Next Article in Journal
Kohler-Polarization Sensor for Glint Removal in Water-Leaving Radiance Measurement
Previous Article in Journal
RaDiT: A Differential Transformer-Based Hybrid Deep Learning Model for Radar Echo Extrapolation
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Adaptive Analysis of Ecosystem Stability in China to Soil Moisture Variations: A Perspective Based on Climate Zoning and Land Use Types

1
State Key Laboratory of Ecological Safety and Sustainable Development in Arid Lands, Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi 830011, China
2
University of Chinese Academy of Sciences, Beijing 100049, China
3
Cele National Station of Observation and Research for Desert-Grassland Ecosystems, Cele 848300, China
4
Polish-Chinese Centre for Environmental Research, Institute of Earth Sciences, University of Silesia in Katowice, 40-007 Katowice, Poland
*
Author to whom correspondence should be addressed.
Remote Sens. 2025, 17(12), 1971; https://doi.org/10.3390/rs17121971
Submission received: 26 April 2025 / Revised: 28 May 2025 / Accepted: 4 June 2025 / Published: 6 June 2025

Abstract

:
In this study, we investigate the impact of soil moisture at varying depths on the stability of Chinese ecosystems, with ecosystem stability assessed using the Enhanced Vegetation Index (EVI) and Gross Primary Productivity (GPP). A multi-perspective analysis is conducted across different climatic zones and land cover types. Sen’s Slope Estimation and the Mann–Kendall trend test, combined with linear regression and correlation analyses, are employed to analyze the long-term trends of EVI and GPP in different climatic zones and land cover types and to assess the effects of soil moisture changes on ecosystem stability. The research reveals the following findings: (1) On a national scale, both EVI and GPP exhibit positive growth trends, with more significant increases in humid areas and relatively slower growth in arid areas. In addition, EVI and GPP of different land cover types exhibit positive inter-annual variation trends, reflecting a gradual enhancement in ecosystem productivity. (2) Cluster analysis shows that EVI has strong spatial correlation, with a distribution pattern of low–low (L-L) clusters in the north and high–high (H-H) clusters in the south. L-H clusters are concentrated in the Huaihai, Southwest Rivers, and Pearl River basins, while H-L clusters are scattered along the eastern coast. The spatial correlation of GPP is mainly concentrated in the south and the northeast, with a distribution pattern of L-L in the northeast, L-H in the Yangtze River basin, and H-H in the south. H-L clusters are dispersed in the downstream area of the Yangtze River. Both EVI and GPP show a tendency for high-value aggregation in space, with high-value areas of EVI located in the south and low-value areas in the central and western regions. High-value areas of GPP are in the south, while low-value areas are in the northeast, particularly in the Yangtze River Delta. (3) The correlation between EVI, GPP, and soil moisture varies significantly across different climatic regions. Arid and semi-humid regions show significant correlations between specific soil moisture depths and EVI and GPP, while such correlations are not significant in humid regions. The EVI and GPP values of croplands and grasslands are significantly and negatively correlated with soil moisture at depths of 150–200 cm (SM4). Conversely, wetland GPP values increase significantly with increasing soil moisture. Other vegetation types do not show significant correlations with soil moisture. The results of this study provide an important basis for understanding the impact of climate change on ecosystem stability and offer scientific guidance for ecological protection and water resource management.

1. Introduction

Soil moisture, a critical ecological factor, is essential for maintaining ecosystem stability [1,2]. It directly influences vegetation growth and productivity and indirectly affects ecosystem functions and stability by modulating soil microbial activity, nutrient cycling, and soil physical properties [3]. Global warming and altered precipitation patterns have significantly impacted the spatial and temporal distribution and availability of soil moisture, thereby increasing ecosystem vulnerability [4]. Thus, understanding how ecosystem stability responds to changes in soil moisture is crucial for predicting future climate change impacts on ecosystems [3].
Ecosystem stability is defined as the ability of an ecosystem to maintain its structure and function relatively stably when facing external disturbances such as climate change and human activities [5]. It reflects the capacity of an ecosystem to maintain its own balance and resilience in dynamic changes and is an important indicator for measuring the health and sustainability of ecosystems [6]. The Enhanced Vegetation Index (EVI) and Gross Primary Productivity (GPP) are commonly used indicators to characterize ecosystem stability. EVI reflects the growth status and cover of vegetation through remote sensing data and can intuitively show the response of vegetation to environmental changes [7]. GPP directly reflects the photosynthetic efficiency and productivity level of vegetation and is the basis for energy flow and material cycling in ecosystems [8]. These two indicators, from the perspectives of vegetation growth and productivity, can effectively reflect the stability and health status of ecosystems and are, therefore, widely used in ecosystem stability research. However, most current studies focus on a single indicator of EVI or GPP, while relatively few studies integrate the two for comprehensive research.
China’s vast territory encompasses a wide array of ecosystem types, and soil moisture conditions exhibit substantial spatial variability across different regions, which has a profound impact on ecosystem stability. Exploring the effects of soil moisture at varying depths on the Enhanced Vegetation Index (EVI) and Gross Primary Productivity (GPP) could potentially uncover the underlying mechanisms linking soil moisture to ecosystem stability. Previous studies have extensively investigated the impact of soil moisture on ecosystem stability in specific regions or particular ecosystem types [9,10,11,12,13]. However, these studies have predominantly concentrated on specific regions or particular ecosystem types, and there is a lack of a comprehensive investigation into the diverse ecosystem types in China under a wide range of soil moisture conditions at various depths. Moreover, the majority of studies have focused on the impact of shallow soil moisture on ecosystem stability, with limited exploration into the influence mechanisms of the vertical distribution of soil moisture on ecosystem stability.
To address these knowledge gaps, this study investigates the responses of key indicators of ecosystem stability in China—specifically, the Enhanced Vegetation Index (EVI) and Gross Primary Productivity (GPP)—to soil moisture at four depths (0–50 cm, 50–100 cm, 100–150 cm, and 150–200 cm). The research objectives include (1) examining the spatiotemporal variation characteristics of EVI and GPP across different climatic zones and land cover types; (2) analyzing the spatial auto-correlation characteristics of EVI and GPP using global and local Moran’s I; (3) investigating the spatiotemporal variation characteristics of soil moisture at different depths across different climatic zones and land cover types; and (4) conducting correlation analyses between EVI, GPP, and soil moisture at different depths across different climatic zones and land cover types. These research objectives will enhance our understanding of the mechanisms by which soil moisture affects ecosystem functions in China.

2. Research Data and Research Methods

2.1. Overview of the Study Area

The study area covers the entire Chinese mainland (3°51′N–53°31′N, 73°40′E–135°2′E), with a total area of approximately 9.6 million square kilometers (excluding the South China Sea islands). Located in Eastern Asia, China features a diverse climate, predominantly temperate with a small tropical region in the south. The country exhibits distinct monsoonal characteristics, characterized by cold, dry winters and warm, humid summers. China’s topography is complex, with plateaus, mountains, and hills comprising a significant portion of its landmass. The terrain slopes from high in the west to low in the east in a stepped pattern. The varied orientations of mountain ranges lead to significant regional differences in temperature and precipitation, resulting in a wide range of climatic types, including tropical monsoon, subtropical monsoon, temperate monsoon, temperate continental, and alpine plateau climates. Variations in heat and humidity also exist within the same climatic categories. Moreover, China’s mountains and hills provide a rich array of habitats for vegetation, encompassing nearly all types of vegetation found in the Northern Hemisphere, from tropical rainforests to cold temperate coniferous forests. The eastern regions are primarily dominated by forests and grasslands, while the western regions are characterized by deserts and grasslands, demonstrating spatial heterogeneity in land cover types (Figure 1).

