed in habitat quality in the MYRB, particularly in the northern, southern, and eastern-central regions, with the most pronounced decrease in the eastern-central region, dominated by construction land, and near tributaries in the southern part. Consistent with the observations of Zhai et al. (2021), ecological risks in wetlands were reduced in the core urban areas within the Yangtze River Economic Belt (YREB), whereas suburban regions faced increased risks owing to construction activities, which were identified as significant threats to wetland habitat quality. Moderated human activities resulting from ecological protection measures have partially enhanced the wetland habitat quality in these areas (Zhai et al., 2021). Moreover, the SSP5-8.5 scenario predicts a significant degradation of wetland habitat quality in the MYRB, consistent with observations from the Sokoto-Rima Basin, where crop expansion and ongoing business development have contributed to similar trends (Raji et al., 2022). A decreasing trend in wetland habitat quality has been observed in both the inland wetlands and wildlife protection areas within the YRB (Xu et al., 2020). The accelerating degradation of wetland habitat quality is further exacerbated by irrational human activities, particularly rapid urbanization, which continues to decrease wetland coverage (Zhang et al., 2020). Under global change, the increasing severity of wetland degradation and decrease in habitat quality necessitate a heightened focus on land-use planning and restoration efforts.
Studies have indicated a discernible decline in habitat quality, particularly in urbanized areas where human activities are concentrated, posing an alarming threat to both biodiversity and ecosystem services provided by wetlands. Different SSP scenarios indicate that without adequate intervention, wetland habitats may experience severe degradation, particularly under high-development scenarios, such as SSP5-8.5. Despite existing challenges, opportunities for ecological restoration persist as enhanced awareness and protective measures can improve wetland quality, as evidenced by studies showing denser and less fragmented wetlands resulting from conservation efforts.

6 Conclusions

In this study, we integrated remote sensing imagery to analyze the LULC in the MYRB between 2000 and 2020. Using the PLUS model, LULC and its drivers were predicted under future scenarios to investigate the historical and prospective ecological and habitat quality in the MYRB. The major conclusions of this study are as follows:
(1) From 2001 to 2020, the predominant LULC in the MYRB was forest land, followed by agricultural land, grasslands, wetlands, and water bodies. Elevation, population, GDP, precipitation, and slope play significant roles in the LULC distribution, with temperature particularly affecting variations in water bodies. The anticipated trends indicate an expansion of forests and water bodies alongside a reduction in farmland. Under the SSP2-4.5 scenario, we observed a notable increase in farmland, decrease in water bodies, and conversion of forests to farmland. In contrast, the SSP5-8.5 scenario predicted a significant expansion of built-up areas, coupled with a substantial reduction in grasslands and forests. The primary influencing factors were GDP, population, temperature, and precipitation.
(2) From 2001 to 2020, counties with superior ecological and environmental quality were primarily located in the southwest. Variations in ecological quality were predominantly concentrated at the northern boundary under the future scenarios. Notably, the SSP2-4.5 scenario indicated substantial changes, marked by a 13.3% reduction in the wetland area in the northwest from 2035 to 2080. Despite frequent changes in the SSP5-8.5 scenario, relatively minor changes were observed in wetland area.
(3) Areas characterized by lower habitat quality were primarily concentrated in urbanized zones with frequent human activity, indicating a discernible degradation in habitat quality. Overall, we identified prospective trends in land use and changes in habitat quality in the MYRB under future scenarios. This highlights the crucial effects of different policies on the sustainable development of wetland ecosystems.