2.2. Research Data

2.2.1. EVI Data

The Enhanced Vegetation Index (EVI) was initially developed to address the potential saturation of vegetation signals in high-biomass areas, thereby enhancing sensitivity to vegetation conditions. As a measure of the “greenness” of vegetation canopies, EVI has become a key indicator for studying the spatial and temporal dynamics of terrestrial vegetation photosynthetic activity [14]. The EVI data used in this study were sourced from Zenodo (https://doi.org/10.5281/zenodo.7979989, accessed on 6 December 2024), which employs advanced spatial and temporal reconstruction techniques based on the Moderate-Resolution Imaging Spectroradiometer (MODIS) MOD13Q1 product. This dataset covers China from 2000 to 2022, with a spatial resolution of 250 m and a temporal resolution of 16 days. The dataset reconstruction method used, known as spatial–inter-annual reconstruction (SIR), incorporates both spatial and inter-annual information to fill cloud-contaminated pixels, effectively addressing the issue of cloud masking and ensuring high accuracy of the data. The SIR method utilizes MODIS Quality Control (QC) flags to identify and retain pixels with good quality or those that meet specific criteria, such as having a Vegetation Index higher than 80% of the multi-year average on the same date [15]. Using Geographic Information Systems (GISs), the original dataset was integrated and converted into a monthly scale and then aggregated to an annual scale for long-term trend analysis. For this research, EVI data from 2001 to 2020 were selected and resampled to a 1 km × 1 km spatial resolution using the nearest neighbor method to match the scale requirements of the study area. The nearest neighbor method was chosen to preserve the original values of the EVI data, ensuring that the Vegetation Index values remain consistent with the source dataset without introducing any interpolated values that could alter the interpretation of vegetation conditions.

2.2.2. GPP Data

Gross Primary Productivity (GPP) refers to the total amount of primary productivity generated by plants through photosynthesis in an ecosystem. It is a key indicator for measuring energy flow and biomass accumulation [16]. The GPP data used in this study were sourced from the Zenodo platform (https://zenodo.org/records/6518002, accessed on 3 December 2024). This dataset was constructed using the GLOBMAP Leaf Area Index (LAI), CRUJRA meteorological data, and ESA-CCI land cover data, in conjunction with an updated Two-Leaf Light Utilization Efficiency Model (TL-LUE). The model was calibrated and validated against Eddy covariance measurements from 68 FLUXNET sites (480 site years) for parameter optimization, and an independent validation using 25 FLUXNET sites (170 site years) yielded an overall coefficient of determination (R2) of 0.76 and root mean square error (RMSE) of 2.1 g C m−2 day−1 [17]. The resulting global GPP dataset has a spatial resolution of 0.05° and a temporal resolution of 8 days, covering the period from 1992 to 2020. For this research, GPP data from 2001 to 2020 were selected and resampled to a 1 km × 1 km spatial resolution using the nearest neighbor method to match the scale requirements of the study area.

2.2.3. Land Cover Data

The land cover dataset utilized in this study is sourced from Zenodo (https://doi.org/10.5281/zenodo.12779975, accessed on 8 August 2024). This dataset, known as the China Land Cover Dataset (CLCD), was constructed using 335,709 Landsat images provided by Google Earth Engine and covers the period from 1985 to 2020. The dataset’s preparation involved integrating stable sample points from the China Land Cover Dataset (CLUD) and visually interpreted sample points from satellite time-series data, Google Earth, and Google Maps. These sample points were used during the training phase of the dataset’s development. A series of temporal metrics derived from comprehensive Landsat imagery were employed as input variables for the Random Forest classifier to generate land cover classification results. To enhance the spatial and temporal consistency of the CLCD, a post-processing method incorporating spatial–temporal filtering and logical reasoning was applied. This optimization ensured the accuracy and reliability of the dataset.

2.2.4. Soil Moisture Data

The soil moisture data employed in this study were sourced from the ERA5-Land reanalysis dataset provided by the Copernicus Climate Change Service (available at https://cds.climate.copernicus.eu/datasets/reanalysis-era5-land?tab=download, accessed on 11 November 2024). This dataset, based on advanced meteorological and climatological models, provides monthly soil moisture data from 1950 to 2023. It constitutes an essential tool for examining spatial and temporal variations in soil moisture. However, it is important to note that ERA5-Land has documented biases in high-elevation and arid regions. Using the Geographic Information System (GIS) platform, the raw dataset was processed and converted to an annual scale to support long-term trend analysis. For this study, soil moisture data for the period 2001–2020 at four depths (0–50 cm [SM1], 50–100 cm [SM2], 100–150 cm [SM3], and 150–200 cm [SM4]) were selected. The spatial resolution was resampled to 1 km × 1 km using the nearest neighbor method to match the scale requirements of the study area.

2.3. Research Methods

2.3.1. Sen’s and Mann–Kendall Trend Test

Sen’s Slope Estimator is a robust non-parametric statistical method for trend analysis in long time-series data. It is favored for its resilience to outliers and lack of assumptions about data distribution, allowing it to extract trend information directly from data. Due to these features, it is widely applied in climate and vegetation change studies [18]. The calculation formula is as follows:
S l o p e = m e d i a n ( x i x j i j ) , i > j
In the formula, x i and x j are the values of meteorological elements for the i and j years, respectively, and median belongs to the median function. When Slope > 0, it indicates an upward trend. When Slope = 0, it means that the elements remain basically unchanged. If Slope < 0, it indicates a downward trend.
The Mann–Kendall trend test, a non-parametric statistical method, is widely used for analyzing non-linear trends in non-normally distributed variables due to its distribution-free nature [19]. Recognized by the World Meteorological Organization (WMO) as an effective non-parametric trend analysis tool, it is particularly suited for long-term trend analysis of meteorological and hydrological sequences. In this study, the Mann–Kendall test is employed to identify and quantify potential trends in meteorological variables, thereby providing a rigorous statistical framework for understanding the impact of climate change on the stability of arid ecosystems. The calculation formula for the MK trend test statistic is as follows:
S = i = 1 n 1 j = i + 1 n sgn x j x i
sgn x j x i = + 1 0 1 i f i f i f x j x i x j x i x j x i > = < 0 0 0
V a r S = 1 18 n n 1 2 n + 5 q = 1 p t q t q 1 2 t q + 5
Z m k = S V a r S i f   S > 0 0 i f   S = 0 S 1 V a r S i f   S < 0
In the formula, x i and x j represent the values of the time series in the i and j years, sgn is a symbolic function, n is the length of the time-series data, t q corresponds to the q term, and z m k represents the trend of the time-series data.