References

List( Publishing order | Descend order by publishing year | Descend order by cited within ) Chart analysis
[1]
Bai L, Xiu C, Feng X et al., 2019. Influence of urbanization on regional habitat quality: A case study of Changchun city. Habitat International, 93: 102042.
Cited in this article [1]
[2]
Cai W, Yang P, Xia J et al., 2023. Analysis of climate change in the middle reaches of the Yangtze River Basin using principal component analysis. Theoretical and Applied Climatology, 151(1): 449-465.
Cited in this article [2]
[3]
Chen G, Li X, Liu X et al., 2020a. Global projections of future urban land expansion under shared socioeconomic pathways. Nature Communications, 11(1): 1-12.
Cited in this article [2]
[4]
Chen Q, Chen H, Zhang J et al., 2020b. Impacts of climate change and LULC change on runoff in the Jinsha River Basin. Journal of Geographical Sciences, 30(1): 85-102.
Cited in this article [1]
[5]
Chen W, Zhao H, Li J et al., 2020c. Land use transitions and the associated impacts on ecosystem services in the middle reaches of the YREB in China based on the geo-informatic Tupu method. Science of the Total Environment, 701:134690.
Cited in this article [2]
[6]
Dale V H, 1997. The relationship between land-use change and climate change. Ecological Applications, 7: 753-769.
Cited in this article [1]
[7]
Deng H, Chen Y, Li Y, 2019. Glacier and snow variations and their impacts on regional water resources in mountains. Journal of Geographical Sciences, 29(1): 84-100.
https://doi.org/10.1007/s11442-019-1585-2
Cited in this article [1] Abstract