2.3.2. Spatial Auto-Correlation Analysis Based on Moran’s I

Spatial auto-correlation represents the potential interdependence between spatial units and their adjacent units in the form of certain characteristic values, typically measured using Moran’s I index [20]. In our spatial auto-correlation analysis, we employed a spatial weighting matrix based on adjacency to capture the spatial relationships between neighboring units. Moran’s I is divided into Global Moran’s I and Local Moran’s I. Global Moran’s I indicates whether spatial clustering occurs but does not identify the specific clustering areas. In contrast, Local Moran’s I reflects the spatial concentration of regional units and can reveal where spatial clustering is located. Even if the Global Moran’s I does not show significant clustering, Local Moran’s I can highlight local spatial clustering phenomena. Local indicators of spatial association (LISA) are used to analyze the correlation between adjacent spatial units, measuring the similarity (positive correlation) or difference (negative correlation) between the attribute values of an observation unit and its surrounding units. The calculation formulae are as follows:
G l o b a l M o r a n s I = n i = 1 n j = 1 n W i j x i x ¯ x j x ¯ i = 1 n x i x ¯ j = 1 n x j x ¯
L o c a l M o r a n s I = i = 1 n j = 1 n W i j x i x ¯ x j x ¯ S 2 i = 1 n j = 1 n W i j x i x ¯
In the formula, n represents the number of units i and j ; W i j denotes the spatial weight between x i and x j ; x i and x j , respectively, represent the EVI (GPP) values of unit i and j .
The global Moran’s I index ranges from −1 to 1. A value greater than 0 indicates spatial clustering, with larger values suggesting more pronounced clustering; a value less than 0 indicates a dispersed spatial distribution, with smaller values indicating more pronounced dispersion; a value of 0 suggests that the observed data are randomly distributed in space [21]. The local Moran’s I index results can be categorized into five types: (a) high–high (H-H): spatial units with high EVI (GPP) values are adjacent to other spatial units with high EVI (GPP) values; (b) low–low (L-L): spatial units with low EVI (GPP) values are adjacent to other spatial units with low EVI (GPP) values; (c) low–high (L-H): spatial units with low EVI (GPP) values are adjacent to spatial units with high EVI (GPP) values; (d) high–low (H-L): spatial units with high EVI (GPP) values are adjacent to spatial units with low EVI (GPP) values; (e) non-significant (random distribution): this indicates that the spatial units do not exhibit clear spatial clustering.

2.3.3. Correlation Analysis

Correlation analysis is frequently employed to assess the degree of association between variables [22]. To elucidate the impact of soil moisture on ecosystem stability, a quantitative analysis of the correlation between Enhanced Vegetation Index (EVI), Gross Primary Productivity (GPP), and soil moisture is conducted. The formula is as follows:
r = i = 1 n x i x ¯ y i y ¯ i = 1 n x i x ¯ 2 i = 1 n y i y ¯ 2
In the formula, r is the linear correlation coefficient of x and y variables; x i and y i are the values of x and y in year i , respectively; x ¯ and y ¯ represent the average value of the two variables over n years, which in this study was 20 years. r > 0, indicating that the two variables are positively correlated; r < 0 indicates that the two variables are negatively correlated. In addition, t test was used to determine the significance of the correlation, and the results were divided into four grades: extremely significant (p ≤ 0.01), significant (0.01 < p ≤ 0.05), weakly significant (0.05 < p ≤ 0.1), and not significant (p > 0.1).

3. Results and Analysis

3.1. Spatial and Temporal Distribution Characteristics of EVI and GPP

The spatial distribution and trends of the Enhanced Vegetation Index (EVI) and Gross Primary Productivity (GPP) across China from 2001 to 2020 are depicted in Figure 2. Both indices exhibit higher values in the southeast and lower values in the northwest. The multi-year average EVI ranges from 0 to 0.64 (Figure 2a), with an annual change slope between −0.0033 and 0.0051 per year (Figure 2c). The multi-year average GPP ranges from 0 to 408.85 gC/m2·a (Figure 2b), with an annual change slope ranging from −4.64 to 3.98 gC/m2·a (Figure 2d). Higher growth rates are observed in humid and semi-humid regions, while lower or declining rates are evident in other regions. On a national scale, regions with decreasing EVI account for 15.95%, while those with increasing EVI account for 84.05%. In different climate zones, the proportions of decreasing regions are 34.45% in arid areas, 11.16% in semi-arid areas, 11.82% in semi-humid areas, and 4.2% in humid areas. Regarding land cover types, the proportions of decreasing regions are 4.49% in croplands, 3.98% in forests, 2.17% in shrubs, 13.76% in grasslands, and 37.29% in wetlands. For GPP, on a national scale, decreasing regions account for 34.11%, while increasing regions account for 65.89%. In different climate zones, the proportions of decreasing regions are 77.23% in arid areas, 28.38% in semi-arid areas, 8.12% in semi-humid areas, and 12.3% in humid areas. In terms of land cover types, the proportions of decreasing regions are 13.24% in croplands, 6.56% in forests, 3.14% in shrubs, 27.76% in grasslands, and 21.62% in wetlands. Overall, both EVI and GPP are predominantly increasing at the national level. However, significant decreases are observed in specific regions and land cover types, such as arid areas and wetlands (Figure 3).
Figure 4 illustrates the inter-annual variations in EVI and GPP across China and various climatic zones from 2001 to 2020, revealing an overall increasing trend for both parameters. The annual rate of change for EVI is 0.001277 per year (Figure 4a), while for GPP, it is 0.933684 gC/m2/year (Figure 4f). Notably, significant differences in the inter-annual rates of change for EVI and GPP are observed among different climatic zones. In arid regions, the inter-annual rates of change are 0.000332 per year for EVI and 0.22109 gC/m2/year for GPP (Figure 4b,g). Semi-arid regions exhibit rates of 0.000913 per year for EVI and 0.500985 gC/m2/year for GPP (Figure 4c,h). Semi-humid regions show rates of 0.001617 per year for EVI and 1.109586 gC/m2/year for GPP (Figure 4d,i). Humid regions demonstrate the most pronounced increases, with rates of 0.002198 per year for EVI and 1.420436 gC/m2/year for GPP (Figure 4e,j). In summary, both EVI and GPP exhibit positive trends at the national level, with the most significant growth in humid regions and relatively slower growth in arid regions.
The inter-annual variations in EVI and GPP for different land cover types from 2001 to 2020 are illustrated in Figure 5. These trends are consistently positive across all land cover types, indicating a gradual enhancement in ecosystem productivity and an improvement in ecosystem health. Specifically, the annual change rates are as follows: for croplands, EVI increases at a rate of 0.002246 per year (Figure 5a) and GPP at a rate of 1.452164 gC/m2 per year (Figure 5f); for forests, EVI increases at a rate of 0.002021 per year (Figure 5b) and GPP at a rate of 1.319233 gC/m2 per year (Figure 5g); for shrubs, EVI increases at a rate of 0.002277 per year (Figure 5c) and GPP at a rate of 1.832376 gC/m2 per year (Figure 5h); for grasslands, EVI increases at a rate of 0.001014 per year (Figure 5d) and GPP at a rate of 0.705504 gC/m2 per year (Figure 5i); and for wetlands, EVI increases at a rate of 0.000604 per year (Figure 5e) and GPP at a rate of 0.813150 gC/m2 per year (Figure 5j).

3.2. Spatial Auto-Correlation of EVI and GPP

The Global Moran’s I indices for EVI and GPP in China are 0.554 (z = 106.85, p < 0.01) and 0.206 (z = 189.39, p < 0.01), respectively, indicating significant spatial clustering of high values for both variables. Specifically, the distribution of EVI hotspots and coldspots exhibits pronounced polarization, with high-value areas predominantly located in the southern part of the study area, while low-value areas are mainly distributed in the central and western regions (Figure 6a). Similarly, GPP hotspots and coldspots display clear polarization, with high-value areas primarily concentrated in the southern part of the study area and low-value areas mainly found in the northeastern region. Notably, a clustering of low GPP values is observed in the downstream area of the Yangtze River (Figure 6b).
Through cluster analysis of EVI and GPP, we found that EVI across China exhibits strong spatial correlation, whereas the spatial correlation of GPP is primarily concentrated in the southern and northeastern regions, consistent with the results of hotspot analysis. Specifically, the spatial distribution of EVI is characterized by two types: low–low (L-L) and high–high (H-H). Low-value aggregations are predominantly located in the northern regions, while high-value aggregations are concentrated in the southern regions. Additionally, low–high (L-H) clusters are mainly distributed in the Huaihe River Basin, the southwestern river basins, and the Pearl River Basin, whereas high–low (H-L) clusters are more dispersed, primarily appearing in the eastern coastal areas (Figure 7a). For GPP, its spatial distribution is characterized by three types: L-L, L-H, and H-H. Low-value aggregations are mainly concentrated in the northeastern regions, the transition from low to high values is primarily found in the Yangtze River Basin, and high-value aggregations are concentrated in the southern regions. H-L clusters are more dispersed, mainly distributed in the downstream areas of the Yangtze River Basin (Figure 7b).

3.3. Correlation Analysis

The spatial distribution of correlation coefficients between EVI, GPP, and soil moisture across China from 2001 to 2020 is illustrated in Figure 8. The correlation patterns between EVI and GPP with soil moisture are spatially similar, showing predominantly negative correlations in the lower reaches of the Haihe, Huaihe, and Yellow River basins, while positive correlations are mainly observed in the Pearl River basin, the Tibetan Plateau, and eastern Inner Mongolia. The areas with positive correlations between EVI and soil moisture at depths SM1, SM2, SM3, and SM4 account for 67.84%, 69.97%, 66.66%, and 54.47%, respectively, whereas the areas with negative correlations are 32.16%, 30.03%, 33.34%, and 45.53% (Figure 8a–d). For GPP, the areas with positive correlations with soil moisture at the same depths are 67.09%, 66.79%, 65.14%, and 52.70%, with negative correlation areas being 32.91%, 33.21%, 34.86%, and 47.30% (Figure 8e–h). Overall, the correlations between EVI and GPP with soil moisture exhibit significant regional and depth variations, with positive correlation areas generally exceeding those of negative correlation.
The correlation between the Enhanced Vegetation Index (EVI), Gross Primary Productivity (GPP), and soil moisture across different climatic zones in China from 2001 to 2020 is illustrated in Figure 9. At the national scale, the correlation coefficients between EVI and soil moisture at depths SM1, SM2, SM3, and SM4 are −0.09, 0.12, −0.02, and −0.46, respectively. Notably, a significant negative correlation is observed between SM4 and EVI, indicating that EVI values decrease as SM4 increases. The correlation coefficients between GPP and soil moisture at depths SM1, SM2, SM3, and SM4 are −0.38, 0.03, 0.15, and −0.07, respectively, none of which are statistically significant (Figure 9a). In arid regions, the correlation coefficients between EVI and soil moisture at depths SM1, SM2, SM3, and SM4 are 0.05, 0.46, −0.07, and 0.33, respectively. A significant positive correlation exists between SM2 and EVI, indicating that EVI values increase with increasing soil moisture at this depth. For GPP, the correlation coefficients with soil moisture at depths SM1, SM2, SM3, and SM4 are 0.22, 0.48, −0.09, and 0.46, respectively, with significant positive correlations for SM2 and SM4, suggesting that GPP values increase as soil moisture at these depths increases (Figure 9b). In semi-arid regions, the correlation coefficients of EVI with soil moisture at depths SM1, SM2, SM3, and SM4 are 0.09, 0.22, 0.30, and −0.00, respectively, and for GPP, they are 0.08, 0.22, 0.30, and −0.05, respectively. None of these correlations are statistically significant (Figure 9c). In semi-humid regions, the correlation coefficients of EVI with soil moisture at depths SM1, SM2, SM3, and SM4 are −0.25, −0.20, −0.23, and −0.80, respectively, all of which are negative. A significant negative correlation is observed for SM4, indicating that EVI values decrease as soil moisture at this depth increases. For GPP, the correlation coefficients with soil moisture at depths SM1, SM2, SM3, and SM4 are −0.21, −0.16, −0.22, and −0.80, respectively, with a significant negative correlation for SM4, showing that GPP values decrease as soil moisture at this depth increases (Figure 9d). In humid regions, the correlation coefficients of EVI with soil moisture at depths SM1, SM2, SM3, and SM4 are 0.28, 0.28, 0.26, and 0.08, respectively, and for GPP, they are 0.37, 0.37, 0.37, and 0.23, respectively. None of these correlations are statistically significant (Figure 9e). Overall, the correlations between EVI and GPP with soil moisture across different climatic regions exhibit distinct differences. Significant correlations are observed at specific soil moisture depths for EVI and GPP in arid and semi-humid regions, while such correlations are non-significant in the humid region.
The correlation between EVI, GPP, and soil moisture across different land cover types from 2001 to 2020 is depicted in Figure 10. For croplands, the correlation coefficients between EVI and soil moisture layers SM1, SM2, SM3, and SM4 are −0.02, 0.01, −0.08, and -0.71, respectively, whereas for GPP, they are 0.04, 0.08, −0.01, and −0.61, respectively. Notably, SM4 exhibits a significant negative correlation with both EVI and GPP, indicating a decline in these values as SM4 increases (Figure 10a). For forests, the correlation coefficients for EVI with SM1, SM2, SM3, and SM4 are 0.13, 0.14, 0.10, and −0.30, respectively, and for GPP, they are 0.21, 0.23, 0.19, and −0.17, respectively, none of which are statistically significant (Figure 10b). For shrubs, the correlation coefficients for EVI with SM1, SM2, SM3, and SM4 are −0.01, 0.05, 0.04, and −0.27, respectively, and for GPP, they are 0.03, 0.08, 0.07, and −0.19, respectively, all of which are non-significant (Figure 10c). Grasslands show correlation coefficients for EVI with SM1, SM2, SM3, and SM4 of −0.11, 0.09, 0.14, and -0.36, respectively, none of which are significant. The corresponding GPP coefficients are -0.18, 0.02, 0.08, and −0.40, respectively, with SM4 showing a significant negative correlation, suggesting a decrease in GPP as SM4 increases (Figure 10d). Wetlands exhibit positive correlations for EVI with SM1, SM2, SM3, and SM4 of 0.23, 0.24, 0.22, and 0.16, respectively, though none are statistically significant. The GPP coefficients for SM1, SM2, SM3, and SM4 are 0.45, 0.48, 0.46, and 0.38, respectively, all of which are positively correlated and statistically significant, indicating an increase in GPP with increasing soil moisture (Figure 10e). SM4 significantly impacts GPP in cropland and grassland ecosystems, with GPP decreasing as soil moisture increases; conversely, GPP in wetlands significantly increases with soil moisture. Other vegetation types show no significant correlation with soil moisture.