Glaciers and snow are major constituents of solid water bodies in mountains; they can regulate the stability of local water sources. However, they are strongly affected by climate change. This study focused on the Tianshan Mountains, using glacier and snow datasets to analyse variations in glaciers, snow, water storage, and runoff. Three typical river basins (Aksu, Kaidou, and Urumqi Rivers) were selected to interpret the impacts of glacier and snow changes on regional water resources in the Tianshan Mountains. The results exhibited a nonlinear functional relationship between glacial retreat rate and area, demonstrating that small glacial retreat is more sensitive under climate change. Further, the glacial retreat rate at the low-middle elevation zone was seen to be faster than that at the high elevation zone. The regional average terrestrial water storage (TWS) decrease rate in the Tianshan Mountains was -0.7±1.53 cm/a during 2003-2015. The highest TWS deficit region was located in the central part of the Tianshan Mountains, which was closely related to sharp glacial retreats. The increases in glacier and snow meltwater led to an increase in runoff in the three typical river basins, especially that of the Aksu River (0.4×108 m3/a). The decreasing and thinning of areas, and increasing equilibrium line altitude (ELV) of glaciers have been the major causes for the decrease in runoff in the three river basins since the mid-1990s. Therefore, the results reveal the mechanisms causing the impacts of glaciers and snow reduction in mountains on regional water resources under climate change, and provide a reference for water resources management in the mountainous river basins.
[8]
Floreano I X, de Moraes L A F, 2021. Land use/land cover (LULC) analysis (2009-2019) with Google Earth Engine and 2030 prediction using Markov-CA in the Rondônia State, Brazil. Environmental Monitoring and Assessment, 193: 239.
Cited in this article [1]
[9]
Fu Q, Hou Y, Wang B et al., 2018. Scenario analysis of ecosystem service changes and interactions in a mountain-oasis-desert system: A case study in Altay Prefecture, China. Scientific Reports, 8(1): 1-13.
Cited in this article [1]
[10]
Guha S, Govil H, Gill N et al., 2021. A long-term seasonal analysis on the relationship between LST and NDBI using Landsat data. Quaternary International, 575: 249-258.
Cited in this article [3]
[11]
Gupta K, Kumar P, Pathan S K et al., 2012. Urban neighborhood green index: A measure of green spaces in urban areas. Landscape and Urban Planning, 105(3): 325-335.
Cited in this article [1]
[12]
Hong C, Burney J A, Pongratz J et al., 2021. Global and regional drivers of land-use emissions in 1961-2017. Nature, 589(7843): 554-561.
Cited in this article [1]
[13]
Hurtt G C, Chini L P, Frolking S et al., 2011. Harmonization of land-use scenarios for the period 1500-2100: 600 years of global gridded annual land-use transitions, wood harvest, and resulting secondary lands. Climatic Change, 109(1), 117-161.
Cited in this article [2]
[14]
Lawler J J, Lewis D J, Nelson E et al., 2014. Projected land-use change impacts on ecosystem services in the United States. Proceedings of the National Academy of Sciences, 111(20): 7492-7497.
[15]
Li C, Wu Y, Gao B et al., 2021. Multi-scenario simulation of ecosystem service value for optimization of land use in the Sichuan-Yunnan ecological barrier, China. Ecological Indicators, 132:108328.
Cited in this article [1]
[16]
Li S, Dong B, Gao X et al., 2022. Study on spatio-temporal evolution of habitat quality based on land-use change in Chongming Dongtan, China. Environmental Earth Sciences, 81(7): 220.
[17]
Liang X, Guan Q, Clarke K C et al., 2021. Understanding the drivers of sustainable land expansion using a patch-generating land use simulation (PLUS) model: A case study in Wuhan, China. Computers, Environment and Urban Systems, 85: 101569.
Cited in this article [3]
[18]
Liao W, Liu X, Xu X et al., 2020. Projections of land use changes under the plant functional type classification in different SSP-RCP scenarios in China. Science Bulletin, 65(22): 1935-1947.
https://doi.org/10.1016/j.scib.2020.07.014
https://www.ncbi.nlm.nih.gov/pubmed/36738059
Cited in this article [1] Abstract
Land use projections are crucial for climate models to forecast the impacts of land use changes on the Earth's system. However, the spatial resolution of existing global land use projections (e.g., 0.25°×0.25° in the Land-Use Harmonization (LUH2) datasets) is still too coarse to drive regional climate models and assess mitigation effectiveness at regional and local scales. To generate a high-resolution land use product with the newest integrated scenarios of the shared socioeconomic pathways and the representative concentration pathways (SSPs-RCPs) for various regional climate studies in China, here we first conduct land use simulations with a newly developed Future Land Uses Simulation (FLUS) model based on the trajectories of land use demands extracted from the LUH2 datasets. On this basis, a new set of land use projections under the plant functional type (PFT) classification, with a temporal resolution of 5 years and a spatial resolution of 5 km, in eight SSP-RCP scenarios from 2015 to 2100 in China is produced. The results show that differences in land use dynamics under different SSP-RCP scenarios are jointly affected by global assumptions and national policies. Furthermore, with improved spatial resolution, the data produced in this study can sufficiently describe the details of land use distribution and better capture the spatial heterogeneity of different land use types at the regional scale. We highlight that these new land use projections at the PFT level have a strong potential for reducing uncertainty in the simulation of regional climate models with finer spatial resolutions.Copyright © 2020 Science China Press. Published by Elsevier B.V. All rights reserved.
[19]
Liu J, Liu M, Zhuang D et al., 2003. Study on spatial pattern of land-use change in China during 1995-2000. Science in China Series D, 46: 373-384.
Cited in this article [1]
[20]
Liu Y J, Yang P, Zhang S Q et al., 2022. Dynamic identification and health assessment of wetlands in the middle reaches of the Yangtze River Basin under changing environment. Journal of Cleaner Production, 345: 131105.
Cited in this article [2]
[21]
Mccarty D A, Kim H W, Lee H K, 2020. Evaluation of light gradient boosted machine learning technique in large scale land use and land cover classification. Environments, 7(10): 84.
Cited in this article [1]
[22]
Mind’je R, Mindje M, Kayumba P M, 2021. The influence of anthropogenic activities on wetland integrity dynamics: A case study of the Rwampara wetland in Rwanda. Environmental Sustainability, 4(4): 833-849.
Cited in this article [2]
[23]
Nelson E, Mendoza G, Regetz J et al., 2009. Modeling multiple ecosystem services, biodiversity conservation, commodity production, and tradeoffs at landscape scales. Frontiers in Ecology and the Environment, 7(1): 4-11.
Cited in this article [1]
[24]
Ning J, Liu J, Kuang W et al., 2018. Spatiotemporal patterns and characteristics of land-use change in China during 2010-2015. Journal of Geographical Sciences, 28(5): 547-562.
https://doi.org/10.1007/s11442-018-1490-0
Cited in this article [1] Abstract