4. Discussion

4.1. Spatial and Temporal Distribution Characteristics of EVI and GPP

Analyzing the spatial and temporal dynamics of EVI (Enhanced Vegetation Index) and GPP (Gross Primary Productivity) is crucial for a comprehensive understanding of regional disparities in ecosystem productivity and vegetation health, as well as their responses to environmental changes. Both EVI and GPP exhibit positive growth trends across the nation and in various climatic zones, indicating significant improvements in vegetation growth conditions and steady increases in ecosystem productivity. Specifically, in humid regions, the substantial growth of EVI and GPP is likely closely associated with positive correlations to temperature and precipitation. Vegetation in these regions is highly sensitive to changes in temperature and precipitation. Warmer climates and increased precipitation provide favorable conditions for vegetation growth and enhanced photosynthesis, thereby promoting better vegetation health and increased ecosystem productivity [23,24,25]. Conversely, in arid regions, the growth of EVI and GPP is relatively slow, likely due to limited water resources. Water stress is more pronounced in arid regions, restricting vegetation growth and productivity improvements. Moreover, vegetation responses to climate change vary significantly across different climatic regions. Vegetation in humid regions, with abundant water resources, is more sensitive to temperature changes, whereas vegetation in arid regions is more constrained by water availability [26,27].
During the study period, all land cover types exhibited a positive inter-annual change trend. This trend reflects the dynamism of land cover types in the context of global change and their response to environmental and socioeconomic factors. The positive change trend suggests an improvement in land cover quality, which may be closely related to the global greening trend, urbanization processes, and agricultural expansion [28,29]. However, significant differences in change trends were observed among different land cover types, which may be closely related to their respective ecological functions, geographical distribution, intensity of human activities, and policy orientation [30]. These differences indicate that land cover change is a complex process influenced by multiple factors. It is necessary to comprehensively consider the mechanisms of natural and human factors and their interactions [31,32]. Moreover, land cover change has far-reaching impacts on ecosystem services, biodiversity conservation, and global climate change. Therefore, an in-depth understanding of these change trends and their driving forces is crucial for formulating effective land management policies and sustainable development strategies.
The spatial aggregation trends and clustering characteristics of the Enhanced Vegetation Index (EVI) and Gross Primary Productivity (GPP) can reflect the vegetation growth conditions and ecosystem productivity in different regions of China. High EVI values are predominantly concentrated in the southern regions, likely due to favorable climatic conditions (e.g., higher temperatures and precipitation) and the prevalence of vegetation types such as evergreen broadleaf forests. These conditions support robust vegetation growth and photosynthesis. Conversely, the low-value aggregation of GPP in the Yangtze River Delta region may be attributed to urbanization and human activities, which can alter land use patterns and degrade the ecological environment [33,34]. Additionally, the clustering analysis of EVI and GPP reveals significant differences in vegetation growth conditions between the northern and southern regions, which may be related to the distinct climatic zones and vegetation types. The low–low (L-L) distribution in the northern regions is likely associated with arid and semi-arid climates, while the high–high (H-H) distribution in the southern regions corresponds to humid and subtropical climates. These findings align with the hypothesis of this study that the spatial distribution and clustering characteristics of EVI and GPP can reflect the ecological environment and vegetation growth conditions across different regions. However, the low-value aggregation of GPP in the Yangtze River Delta region is inconsistent with the expected high-productivity areas, suggesting the presence of unique environmental pressures or human-induced impacts in this region.

4.2. Response of EVI and GPP to Soil Moisture

Studying the impact of soil moisture at different depths on EVI and GPP is essential for understanding and predicting ecosystem functions, water resource management, and climate change adaptation strategies [35]. In this research, significant correlations were observed between soil moisture at specific depths and both EVI and GPP in arid and semi-humid regions, whereas such correlations were not significant in humid regions. These findings are consistent with those of Niu Shuli’s team, who demonstrated that ecosystem photosynthesis adapts to soil moisture levels, exhibiting a hump-shaped response curve of GPP to increasing soil moisture, with an optimal soil moisture level at which GPP peaks [36]. This suggests that the optimal soil moisture for arid ecosystems is typically higher than that for humid ecosystems, reflecting the water adaptability of ecosystems. Additionally, Xu Qingchen’s (2021) [36] research indicates that in arid and semi-arid regions, water use efficiency (WUE) is positively correlated with precipitation, suggesting that increased rainfall enhances vegetation’s carbon assimilation rate more than the rate of ecosystem water loss. Collectively, these findings elucidate the impact of soil moisture on ecosystem productivity and its variability across different climatic zones, providing new insights into understanding and predicting carbon-climate feedbacks. In particular, the influence of soil moisture on vegetation growth and photosynthesis is more pronounced in arid and semi-humid regions, likely due to the scarcity of water in these areas, which heightens vegetation’s sensitivity to water use efficiency and response. In contrast, in humid regions, the impact of soil moisture on EVI and GPP may be overshadowed by other environmental factors, resulting in non-significant correlations.
The EVI and GPP of croplands and grasslands exhibited significant negative correlations with SM4. In contrast, the GPP of wetlands significantly increased with rising soil moisture. This finding aligns with previous research suggesting a unimodal relationship between GPP and soil moisture, indicating an optimal soil moisture level at which GPP reaches its maximum value [3]. For croplands and grasslands, exceeding optimal soil moisture levels can trigger root oxygen stress. Under hypoxic conditions, aerobic respiration in roots is significantly weakened, resulting in insufficient ATP synthesis and severely restricting the active absorption of key nutrients like nitrogen and phosphorus. Meanwhile, the accumulation of harmful substances like ethanol produced by anaerobic respiration within cells can damage cell membrane structure and function. Once root physiological functions are impaired, the synthesis of photosynthetic pigments in the aboveground parts is affected, stomatal conductance is reduced, and carbon dioxide uptake is decreased, thereby inhibiting photosynthetic efficiency. Ultimately, this results in a significant negative correlation between EVI, GPP, and deep soil moisture [37,38]. Conversely, wetland ecosystems, characterized by their unique hydrological conditions and plant adaptations, may more effectively utilize increased soil moisture, thereby promoting increases in EVI and GPP. Additionally, the correlation between soil moisture and other vegetation types did not reach a significant level, possibly due to differences in water use strategies and ecological adaptations among vegetation types. For instance, some vegetation types may access water through deep root systems, demonstrating greater resilience under water stress [39,40], which may account for the less pronounced relationship between EVI and GPP and soil moisture compared to croplands and grasslands. These findings underscore the importance of considering soil moisture conditions in global change research and highlight the need for future studies to further investigate the response mechanisms of different vegetation types to changes in soil moisture and how these responses affect ecosystem productivity and carbon cycling.

4.3. Limitations and Future Research

Although this study focused on the impact of soil moisture at different depths on ecosystem stability (measured by EVI and GPP), it primarily examined soil moisture as the main variable. Other factors influencing ecosystem stability, such as atmospheric CO2 concentration, temperature, precipitation, and radiation, were not considered. It is worth noting that potential uncertainties stemming from data sources and modeling approaches, such as MODIS signal saturation effects, inherent assumptions in GPP model formulations, and potential biases in reanalysis datasets, were not explicitly discussed in this study. These aspects warrant careful consideration and evaluation in future research endeavors. Previous studies have shown that changes in atmospheric CO2 concentration significantly affect GPP, with its contribution varying spatially across regions. While some of these variables also contribute importantly to ecosystem stability, they were excluded from our study. Moreover, changes in ecosystems are not solely driven by natural factors; human activities, such as land use changes and irrigation, may interfere with and even amplify or weaken the impact of soil moisture on ecosystem stability. Our study did not quantitatively analyze the effects of such human activities. Therefore, further research is needed to comprehensively and quantitatively evaluate the impact of various environmental factors on ecosystem stability. This study used EVI and GPP data to assess the influence of soil moisture at different depths on ecosystem stability across different climatic zones in China, thereby enhancing our understanding of potential changes in ecosystem stability under future climate change.