Land use/cover change is an important theme on the impacts of human activities on the earth systems and global environmental change. National land-use changes of China during 2010-2015 were acquired by the digital interpretation method using the high-resolution remotely sensed images, e.g. the Landsat 8 OLI, GF-2 remote sensing images. The spatiotemporal characteristics of land-use changes across China during 2010-2015 were revealed by the indexes of dynamic degree model, annual land-use changes ratio etc. The results indicated that the built-up land increased by 24.6×103 km2 while the cropland decreased by 4.9×103 km2, and the total area of woodland and grassland decreased by 16.4×103 km2. The spatial pattern of land-use changes in China during 2010-2015 was concordant with that of the period 2000-2010. Specially, new characteristics of land-use changes emerged in different regions of China in 2010-2015. The built-up land in eastern China expanded continually, and the total area of cropland decreased, both at decreasing rates. The rates of built-up land expansion and cropland shrinkage were accelerated in central China. The rates of built-up land expansion and cropland growth increased in western China, while the decreasing rate of woodland and grassland accelerated. In northeastern China, built-up land expansion slowed continually, and cropland area increased slightly accompanied by the conversions between paddy land and dry land. Besides, woodland and grassland area decreased in northeastern China. The characteristics of land-use changes in eastern China were essentially consistent with the spatial govern and control requirements of the optimal development zones and key development zones according to the Major Function-oriented Zones Planning implemented during the 12th Five-Year Plan (2011-2015). It was a serious challenge for the central government of China to effectively protect the reasonable layout of land use types dominated with the key ecological function zones and agricultural production zones in central and western China. Furthermore, the local governments should take effective measures to strengthen the management of territorial development in future.
[25]
Qi G Z, She D X, Xia J et al., 2024. Soil moisture plays an increasingly important role in constraining vegetation productivity in China over the past two decades. Agricultural and Forest Meteorology, 356: 110193.
Cited in this article [1]
[26]
Raji S A, Odunuga S, Fasona M, 2022. Spatially explicit scenario analysis of habitat quality in a tropical semi-arid zone: Case study of the Sokoto-Rima Basin. Journal of Geovisualization and Spatial Analysis, 6(1): 11.
Cited in this article [1]
[27]
Rehmani M I A, Ding C, Li G et al., 2021. Vulnerability of rice production to temperature extremes during rice reproductive stage in Yangtze River Valley, China. Journal of King Saud University-Science, 33(8): 101599.
Cited in this article [1]
[28]
Shao Z F, Ding L, Li D et al., 2020. Exploring the relationship between urbanization and ecological environment using remote sensing images and statistical data: A case study in the Yangtze River Delta, China. Sustainability, 12(14): 5620.
Cited in this article [1]
[29]
Shi X R, Yang P, Xia J et al., 2024. Effects of precipitation on vegetation and surface water in the Yellow River Basin during 2000-2021. Journal of Geographical Sciences, 34(4): 633-653.
https://doi.org/10.1007/s11442-024-2221-3
Cited in this article [1] Abstract

The Yellow River Basin (YRB) is a vital ecological zone in China owing to its sensitive and fragile environment. Under the long-term influence of climate changes and artificial factors, the relationship between precipitation, vegetation, and surface water in the YRB has changed drastically, ultimately affecting the water resources and environmental management. Therefore, we applied multivariate statistical analysis to investigate the precipitation, normalized difference vegetation index (NDVI), and surface water changes in the YRB from 2000 to 2021. Furthermore, we attempted to clarify the ecological effects of precipitation by explaining the relationship between precipitation and vegetation in terms of the time-lag relationship using the Integrated Multi-satellite Retrievals for Global Precipitation Measurement algorithm, Moderate Resolution Imaging Spectroradiometer, and hydrological databases. Precipitation, vegetation, and area of surface water in the YRB showed increasing trends from 2000-2021 (e.g., 7.215 mm/yr, 0.004 NDVI/yr, and 0.932 km2/yr, respectively). The water level in the upper reaches of the YRB showed a downward trend, whereas that in the middle and lower reaches exhibited an upward trend. Changes in precipitation had a positive effect on vegetation and surface water in the YRB, with correlation coefficients of 0.63 and 0.51, respectively. The responses of NDVI and surface water elevation to precipitation were heterogeneous and delayed, with the majority showing a lag time of approximately ≤ 16 days. Moreover, the lag times of Longyangxia Lake and Ngoring-Co Lake were 0 and 8 days, respectively. We showed that precipitation variability can effectively explain vegetation improvement and increases in surface water elevation, while providing a proven scenario for predicting the surface water and vegetation productivity under the influence of climate change.
[30]
Sivakumar R, Ghosh S, 2016. Wetland spatial dynamics and mitigation study: An integrated remote sensing and GIS approach. Natural Hazards, 80(2): 975-995.
Cited in this article [1]
[31]
Song X Q, Wen M, Shen Y et al., 2020. Urban vacant land in growing urbanization: An international review. Journal of Geographical Sciences, 30(4): 669-687.
https://doi.org/10.1007/s11442-020-1749-0
Cited in this article [1] Abstract