5. Conclusions

Based on long-term observational data from across the country, this study conducted a comprehensive analysis of the Ecosystem Vegetation Index (EVI), Gross Primary Productivity (GPP), and soil moisture conditions. The findings reveal that ecosystem productivity, as indicated by EVI and GPP, shows clear spatial clustering and positive long-term trends nationwide. Specifically, the most significant increases in EVI and GPP were observed in humid regions, while arid regions exhibited relatively slower growth. This highlights the overall enhancement of ecosystem productivity in China, particularly in humid areas. Across various land cover types, both EVI and GPP demonstrated positive inter-annual variation trends, further confirming the gradual strengthening of ecosystem productivity.
The spatial correlation of EVI in China is generally strong, characterized by low–low (L-L) clusters in the north and high–high (H-H) clusters in the south. Low–high (L-H) clusters are concentrated in the Huaihai Plain, southwestern rivers, and the Pearl River Basin, while high–low (H-L) clusters are scattered along the east coast. The spatial correlation of GPP is more pronounced in the southern and northeastern regions, exhibiting L-L clusters in the northeast, L-H clusters in the Yangtze River Basin, and H-H clusters in the south. H-L clusters are dispersed in the downstream area of the Yangtze River. Both EVI and GPP show a significant tendency for spatial aggregation of high values. Specifically, the distribution of EVI hotspots and coldspots exhibits clear polarization, with high-value areas predominantly located in the southern part of the study area and low-value areas mainly distributed in the central and western regions. For GPP, hotspots and coldspots also display distinct polarization, with high-value areas concentrated in the southern part and low-value areas concentrated in the northeastern region. Notably, a low-value aggregation is observed in the Yangtze River Delta region.
In distinct climatic zones, the correlation between EVI and GPP with soil moisture shows significant variation. In arid and semi-humid regions, a pronounced correlation exists between specific soil moisture depths and both EVI and GPP. However, this correlation is not significant in humid regions. When examining different land cover types, EVI and GPP values for croplands and grasslands exhibit a significant negative correlation with SM4, indicating that SM4 is a key constraint in these areas. Conversely, GPP in wetlands increases significantly with increasing soil moisture, highlighting contrasting ecosystem strategies. These findings suggest that wetlands benefit from high soil moisture, while croplands and grasslands may face challenges when soil moisture exceeds optimal levels.
These insights are essential for understanding the impacts of climate change on China’s ecosystems and for guiding ecological conservation and land management practices. In arid regions, water-saving measures should be implemented to optimize soil moisture conditions and enhance ecosystem productivity. In contrast, wetland protection strategies should focus on maintaining high soil moisture levels to support robust GPP. Future research should investigate the mechanisms underlying the interactions between soil moisture and ecosystem productivity under different climatic conditions and explore strategies to enhance ecosystem adaptability and stability through scientific management.

Author Contributions

Conceptualization, Y.L. and I.M.; methodology, Y.L. and Z.G.; software, Y.L., L.S. and C.L.; validation, Y.L., J.H. and L.S.; formal analysis, Y.L.; investigation, M.W.; resources, X.D.; data curation, Y.L.; writing—original draft preparation, Y.L.; writing—review and editing, Y.L.; visualization, Y.L.; supervision, Y.Y. and R.Y.; project administration, Y.Y.; funding acquisition, Y.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Key Research and Development Program of China (2023YFF0805603); Key Research and Development Program of Xinjiang: 2022B01032-4.