Urban growth and shrinkage constitute the overall pattern of growing urbanization across the globe. Studies on urban vacant land (UVL) are few, and have proved to be mainly rudimentary and subjective. This paper first presents the definition of UVL based on bibliometric analysis. Typology, morphology, proximate causes, and the multiple functions of UVL are then analyzed at parcel, transect, city, and national levels based on an international review. Results show that UVL can be categorized by land cover, land usage, and land ownership. Worldwide, UVL has been widespread and extensive. For example, the occurrence probabilities of UVL in the cases of Guangzhou and New York are 8.46%-8.88% and 3.17%-5.08%, respectively. The average vacancy rate of residential land amounts to 11.48% in 65 U.S. cities. Generally, UVL shows fragmentation and irregular shape, and significant spatial differences exist at parcel, transect, city, and national levels. Proximate causes, such as excessive land division, irregularly shaped land parcels, decreases in resident population, deindustrialization, land speculation, insufficient investment, and environmental concerns, can all result in UVL. Currently, UVL has become a gray area of social, economic, and ecological space. However, it can also be considered a potential resource for enhancing urban sustainability. Policy implications to promote urban sustainability using monitoring, control, and differential revitalization of UVL are presented.
[32]
Verburg P H, Neumann K, Nol L, 2011. Challenges in using land use and land cover data for global change studies. Global Change Biology, 17(2): 974-989.
Cited in this article [1]
[33]
Wang D, Gong J, Chen L et al., 2012. Spatio-temporal pattern analysis of land use/cover change trajectories in Xihe watershed. International Journal of Applied Earth Observation and Geoinformation, 14(1): 12-21.
Cited in this article [1]
[34]
Wang R, Li F, Hu D et al., 2011. Understanding eco-complexity: social-economic-natural complex ecosystem approach. Ecological Complexity, 8(1): 15-29.
Cited in this article [2]
[35]
Wang W Y, Yang P, Xia J et al., 2022. Coupling analysis of surface runoff variation with atmospheric teleconnection indices in the middle reaches of the Yangtze River. Theoretical and Applied Climatology, 148(3): 1513-1527.
Cited in this article [2]
[36]
Wei Y H, She D X, Xia J et al., 2024. Climate change dominates the increasing exposure of global population to compound heatwave and humidity extremes in the future. Climate Dynamics, 1-15.
Cited in this article [1]
[37]
Xu X, 2017a. China GDP Spatial Distribution Kilometer Grid Dataset. Resource and Environment Science Data Publishing and Registration System (http://www.resdc.cn/DOI). doi: 10.12078/2017121102.
Cited in this article [2]
[38]
Xu X, 2017b. China Population Spatial Distribution Kilometer Grid Dataset. Resource and Environment Science Data Publishing and Registration System (http://www.resdc.cn/DOI). doi: 10.12078/2017121101.
Cited in this article [2]
[39]
Xu X, Jiang B, Chen M et al., 2020. Strengthening the effectiveness of nature reserves in representing ecosystem services: The Yangtze River Economic Belt in China. Land Use Policy, 96: 104717.
Cited in this article [2]
[40]
Yang J L, Yang P, Zhang S Q et al., 2023. Evaluation of water resource carrying capacity in the middle reaches of the Yangtze River Basin using the variable fuzzy-based method. Environmental Science and Pollution Research, 30(11): 30572-30587.
Cited in this article [1]
[41]
Yin H, Liu G, Pi J et al., 2007. On the river-lake relationship of the middle Yangtze reaches. Geomorphology, 85: 197-207.
Cited in this article [1]
[42]
Zhai T, Wang J, Fang Y et al., 2021. Identification and prediction of wetland ecological risk in key cities of the Yangtze River Economic Belt: From the perspective of land development. Sustainability, 13: 411.
Cited in this article [3]
[43]
Zhang Q, She D X, Zhang L P et al., 2022a. High sensitivity of compound drought and heatwave events to global warming in the future. Earth’s Future, 10(11): e2022EF002833.
Cited in this article [2]
[44]
Zhang S Q, Yang P, Xia J et al., 2021a. Research and analysis of ecological environment quality in the middle reaches of the Yangtze River Basin between 2000 and 2019. Remote Sensing, 13(21): 4475.
Cited in this article [1]
[45]
Zhang S Q, Yang P, Xia J et al., 2023a. Remote sensing inversion and prediction of land use land cover in the middle reaches of the Yangtze River basin, China. Environmental Science and Pollution Research, 30(16): 46306-46320.
Cited in this article [1]
[46]
Zhang S Q, Yang P, Xia J et al., 2022b. Land use/land cover prediction and analysis of the middle reaches of the Yangtze River under different scenarios. Science of the Total Environment, 833: 155238.
Cited in this article [5]
[47]
Zhang S Q, Zhai X Y, Yang P et al., 2023b. Ecological health analysis of wetlands in the middle reaches of Yangtze River under changing environment. International Journal of Digital Earth, 16(1): 3125-3144.
Cited in this article [1]
[48]
Zhang X, Liu L, Chen X et al., 2021b. GLC_FCS30: global land-cover product with fine classification system at 30m using time-series Landsat imagery. Earth System Science Data Discussions, 13(6): 2753-2776.
Cited in this article [1]
[49]
Zhang X, Song W, Lang Y et al., 2020. Land use changes in the coastal zone of China’s Hebei province and the corresponding impacts on habitat quality. Land Use Policy, 99: 104957.
Cited in this article [1]
[50]
Zhao G, Tian P, Mu X et al., 2014. Quantifying the impact of climate variability and human activities on streamflow in the middle reaches of the Yellow River basin, China. Journal of Hydrology, 519: 387-398.
Cited in this article [1]
[51]
Zhu Z, Woodcock C E, 2014. Continuous change detection and classification of land cover using all available Landsat data. Remote sensing of Environment, 144: 152-171.
Cited in this article [1]

Funding

National Natural Science Foundation of China(42207078)
CUG Scholar - Scientific Research Funds at China University of Geosciences (Wuhan)(2022166)
China Scholarship Council(202306410026)
Opening Fund of State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water Resources and Hydropower Research(IWHR-SKL-KF202217)
PDF(192397 KB)

142

Accesses

0

Citation

51

Altmetric

Detail
Sections
Recommended

/

〈 〉