Data Availability Statement

The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding author.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. He, Q.; Lu, H.; Yang, K.; Zhen, L.; Yue, S.; Li, Y.; Entekhabi, D. Global Patterns of Vegetation Response to Short-Term Surface Water Availability. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2021, 14, 8273–8286. [Google Scholar] [CrossRef]
  2. Fu, Z.; Ciais, P.; Wigneron, J.P.; Gentine, P.; Feldman, A.F.; Makowski, D.; Viovy, N.; Kemanian, A.R.; Goll, D.S.; Stoy, P.C.; et al. Global critical soil moisture thresholds of plant water stress. Nat. Commun. 2024, 15, 4826. [Google Scholar] [CrossRef] [PubMed]
  3. Peng, J.; Tang, J.; Xie, S.; Wang, Y.; Liao, J.; Chen, C.; Sun, C.; Mao, J.; Zhou, Q.; Niu, S. Evidence for the acclimation of ecosystem photosynthesis to soil moisture. Nat. Commun. 2024, 15, 9795. [Google Scholar] [CrossRef] [PubMed]
  4. Li, Y.; Leng, P.; Kasim, A.A.; Li, Z.-L. Spatiotemporal variability and dominant driving factors of satellite observed global soil moisture from 2001 to 2020. J. Hydrol. 2025, 654, 132848. [Google Scholar] [CrossRef]
  5. Xu, X.; Yu, G. Theories of ecosystem vulnerability, adaptability and catastrophe based on the mechanisms of ecological succession. Chin. J. Appl. Ecol. 2022, 33, 623–628. [Google Scholar]
  6. Liu, X.; Zhou, H.; Li, P.; Peng, S. A conceptual analysis of ecosystem stability. Acta Ecol. Sin. 2004, 11, 2635–2640. [Google Scholar]
  7. Higgins, S.I.; Conradi, T.; Muhoko, E. Shifts in vegetation activity of terrestrial ecosystems attributable to climate trends. Nat. Geosci. 2023, 16, 147–153. [Google Scholar] [CrossRef]
  8. Liu, G.; Wang, Y.; Chen, Y.; Tong, X.; Wang, Y.; Xie, J.; Tang, X. Remotely Monitoring Vegetation Productivity in Two Contrasting Subtropical Forest Ecosystems Using Solar-Induced Chlorophyll Fluorescence. Remote Sens. 2022, 14, 1328. [Google Scholar] [CrossRef]
  9. Chen, X.; Li, Y.; Wang, Z. Soil Moisture Dynamics and Its Impact on Vegetation in the Loess Plateau of China. J. Hydrol. 2020, 584, 124703. [Google Scholar]
  10. Zhang, Y.; Wang, L.; Liu, H. Effects of Soil Moisture on the Stability of Grassland Ecosystems in Inner Mongolia. Ecol. Eng. 2019, 134, 143–150. [Google Scholar]
  11. Green, M.; White, J.; Black, S. Impact of Soil Moisture on Carbon Sequestration in Forest Ecosystems. For. Ecol. Manag. 2022, 513, 119567. [Google Scholar]
  12. Brown, R.; Smith, A.; Johnson, B. Effects of Soil Moisture on the Stability of Agricultural Ecosystems. Agric. Syst. 2021, 188, 103056. [Google Scholar]
  13. Zhang, L.; Li, H.; Wang, Y. Soil Moisture and Its Role in Maintaining Biodiversity in Wetland Ecosystems. Wetlands 2020, 40, 123–134. [Google Scholar]
  14. Zhang, X.; Friedl, M.A.; Schaaf, C.B.; Strahler, A.H.; Hodges, J.C.F.; Gao, F.; Reed, B.C.; Huete, A. Monitoring vegetation phenology using MODIS. Remote Sens. Environ. 2003, 84, 471–475. [Google Scholar] [CrossRef]
  15. Yao, R.; Zhang, Y.; Wang, L.; Li, J.; Yang, Q. Reconstructed NDVI and EVI datasets in China (ReVIChina) generated by a spatial-interannual reconstruction method. Int. J. Digit. Earth 2023, 16, 4749–4768. [Google Scholar] [CrossRef]
  16. Pandey, V.; Harde, S.; Rajasekaran, E.; Burman, P.K.D. Gross primary productivity of terrestrial ecosystems: A review of observations, remote sensing, and modelling studies over South Asia. Theor. Appl. Climatol. 2024, 155, 8461–8491. [Google Scholar] [CrossRef]
  17. Bi, W.; He, W.; Zhou, Y.; Ju, W.; Liu, Y.; Liu, Y.; Zhang, X.; Wei, X.; Cheng, N. A global 0.05° dataset for gross primary production of sunlit and shaded vegetation canopies from 1992 to 2020. Sci. Data 2022, 9, 213. [Google Scholar] [CrossRef]
  18. Frimpong, B.F.; Koranteng, A.; Molkenthin, F. Analysis of temperature variability utilising Mann-Kendall and Sen’s slope estimator tests in the Accra and Kumasi Metropolises in Ghana. Environ. Syst. Res. 2022, 11, 24. [Google Scholar] [CrossRef]
  19. Liu, J.; Wei, L.; Zheng, Z.; Du, J. Vegetation cover change and its response to climate extremes in the Yellow River Basin. Sci. Total Environ. 2023, 905, 167366. [Google Scholar] [CrossRef]
  20. Chen, Y. Spatial autocorrelation equation based on Moran’s index. Sci. Rep. 2023, 13, 19296. [Google Scholar] [CrossRef]
  21. Wang, Y.; Lv, W.; Wang, M.; Chen, X.; Li, Y. Application of improved Moran’s I in the evaluation of urban spatial development. Spat. Stat. 2023, 54, 100736. [Google Scholar] [CrossRef]
  22. Gago, T.; Sargisson, R.J.; Milfont, T.L. A meta-analysis on the relationship between climate anxiety and wellbeing. J. Environ. Psychol. 2024, 94, 102230. [Google Scholar] [CrossRef]
  23. Zhao, Q.Q.; Zhang, J.P.; Zhao, T.B.; Li, J. Vegetation Changes and Its Response to Climate Change in China Since 2000. Plateau Meteorol. 2021, 40, 292–301. [Google Scholar]
  24. Li, M.H.; Du, J.K.; Li, W.T.; Li, R.; Wu, S.; Wang, S. Global Vegetation Change and Its Relationship with Precipitation and Temperature Based on GLASS-LAI in 1982–2015. Sci. Geogr. Sin. 2020, 40, 823–832. [Google Scholar]
  25. Fang, Z.; Zhang, W.; Wang, L.; Schurgers, G.; Ciais, P.; Peñuelas, J.; Brandt, M.; Yang, H.; Huang, K.; Shen, Q.; et al. Global increase in the optimal temperature for the productivity of terrestrial ecosystems. Commun. Earth Environ. 2024, 5, 466. [Google Scholar] [CrossRef]
  26. Zhang, S.Z.; Zhu, X.F.; Liu, T.T.; Xu, K.; Guo, R. Response of gross primary production to drought under climate change in different vegetation regions of China. Acta Ecol. Sin. 2022, 42, 3429–3440. [Google Scholar]
  27. Chen, J.; Yang, H.; Jin, T.; Wu, K. Assessment of terrestrial ecosystem sensitivity to climate change in arid, semi-arid, sub-humid, and humid regions using EVI, LAI, and SIF products. Ecol. Indic. 2024, 158, 111511. [Google Scholar] [CrossRef]
  28. Zhang, J.G.; Xiao, J.F.; Wang, L.Y.; Tong, X.J.; Zhang, J.S.; Li, J.; Liu, P.R.; Yu, P.Y.; Meng, P. Comparing the performance of phenocam GCC, MODIS GCC, and MODIS EVI for retrieving vegetation phenology and estimating gross primary production. Ecol. Indic. 2024, 166, 112251. [Google Scholar] [CrossRef]
  29. Aryal, J.; Sitaula, C.; Frery, A.C. Land use and land cover (LULC) performance modeling using machine learning algorithms: A case study of the city of Melbourne, Australia. Sci. Rep. 2023, 13, 13510. [Google Scholar] [CrossRef]
  30. Shi, H.; Li, L.H.; Eamus, D.; Huete, A.; Cleverly, J.; Tian, X.; Yu, Q.; Wang, S.Q.; Montagnani, L.; VMagliulo, V.; et al. Assessing the ability of MODIS EVI to estimate terrestrial ecosystem gross primary production of multiple land cover types. Ecol. Indic. 2017, 72, 153–164. [Google Scholar] [CrossRef]
  31. Zhang, Y.; Xiao, X.; Wu, X.; Zhou, S.; Zhang, G.; Qin, Y.; Dong, J. A global moderate resolution dataset of gross primary production of vegetation for 2000–2016. Sci. Data 2017, 4, 170165. [Google Scholar] [CrossRef]
  32. Jing, Q.; He, J.; Li, Y.; Yang, X.; Peng, Y.; Wang, H.; Yu, F.; Wu, J.; Gong, S.; Che, H.; et al. Analysis of the spatiotemporal changes in global land cover from 2001 to 2020. Sci. Total Environ. 2024, 908, 168354. [Google Scholar] [CrossRef] [PubMed]
  33. Li, H.; Wu, X.; Ma, D.; Wei, C.; Peng, B.; Du, W. Spatiotemporal variation characteristics and influencing factors analysis of the GPP in the Tibetan Plateau. Remote Sens. Technol. Appl. 2024, 39, 727–740. [Google Scholar]
  34. Xu, N.; Luo, J.; Ma, R.; Jia, J.; Chen, Y.; Zhou, X. Dynamic change of land use/cover spatial pattern in Yangtze River Delta from 2000 to 2010. Quat. Sci. 2014, 34, 856–864. [Google Scholar]
  35. Dong, G.; Chen, S.; Liu, K.; Wang, W.; Hou, H.; Gao, L.; Zhang, F.; Su, H. Spatiotemporal variation in sensitivity of urban vegetation growth and greenness to vegetation water content: Evidence from Chinese megacities. Sci. Total Environ. 2023, 905, 167090. [Google Scholar] [CrossRef]
  36. Xu, Q.C. Spatiotemporal variation of water use efficiency and its influencing factors in arid and semi-arid areas of China. Geogr. Sci. Res. 2021, 10, 126–136. [Google Scholar]
  37. Xu, Z.; Zhou, G. Responses of photosynthetic capacity to soil moisture gradient in perennial rhizome grass and perennial bunchgrass. BMC Plant Biol. 2011, 11, 21. [Google Scholar] [CrossRef]
  38. Liu, L.B.; Gudmundsson, L.; Hauser, M.; Qin, D.; Li, S.; Seneviratne, S.I. Soil moisture dominates dryness stress on ecosystem production globally. Nat. Commun. 2020, 11, 4892. [Google Scholar] [CrossRef]
  39. Yang, M.; Gao, X.; Wang, S.; Zhao, X. Quantifying the importance of deep root water uptake for apple trees’ hydrological and physiological performance in drylands. J. Hydrol. 2022, 606, 127471. [Google Scholar] [CrossRef]
  40. Knipfer, T. Future in the past: Water uptake function of root systems. Plant Soil 2022, 481, 495–500. [Google Scholar] [CrossRef]
Figure 1. Overview map of the study area (notes: AR: arid region; SAR: semi-arid region; SHR: semi-humid region; HR: humid region).
Figure 1. Overview map of the study area (notes: AR: arid region; SAR: semi-arid region; SHR: semi-humid region; HR: humid region).
Remotesensing 17 01971 g001
Figure 2. Spatial distribution and Slope values of EVI and GPP (notes: (a): EVI spatial distribution; (b): GPP spatial distribution; (c): EVI Slope value; (d): GPP Slope value).
Figure 2. Spatial distribution and Slope values of EVI and GPP (notes: (a): EVI spatial distribution; (b): GPP spatial distribution; (c): EVI Slope value; (d): GPP Slope value).
Remotesensing 17 01971 g002
Figure 3. Variation trends of EVI and GPP in different climate regions and land cover types (notes: WR: whole region; AR: arid region; SAR: semi-arid region; SHR: semi-humid region; HR: humid region).
Figure 3. Variation trends of EVI and GPP in different climate regions and land cover types (notes: WR: whole region; AR: arid region; SAR: semi-arid region; SHR: semi-humid region; HR: humid region).
Remotesensing 17 01971 g003
Figure 4. Inter-annual changes in EVI and GPP in different climate regions (notes: (ae): inter-annual variation in EVI in the whole region, arid region, semi-arid region, sub-humid region and humid region; (fj): inter-annual variation in GPP in whole region, arid region, semi-arid region, sub-humid region and humid region).
Figure 4. Inter-annual changes in EVI and GPP in different climate regions (notes: (ae): inter-annual variation in EVI in the whole region, arid region, semi-arid region, sub-humid region and humid region; (fj): inter-annual variation in GPP in whole region, arid region, semi-arid region, sub-humid region and humid region).
Remotesensing 17 01971 g004
Figure 5. Inter-annual variation in EVI and GPP for different land cover types (notes: (ae): inter-annual variation in EVI in croplands, forests, shrubs, grasslands and wetlands; (fj): inter-annual variation in GPP in croplands, forests, shrubs, grasslands, and wetlands).
Figure 5. Inter-annual variation in EVI and GPP for different land cover types (notes: (ae): inter-annual variation in EVI in croplands, forests, shrubs, grasslands and wetlands; (fj): inter-annual variation in GPP in croplands, forests, shrubs, grasslands, and wetlands).
Remotesensing 17 01971 g005
Figure 6. Spatial distribution of cold- and hotspots of EVI and GPP in China from 2001 to 2020 (notes: (a): spatial distribution of cold- and hotspots of EVI; (b): spatial distribution of cold- and hotspots of GPP).
Figure 6. Spatial distribution of cold- and hotspots of EVI and GPP in China from 2001 to 2020 (notes: (a): spatial distribution of cold- and hotspots of EVI; (b): spatial distribution of cold- and hotspots of GPP).
Remotesensing 17 01971 g006
Figure 7. Spatial distribution of Local Morans’s I of EVI and GPP in China from 2001 to 2020 (notes: (a): spatial distribution of Local Morans’s I of EVI; (b): spatial distribution of Local Morans’s I of GPP).
Figure 7. Spatial distribution of Local Morans’s I of EVI and GPP in China from 2001 to 2020 (notes: (a): spatial distribution of Local Morans’s I of EVI; (b): spatial distribution of Local Morans’s I of GPP).
Remotesensing 17 01971 g007
Figure 8. Spatial distribution of correlation coefficients between EVI, GPP and soil moisture (notes: (ad): spatial distribution of correlation between EVI and SM1, SM2, SM3 and SM4; (eh): spatial distribution of correlation between GPP and SM1, SM2, SM3 and SM4).
Figure 8. Spatial distribution of correlation coefficients between EVI, GPP and soil moisture (notes: (ad): spatial distribution of correlation between EVI and SM1, SM2, SM3 and SM4; (eh): spatial distribution of correlation between GPP and SM1, SM2, SM3 and SM4).
Remotesensing 17 01971 g008
Figure 9. Correlation between EVI, GPP and soil moisture in different climate zones (notes: (a): the correlation of EVI and GPP with soil moisture in China; (b): correlation between EVI and GPP with soil moisture in arid region; (c): correlation between EVI and GPP with soil moisture in semi-arid region; (d): correlation between EVI and GPP with soil moisture in sub-humid region; (e): correlation between EVI and GPP with soil moisture in humid region. “**” indicates extremely significant correlation (p < 0.01)).
Figure 9. Correlation between EVI, GPP and soil moisture in different climate zones (notes: (a): the correlation of EVI and GPP with soil moisture in China; (b): correlation between EVI and GPP with soil moisture in arid region; (c): correlation between EVI and GPP with soil moisture in semi-arid region; (d): correlation between EVI and GPP with soil moisture in sub-humid region; (e): correlation between EVI and GPP with soil moisture in humid region. “**” indicates extremely significant correlation (p < 0.01)).
Remotesensing 17 01971 g009
Figure 10. Correlation between different land cover types and soil moisture (notes: (a): correlation between EVI and GPP with soil moisture in croplands; (b): correlation between EVI and GPP with soil moisture in forests; (c): correlation between EVI and GPP with soil moisture in shrubs; (d): correlation between EVI and GPP with soil moisture in grasslands; (e): correlation between EVI and GPP with soil moisture in wetlands. “*” indicates significant correlation (p < 0.05); “**” indicates extremely significant correlation (p < 0.01)).
Figure 10. Correlation between different land cover types and soil moisture (notes: (a): correlation between EVI and GPP with soil moisture in croplands; (b): correlation between EVI and GPP with soil moisture in forests; (c): correlation between EVI and GPP with soil moisture in shrubs; (d): correlation between EVI and GPP with soil moisture in grasslands; (e): correlation between EVI and GPP with soil moisture in wetlands. “*” indicates significant correlation (p < 0.05); “**” indicates extremely significant correlation (p < 0.01)).
Remotesensing 17 01971 g010
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Lu, Y.; Yu, Y.; Ding, X.; Sun, L.; Li, C.; He, J.; Guo, Z.; Malik, I.; Wistuba, M.; Yu, R. Adaptive Analysis of Ecosystem Stability in China to Soil Moisture Variations: A Perspective Based on Climate Zoning and Land Use Types. Remote Sens. 2025, 17, 1971. https://doi.org/10.3390/rs17121971

AMA Style

Lu Y, Yu Y, Ding X, Sun L, Li C, He J, Guo Z, Malik I, Wistuba M, Yu R. Adaptive Analysis of Ecosystem Stability in China to Soil Moisture Variations: A Perspective Based on Climate Zoning and Land Use Types. Remote Sensing. 2025; 17(12):1971. https://doi.org/10.3390/rs17121971

Chicago/Turabian Style

Lu, Yuanbo, Yang Yu, Xiaoyun Ding, Lingxiao Sun, Chunlan Li, Jing He, Zengkun Guo, Ireneusz Malik, Malgorzata Wistuba, and Ruide Yu. 2025. "Adaptive Analysis of Ecosystem Stability in China to Soil Moisture Variations: A Perspective Based on Climate Zoning and Land Use Types" Remote Sensing 17, no. 12: 1971. https://doi.org/10.3390/rs17121971

APA Style

Lu, Y., Yu, Y., Ding, X., Sun, L., Li, C., He, J., Guo, Z., Malik, I., Wistuba, M., & Yu, R. (2025). Adaptive Analysis of Ecosystem Stability in China to Soil Moisture Variations: A Perspective Based on Climate Zoning and Land Use Types. Remote Sensing, 17(12), 1971. https://doi.org/10.3390/rs17121971

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop