Next Article in Journal
Defining a Method for Mapping Aeolian Sand Transport Susceptibility Using Bivariate Statistical and Machine Learning Methods—A Case Study of the Seqale Watershed, Eastern Iran
Previous Article in Journal
A Multi-Platform Online Data-Driven Diagnostic Approach for Macro-Level Sustainability of Homestays
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 17
Issue 18
10.3390/su17188233
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Vegetation Response to Climate and Human Interventions on the Loess Plateau: Trends, Variability, and the Influence of the Grain for Green Program

by
Jiangbo Li
1,
Huan Liu
2,*,
Dandong Cheng
1,*,
Hangzhen Zhang
1,
Guizeng Qi
1 and
Weize Wang
3
1
Xi’an Key Laboratory of Environmental Simulation and Ecological Health in the Yellow River Basin, College of Urban and Environmental Sciences, Northwest University, Xi’an 710127, China
2
State Key Laboratory of Water Cycle and Water Security, China Institute of Water Resources and Hydropower Research, Beijing 100048, China
3
State Key Laboratory Base of Eco-Hydraulic Engineering in Arid Area, Xi’an University of Technology, Xi’an 710048, China
*
Authors to whom correspondence should be addressed.
Sustainability 2025, 17(18), 8233; https://doi.org/10.3390/su17188233
Submission received: 29 July 2025 / Revised: 8 September 2025 / Accepted: 9 September 2025 / Published: 12 September 2025
(This article belongs to the Special Issue Vegetation–Soil–Water Nexus: Ecological Solutions for the Environment)
Browse Figures
Versions Notes

Abstract

Since the launch of the Grain for Green (GFG) program in 1999, the Loess Plateau has undergone significant vegetation changes. However, the driving mechanisms behind these changes in the post-GFG period remain insufficiently understood. This study analyzes the spatiotemporal dynamics of vegetation on the Loess Plateau from 1982 to 2015, based on long-term NDVI time series, and quantitatively identifies the relative contributions of climate variability and human activities using partial correlation and multiple regression residual analysis. The results reveal a significant increase in NDVI after 2000, with the annual variation rate rising from 0.0009 to 0.0028, and the proportion of rapidly greening areas expanding from 13.3% to 62.9%. Spatially, vegetation recovery was more prominent in the eastern and lower-latitude regions. While both climate and anthropogenic factors influenced vegetation changes, the latter became dominant after 2000. The area where human activities significantly enhanced vegetation increased from 1.9% to 60.6%, with the most notable improvements observed in forests, followed by croplands and grasslands. Vegetation in the southern plateau was more sensitive to temperature, while the northern region responded more strongly to precipitation. From 2000 to 2015, the GFG program contributed to increases of 17,059.46 km2 in grasslands and 10,105.78 km2 in forests. These findings improve our understanding of vegetation change drivers on the Loess Plateau and offer a scientific basis for ecological restoration, policy-making, and sustainable development in the Yellow River Basin.

1. Introduction

Vegetation is a fundamental component of terrestrial ecosystems, playing a crucial role in regulating energy flows, biogeochemical cycles, and climate systems [1]. Changes in vegetation dynamics reflect environmental variation and are closely linked to ecosystem health and resilience [2,3]. Understanding the spatiotemporal patterns of vegetation change and their driving mechanisms is essential for assessing ecological stability, predicting ecosystem responses under global change scenarios, and informing land use and environmental management policies [4]. The Grain for Green (GFG) program, launched by the Chinese government in 1999, is one of the world’s largest ecological restoration initiatives. Its primary objective was to mitigate soil erosion and land degradation by converting steep-sloped and marginal farmlands into forests and grasslands. The program was implemented through government subsidies and incentives that encouraged farmers to retire croplands and participate in afforestation and revegetation projects [5]. On the Loess Plateau, where fragile ecosystems and severe soil erosion had long constrained sustainable development, the GFG program profoundly reshaped land-use patterns [6]. Large areas of cropland were converted to forest and grassland, leading to significant improvements in vegetation cover as reflected in NDVI trends. This policy-driven transformation represents a major ecological turning point in the region and provides a critical temporal benchmark for analyzing vegetation dynamics before and after 2000 [7]. Understanding the driving mechanisms behind this recovery is crucial for revealing the combined effects of climate change and human activities on regional ecosystems, and for providing theoretical guidance for ecological management and restoration in similarly vulnerable areas.
Vegetation indices (VIs) are widely used in remote sensing, particularly in agriculture and ecological studies. The Normalized Difference Vegetation Index (NDVI) is the most commonly used VI due to its high accuracy, low computational cost, and broad applicability [8,9]. While the Enhanced Vegetation Index (EVI) is more sensitive in dense vegetation areas and corrects for atmospheric and background noise, it can be less effective in areas with sparse vegetation [10]. The Soil-Adjusted Vegetation Index (SAVI) is better suited for arid regions as it reduces the influence of soil brightness [11,12]. However, NDVI is favored for its stability and effectiveness in large-scale, long-term vegetation monitoring. These characteristics make NDVI particularly appropriate for long-term and large-scale assessments such as those conducted in the Loess Plateau.
Numerous studies have demonstrated that climatic factors, particularly precipitation and temperature variability, play a dominant role in regulating vegetation dynamics, while large-scale ecological restoration projects such as the Grain for Green program have also significantly enhanced NDVI in the Loess Plateau [13]. Many studies have already conducted integrated analyses of climatic and anthropogenic drivers, often using data with spatial resolutions ranging from several kilometers down to 30 m [14,15,16]. These advances have greatly improved our understanding of vegetation change. Nevertheless, uncertainties remain regarding the relative contributions of climate variability and human activities, as well as their interactions across different spatial and temporal scales. These factors are broadly categorized into climatic drivers—such as temperature and precipitation—and anthropogenic drivers, including urbanization, grazing, and afforestation (e.g., the Grain for Green program) [17,18]. The sensitivity of vegetation to specific climatic variables may differ regionally. For example, rising temperatures inducing droughts in the Northern Hemisphere were a primary cause of vegetation decline at higher latitudes during the 1980s [19]. Conversely, some studies suggest that global warming has been the primary driver of increased vegetation cover in the Northern Hemisphere [20,21]. Moreover, the relative influence of different climatic factors on the same region can shift over time; in the Loess Plateau, air temperature predominantly influences vegetation cover in spring and fall, whereas rainfall has a more pronounced impact during summer [5]. Recent advances have emphasized that vegetation dynamics are not only shaped by concurrent climate conditions but also by the cumulative and lagged effects of climatic factors. Studies have demonstrated that precipitation, temperature, and solar radiation can exert intra-annual cumulative influences on vegetation growth across different biomes, with effects persisting beyond the immediate season [22]. Compared to climatic factors, human activities exert more complex and variable, yet often substantial, impacts on vegetation. In the 1950s, rapid population growth and industrialization triggered widespread deforestation and land reclamation in the Loess Plateau, leading to significant vegetation loss and aggravated soil erosion [23]. More recently, China and India have contributed over one-third of the global net increase in leaf area from 2000 to 2017—despite accounting for only 9% of the global vegetated area—primarily through large-scale afforestation and intensified agriculture [4]. These findings highlight the growing influence of human-induced land cover change on vegetation dynamics, particularly in regions undergoing ecological restoration or land-use transformation.
The Loess Plateau, situated in a semi-arid and semi-humid climate zone, experiences severe soil erosion and extensive human activity, resulting in a fragile ecological environment [24,25]. This region is crucial for soil and water conservation and vegetation restoration in China [26]. Following the implementation of the GFG program, significant enhancements in vegetation cover have been observed in the Loess Plateau [27]. Consequently, this region has become a critical area for examining the joint effects of climate variability and human activities on ecosystem dynamics. In previous studies, various methods have been employed to analyze vegetation drivers, including correlation analysis, geographic detector, and quantitative attribution models. Correlation-based approaches are commonly used due to their simplicity, yet they often fail to disentangle the overlapping effects of multiple drivers or capture their interactions [28]. The geographic detector method, while effective in identifying spatial heterogeneity and measuring the explanatory power of individual factors, primarily focuses on spatial patterns and lacks temporal depth [29,30].
This study integrates high-resolution climate data, land-use records, and vegetation indices to quantify the relative impacts of climatic and anthropogenic drivers on vegetation recovery in the Loess Plateau (Figure 1). Our objectives are to: (1) provide a spatiotemporally explicit assessment of multiple drivers, (2) reveal the mechanisms by which climate and human activities jointly influence vegetation dynamics, and (3) inform effective strategies for ecological restoration under ongoing environmental change.

2. Materials and Methods

2.1. Study Area

The Loess Plateau, located in central and northern China between latitudes 33.41° N–41.31° N and longitudes 100.64° E–114.74° E, spans approximately 640,000 km2 across the provinces of Henan, Shaanxi, Shanxi, Inner Mongolia, Gansu, Qinghai, and Ningxia. Elevation ranges from 92 to 4921 m, generally decreasing from southwest to northeast (Figure 2). Characterized by a continental monsoon climate, the Loess Plateau experiences average annual temperatures from 3.6 °C to 14.3 °C and precipitation that increases from west to east, varying between 150 and 700 mm annually. The diverse vegetation, predominantly comprising forests, farmlands, and grasslands, mirrors the significant spatial variations in climate and topography across the plateau [31]. In addition to natural variability, the region has been heavily affected by human activities over recent decades. Long-term deforestation, overgrazing, and agricultural expansion have contributed to severe soil erosion and ecological degradation. Since 1999, the implementation of the GFG program has led to widespread afforestation and cropland retirement across the Plateau, significantly altering land cover and vegetation dynamics [5,30]. These intensive land-use changes make the region particularly suitable for examining the combined effects of climate change and anthropogenic interventions on vegetation patterns.

2.2. Datasets

The NDVI data used in this study were obtained from the Advanced Very High Resolution Radiometer (AVHRR) onboard the National Oceanic and Atmospheric Administration (NOAA) satellites (https://www.NASA.gov, accessed on 7 September 2025). The imagery, provided in 32-bit GeoTIFF format, adopted an Albers Conic projection with a spatial resolution of 8 km × 8 km and a temporal resolution of 15 days, covering the period from 1982 to 2015. Monthly data were extracted from the GIMMS NDVI dataset using the Maximum Value Composite (MVC) method, an internationally recognized technique for processing NDVI data [32]. Annual NDVI values were obtained by averaging the 12 monthly NDVI values for each year using the Map Algebra tool in ArcGIS 10.8. This procedure ensures that aggregated NDVI reflects overall vegetation dynamics rather than being biased by extreme monthly values.
Meteorological data, comprising monthly precipitation and average temperature records from each meteorological station within the Loess Plateau and its adjacent areas from 1982 to 2015, were sourced from the China Meteorological Science Data Sharing Service (https://data.cma.cn/en, accessed on 7 September 2025). The spatial resolution of the interpolated meteorological data is 1 km. These data were processed using ANUSPLIN 4.3 software, employing the thin-plate smoothing spline function for effective spatial interpolation, a method that synergistically integrates with DEM data to interpolate meteorological variables accurately.
The China Land Cover Dataset Land Use/Cover Change (CLCD LUCC) was obtained from a comprehensive dataset developed by Yang Jie at Wuhan University, which integrates multi-sensor remote sensing observations spanning the period from 1982 to 2015. The data were processed and analyzed using the Google Earth Engine platform (https://zenodo.org/records/441 (accessed on 7 September 2025)). Land use classification was performed using the Random Forest Classifier, with a spatial resolution of 30 m [33]. Adhering to the National Land Use Remote Sensing Classification System (NLURSC) standards and accounting for the distinct topographic and geomorphological features of the Loess Plateau, land use types were categorized into five principal groups: grasslands, forests, farmland, shrubland, and non-vegetated areas [34].
To ensure clarity regarding spatial resolution, NDVI data were analyzed at their original resolution of 8 × 8 km, while meteorological variables (temperature and precipitation) and land use/cover (LUCC) data were analyzed at a finer resolution of 1 × 1 km. No resampling was applied; each dataset was processed and analyzed independently at its native resolution to preserve its intrinsic characteristics. Given that this study focuses on large-scale vegetation dynamics and spatial gradients across the Loess Plateau, the differences in spatial resolution among datasets are not expected to substantially influence the main conclusions [35,36].

2.3. Methodology

2.3.1. Change Trend Analysis

Trend analysis: univariate linear regression was applied to calculate the inter-annual trend of NDVI, with the slope of the linear regression equation representing the rate of change in NDVI over time [37]. The Slope was calculated using the following formula:
S l o p e = n × ∑ i = 1 n ( i × N D V I i ) − ∑ i = 1 n i ∑ i = 1 n N D V I i n × ∑ i = 1 n i 2 − ∑ i = 1 n i
where the Slope represents the rate of change in the NDVI over time as determined by the univariate linear regression equation fitted to the time variable. Here, i denotes the time variable, ranging from 1 to n; n is the total number of years in the study period; and NDVIi is the average NDVI for year i. The Slope < 0 indicates a decreasing NDVI trend, while the Slope > 0 indicates an increasing trend. The magnitude of the slope reflects the rate of change, with larger absolute values indicating more rapid changes in NDVI.

2.3.2. Analysis of Multiple Regression

Multiple regression residual analysis: Multiple regression residual analysis (MRA) was employed to evaluate the impacts and relative contributions of human activities and climate change on NDVI changes in vegetation [14,38]. This method involves three main steps: (1) A bivariate linear regression model was established using NDVI as the dependent variable and interpolated time series data of temperature and precipitation as independent variables. The model parameters were then calculated [39]. Using these parameters, the predicted NDVI values were computed based on temperature and precipitation data, representing the influence of climate factors on vegetation NDVI. (2) The predicted NDVI values (NDVICC), calculated from the temperature and precipitation data and the regression model parameters, were used to quantify the impact of climatic factors on vegetation NDVI [40]. (3) The NDVI residual (NDVIHA), defined as the difference between the observed NDVI and NDVICC, was calculated to represent the effect of human activities on vegetation NDVI. The specific calculation formula is as follows:
N D V I C C = a × T + b × P + C
N D V I H A = N D V I o b s − N D V I C C
where NDVICC refers to the NDVI predicted by the regression model, and NDVIobs represents the observed NDVI obtained from remote sensing images (dimensionless); a, b, and c are the model parameters; T and P denote the mean temperature (°C) and cumulative precipitation (mm), respectively; NDVIHA represents the residual value.

2.3.3. Evaluation and Influence of Vegetation NDVI Change

Determination of vegetation NDVI change drivers and their impacts: The linear trend rates of NDVICC and NDVIHA were calculated according to Equation (1) for the Loess Plateau from 1982 to 2015, reflecting the NDVI change trends under the influence of climate change and human activities, respectively [38]. A positive trend rate indicates that climate change or human activities have facilitated an increase in vegetation NDVI, thereby supporting vegetation recovery. Conversely, a negative trend rate suggests a decrease in vegetation NDVI, indicating an inhibitory effect on vegetation recovery. To refine the assessment of the impacts of climate change and human activities on vegetation growth, these effects were categorized into seven levels: ‘obviously inhibit’, ‘moderately inhibit’, ‘slightly inhibit’, ‘no effect’, ‘slightly promote’, ‘moderately promote’, and ‘obviously promote’, based on the trends of NDVICC and NDVIHA (Table 1) [1]. Furthermore, the primary drivers of NDVI changes during the growing season on the Loess Plateau were identified, and the relative contributions of climate change and human activities to NDVI changes during the growing season were quantified (Table 2) [14].

2.3.4. Partial Correlation Analysis

Partial correlation analysis is a geostatistical method that refines traditional correlation analysis by isolating the influence of a third variable [41,42]. This technique allows for the precise assessment of the correlation between two variables while controlling for their simultaneous association with a third variable [43]. This approach yields more accurate and reliable results than simple correlation analysis and is extensively utilized in research investigating the effects of climate change on vegetation growth. The calculation is conducted using the following formula:
R XY , Z = R XY − R XZ R YZ 1 − R XZ 2 1 − R YZ 2
where RXY,Z is the partial correlation coefficient between temperature and vegetation NDVI after controlling for precipitation; RXY is the correlation coefficient between temperature and vegetation NDVI; RXZ is the correlation coefficient between precipitation and vegetation NDVI; and RYZ is the correlation coefficient between temperature and precipitation. Similarly, the partial correlation coefficient between precipitation and vegetation NDVI can be calculated by controlling for temperature.
To facilitate interpretation of spatial patterns, the resulting partial correlation coefficients were further categorized based on the sign of the coefficient and its statistical significance (p-value < 0.05) as follows: Significant positive correlation (p < 0.05, r > 0), Non-significant positive correlation (p ≥ 0.05, r > 0), Significant negative correlation (p < 0.05, r < 0), Non-significant negative correlation (p ≥ 0.05, r < 0). This classification allows for a clear depiction of the direction and reliability of relationships across the study area and ensures that non-significant correlations are distinguished from statistically meaningful patterns.

3. Results

3.1. Characteristics of Spatial and Temporal Changes in Vegetation NDVI

NDVI, temperature (T), and precipitation (Pre) across the Loess Plateau exhibit similar spatial distribution patterns, all showing a gradually decreasing gradient from the southeast to the northwest (Figure 3). The annual NDVI of the Loess Plateau region showed a fluctuating but overall increasing trend from 1982 to 2015 (Figure 3b). Taking 1999 as a reference, the average NDVI trend rate was 0.9 × 10−3 a−1 from 1982 to 1999. Following the implementation of the GFG program, the trend rate significantly increased to 2.8 × 10−3 a−1 from 2000 to 2015. During the study period, precipitation initially declined and then rose, with trend rates of −2.77 mm/a before 1999 and 2.45 mm/a after 1999, respectively. Temperature trends were 0.06 °C/a before 1999 and 0.01 °C/a after 1999, indicating a slowdown in the warming trend after 1999. Comparing the periods 1982–1999 and 2000–2015, the proportion of areas with faster NDVI growth (slope ≥ 2 × 10−3 a−1) expanded from 13.31% to 62.88% of the Loess Plateau, representing an increase of nearly 50% (Figure 4). These areas are predominantly located in Shaanxi and Shanxi provinces. The GFG program has substantially contributed to vegetation recovery, highlighting the significant role of human activities.

3.2. Impacts of Climate Change and Human Activities on Vegetation Cover

Between 1982 and 1999, climate change correlated with NDVI growth across approximately 87.7% of the Loess Plateau region (Figure 5a). Within this area, 26.8% exhibited moderate promotion, predominantly in the southern Loess Plateau. In contrast, from 2000 to 2015, the influence of climate change on NDVI growth dropped to 45.7% (Figure 5b), with only 5.8% showing moderate promotion, primarily in the eastern Loess Plateau. This change indicates a significant reduction in the influence of climate change on NDVI growth after the year 2000.
Between 1982 and 1999, human activities had little impact on NDVI in 24.5% of the Loess Plateau (Figure 5c). During this period, human activities positively affected NDVI growth in 41.5% of the area, with moderate promotion in 11.4%. Conversely, NDVI growth was inhibited in 33.9% of the Loess Plateau. From 2000 to 2015, the impact of human activities on NDVI growth became more pronounced, positively affecting 88.4% of the Loess Plateau, with significant promotion in 60.5% of these areas (Figure 5d). These findings demonstrate a substantial increase in the positive impact of human activities on NDVI growth after 2000.

3.3. Spatial Distribution of Drivers of Vegetation Change

The distribution of the dominant impacts of anthropogenic activities and climate change exhibits significant spatial and temporal heterogeneity (Figure 6). From 1982 to 1999, NDVI increased in 87.4% of the study area, with 50.01% of this improvement attributed to the combined effects of human activities and climate change, primarily in the southwest and northeast of the Loess Plateau (Figure 6a). Climate change alone accounted for 34.72% of the improvement, mainly in Yan’an, Yulin, and Ordos, while human activities alone contributed 2.67%. Conversely, NDVI degradation was observed in only 12.6% of the study area, where 9.87% was mainly driven by anthropogenic activities, 0.6% by climate change, and 2.13% by their combined impact.
From 2000 to 2015, NDVI improved in 90.19% of the study area. Of this, 74.12% resulted from the combined impact of human activities and climate change, widely distributed across Qinghai, Gansu, Ningxia, Inner Mongolia, Shanxi, and northern Shaanxi provinces (Figure 6b). Human activities alone were responsible for 15.33% of the improvement, mainly in the southern Loess Plateau, while climate change alone accounted for just 0.74%. NDVI degradation was limited to 9.81% of the study area, with 3.52% primarily influenced by anthropogenic activities, 0.59% by climate change, and 5.69% by their combined influence.

3.4. Spatial Distribution of Drivers of Change in Different Vegetation Types

Since 2000, there have been notable shifts in the impact of human activities and climate change on various types of vegetation (Figure S1). Specifically, the area where human activities alone inhibited the increase in NDVI of forest vegetation decreased from 10.2% to 2.9%. Conversely, the area where human activities alone promoted NDVI growth in forests surged dramatically from 3.1% to 32.7%. In contrast, the area where climate change alone contributed to forest NDVI growth declined sharply from 37.51% to 0.16%. For farmland vegetation, the area impacted solely by human activities that facilitated NDVI growth expanded from 2.22% to 19.44%, while the area where climate change alone supported growth dropped from 30% to 0.6%, with a more substantial proportion of areas experiencing co-promotion of growth. Similar trends were observed in grassland vegetation, where the area of co-promoted growth increased markedly from 50% to 86.4%. The contribution of climate change alone to grassland vegetation growth decreased from 33.8% to 0.7%, alongside a slight rise in areas influenced by human activities. The extent of anthropogenic impacts on different vegetation types follows this order: forests > farmland > grasslands.

3.5. Contribution and Spatial Distribution of Different Drivers to Changes in Vegetation NDVI

Between 1982 and 1999, climate change positively influenced NDVI in approximately 87.5% of the Loess Plateau region (Figure 7a). Areas where the contribution of climate change exceeded 80% were more extensive, covering around 41.2% of the total area, primarily located in northern Shaanxi and Inner Mongolia. Regions where climate change contributed between 20% and 40%, and between 40% and 60%, accounted for 10% and 13.5% of the affected areas, respectively, and were mainly situated in Qinghai, Gansu, and Shanxi. Approximately 12.5% of the region experienced a negative contribution from climate change, concentrated in the central Loess Plateau.
From 1982 to 1999, human activities positively contributed to NDVI changes in approximately 64% of the Loess Plateau (Figure 7c). Areas where the contribution rate exceeded 80% constituted 16.7% of the total and were concentrated in central Loess Plateau. Conversely, regions with a negative contribution rate covered 35.3% of the area, with 28% experiencing a negative contribution rate above 20%, predominantly in Gansu, Inner Mongolia, and northern Shaanxi. Generally, climate change contributed more to NDVI increases across the Loess Plateau than human activities; however, in northern Loess Plateau, human activities notably suppressed NDVI growth.
Between 2000 and 2015, climate change had a positive impact on NDVI change in approximately 81% of the Loess Plateau (Figure 7b). The largest area, accounting for 68.4% of the total, had a contribution rate between 0% and 20%. Conversely, 19% of the region, primarily in southern Shaanxi, experienced a negative impact from climate change. However, human activities positively influenced NDVI in 98% of the Loess Plateau, with 87.2% of the area having a contribution rate over 80% (Figure 7d). These results suggest that human activities have become the primary driver of vegetation NDVI recovery across most of the Loess Plateau.
The contribution of climate change to the growth of all vegetation types declined markedly after 2000 (Figure S2). Areas where climate change contributed more than 80% to the growth of forests, croplands, and grasslands decreased substantially, from 52%, 41%, and 44% to 1.3%, 2.2%, and 1.4%, respectively. Meanwhile, areas with a contribution rate between 0% and 20% increased from 3.0%, 3.8%, and 5.1% to 50.9%, 66.4%, and 76.1%, respectively. Conversely, the contribution of human activities to vegetation growth increased significantly (Figure S3). Following the implementation of the GFG policy, areas where human activities contributed more than 80% to the growth of forests, croplands, and grasslands expanded from 16.4%, 14.7%, and 18.1% to 86.5%, 89.2%, and 86.7%, respectively. Additionally, the degree of suppression across all vegetation types was markedly reduced.

3.6. Sensitivities of NDVI Variations to Climate Related Variables

From 1982 to 2015, the correlations between precipitation, temperature, and NDVI changes on the Loess Plateau exhibited significant temporal and spatial heterogeneity. Temporally, the mean, maximum, and minimum partial correlation coefficients between NDVI changes and precipitation were recorded at 0.16/0.24, 0.77/0.84, and −0.65/−0.72 for the periods 1982–1999 and 2000–2015, respectively (Table 3). Spatially, from 1982 to 1999, 76.18% of the Loess Plateau exhibited a positive correlation between NDVI changes and precipitation, with 7.44% displaying a significant positive correlation (p < 0.05), predominantly in the western Ordos Plateau (Figure 8a). From 2000 to 2015, this positive correlation expanded to 84.66% of the Loess Plateau, with the proportion of significantly positive areas increasing to 11.42% (p < 0.05), concentrated in Zhongwei (Ningxia), eastern Yulin (Shaanxi), and near Taiyuan (Shanxi) (Figure 8b). From 1982 to 1999, 23.81% of the Loess Plateau exhibited a negative correlation between NDVI changes and precipitation, primarily in the southern Loess Plateau, with 0.62% of these areas showing a significant negative correlation (p < 0.05) (Figure 8a). This negative correlation decreased from 2000 to 2015, affecting 15.35% of the Loess Plateau, predominantly in southern Shaanxi, with only 0.43% demonstrating significant negative correlation (p < 0.05) (Figure 8b). Overall, the positive correlations between NDVI and precipitation have increased significantly over time.
Temporally, the mean, maximum, and minimum partial correlation coefficients between NDVI and temperature in the Loess Plateau were 0.24 and 0.14, 0.85 and 0.80, and −0.67 and −0.72 for the periods 1982–1999 and 2000–2015, respectively (Table 3). Spatially, 78.96% of the Loess Plateau exhibited a positive correlation between NDVI changes and temperature from 1982 to 1999 (Figure 8c). Of these, 20.71% displayed a significant positive correlation (p < 0.05), primarily concentrated in the southern Loess Plateau. From 2000 to 2015, 80.86% of the Loess Plateau exhibited a positive correlation between NDVI changes and temperature (Figure 8d), with 1.92% of the area showing a significant positive correlation (p < 0.05). In contrast, from 1982 to 1999, 21.05% of the Loess Plateau exhibited a negative correlation between NDVI changes and temperature, primarily in the northwestern region, with 0.24% showing a significant negative correlation (p < 0.05) (Figure 8c). From 2000 to 2015, the negative correlation decreased slightly, covering 19.14% of the Loess Plateau, with 0.16% of the area exhibiting a significant negative correlation (p < 0.05) (Figure 8d). Overall, the positive correlation between NDVI changes and temperature decreased significantly over time.

3.7. Land Use/Cover Changes During 1982–2015

LUCC on the Loess Plateau are primarily composed of grasslands, farmlands, and forests, which accounted for 48.86%/50.71%, 31.63%/28.73%, and 12.20%/13.75% of the total area in 2000 and 2015, respectively (Figure S4). Grasslands are primarily located in the northern Loess Plateau, where they occupy vast expanses. Farmlands are closely associated with human settlements, reflecting the anthropogenic influence on land use patterns. Forests are mainly distributed in the southeastern Loess Plateau, near the Qinling and Lvliang mountain ranges, indicating a concentration in areas conducive to forest growth.
From 2000 to 2015, the area of grasslands increased by 17,059.46 km2, forest land expanded by 10,105.78 km2, while non-vegetated areas decreased by 4861.72 km2, farmland reduced by 20,511.88 km2, and shrubland declined by 1791 km2 (Table 4). These changes encompass a total area that accounts for 15.2% of the Loess Plateau. The most significant changes were observed in cropland and grassland, comprising 43.6% and 36.1% of the altered area, respectively. The main land transitions included the conversion of farmland to grasslands (35.39%), farmland to non-vegetated areas (5.22%), and farmland to forests (3.02%). Other notable transitions were from grasslands to farmland (22.54%), grasslands to forests (8.23%), and from non-vegetated areas to grasslands (12.05%).
The vegetation changes across the provinces of the Loess Plateau show significant differences (Figure 9). The most significant decreases in farmland occurred in Shaanxi (6389 km2), followed by Gansu (5490 km2), Shanxi (4075 km2), and Ningxia (2726 km2). The largest increases in forest land were observed in Shanxi (4654 km2), followed by Shaanxi (3767 km2) and Gansu (1258 km2). Grassland areas expanded most notably in Inner Mongolia (5562 km2), followed by Gansu (4163 km2), Shaanxi (3866 km2), Ningxia (3470 km2), and Qinghai (1119 km2). In Inner Mongolia, the increase in grasslands was primarily due to the conversion of non-vegetated areas. In Gansu, grassland expansion surpassed forest growth, mainly driven by the conversion of farmland. In Shanxi, the primary increase was in forests, resulting from the conversion of farmland and some shrub. Shaanxi experienced nearly equal increases in forests and grasslands, both primarily converted from farmland. In Ningxia, grassland growth was predominant, largely due to the conversion of farmland.

4. Discussion

4.1. Trend of NDVI

This study examined the trends in vegetation NDVI and the contributions of climate change and human activities to vegetation changes in the Loess Plateau from 1982 to 2015. The analysis employed trend analysis and multiple regression residuals to assess the spatial and temporal variability of vegetation changes on the Loess Plateau. The increasing trend of NDVI from 2000 to 2015 was nearly three times higher than that from 1982 to 1999 (Figure 3). The implementation of the GFG policy significantly expanded vegetation cover, leading to a marked increase in the annual NDVI mean value. Vegetation recovery was observed across most areas, with human activities identified as the primary driver of NDVI change, aligning with previous studies [14,44,45]. Moreover, advancements in planting technology have enhanced vegetation vigor in cultivated areas [29]. Human activities have emerged as the dominant factor driving NDVI growth in the Loess Plateau. Concurrently, the intensification of global warming since the 1980s has resulted in rising temperatures, advancing the vegetation growth cycle and extending its duration, thereby promoting vegetation growth and increasing NDVI [20,46].
Significant spatial variations in NDVI were observed across the Loess Plateau. Notably, the central-eastern regions, including northern Shaanxi, Yulin, and the Luliang Mountains, exhibited significantly higher NDVI increases than other areas (Figure 4). The Loess Plateau is characterized by an elevation gradient, with higher elevations in the west and lower in the east [47]. Previous research has demonstrated that NDVI variations correlate with elevation, with lower elevation areas being more conducive to NDVI growth [48,49]. Consequently, altitude variation significantly contributes to the spatial heterogeneity of NDVI across the Loess Plateau. Additionally, the uneven distribution of precipitation across the Loess Plateau further influences vegetation cover [27,50]. Precipitation patterns in the Loess Plateau are characterized by higher amounts in the south than the north, and more in the east than the west, generally decreasing from southeast to northwest. The annual precipitation contours orient from northeast to southwest [28]. Lower precipitation levels contribute to slower NDVI recovery in the western Loess Plateau, while more pronounced vegetation recovery along the Yellow River is supported by erosion control efforts in the Yellow River basin [51,52,53].

4.2. Influence Factors of NDVI

4.2.1. Human Activity

After 2000, the influence of human activities on NDVI growth in the Loess Plateau increased significantly, while the impact of climate change on NDVI diminished (Figure 5). This shift indicates that human activities have emerged as the primary driver of NDVI growth in parts of the Loess Plateau after 2000. This phenomenon can be attributed to two main factors: firstly, the enhancement of vegetation cover due to the implementation of the GFG program and land desertification control measures [54,55]. Secondly, changes in land use patterns, shifts in agricultural production methods, and the extensive use of chemical fertilizers, all of which have significantly enhanced vegetation activity [51]. Excessive human interference has also indirectly reduced the sensitivity of vegetation cover to climatic factors, resulting in a lower correlation between vegetation cover and climatic variables [56,57]. Consequently, the impact of climate change on NDVI changes diminished after 2000. However, in and around densely populated urban centers such as Xi’an and Luoyang, human-induced disturbances remain a prominent factor limiting vegetation regeneration. This persistent inhibitory effect is likely associated with high land-use intensity, urban expansion, and infrastructure development, which can offset the benefits of ecological restoration efforts [58].
Of all vegetation types impacted by human activities, forests underwent the most pronounced transformation in the period surrounding the year 2000 (Figure S3) [59]. Since 2000, the area where human activities alone contributed to forest growth increased from 3.1% to 32.7%. Moreover, the regions where human activities contributed more than 80% to forest growth expanded substantially. These changes can be attributed to the implementation of ecological and environmental projects and increased environmental awareness among the population [60]. The adoption of sustainable development strategies, advancements in production methods, and the development of clean energy sources have decreased the utilization rate of forest resources, thereby contributing to the growth of forest vegetation [45,60]. Although the GFG program has reduced some cultivated land, advancements in agricultural technology and improvements in saline land have significantly enhanced the vitality of farmland vegetation. Consequently, human activities continue to exert a strong positive influence on arable land vegetation [21,57]. While human activities have also positively impacted grasslands, their effect is relatively less pronounced [59]. Grasslands, which occupy nearly half of the vegetation area of the Loess Plateau, are predominantly located far from human settlements. Grasslands, which occupy nearly half of the vegetation area of the Loess Plateau, are mainly distributed in areas far from human settlements. In recent years, the implementation of ecological restoration projects and grassland management policies (e.g., artificial seeding, enclosure protection, and controlled grazing) has significantly reduced grazing pressure, thereby promoting the recovery of grassland vegetation. Consequently, the growth of these grasslands is influenced by both climate change and human management interventions. The contribution of climate change to all vegetation types shows a decreasing trend, likely due to changes in thermal and hydrological conditions associated with global warming [59,61].
Over the past 30 years, land use and land cover changes (LUCC) on the Loess Plateau have been characterized by the shrinkage of farmland and the expansion of forests and grasslands (Table 4). This trend aligns with the expectations of the GFG policy. Grasslands experienced the largest increase, expanding by 17,059.46 km2, followed by forest land, which grew by 10,105.78 km2. Farmland was primarily converted to forests in southern Shaanxi and Shanxi [30], while conversion to grasslands occurred mainly in Inner Mongolia, Gansu, Ningxia, and northern Shaanxi (Figure 9 and Figure S4). Additionally, some grasslands were transformed into forests in the southeastern Loess Plateau, consistent with existing studies [30]. The distribution of LUCC on the Loess Plateau reflects regional vegetation patterns, with forests and farmland predominating in the southeastern region and grasslands dominating the central and northwestern areas [3]. This distribution is influenced by the uneven spatial distribution of precipitation, with the southeast receiving ample rainfall, creating favorable conditions for forest growth [50]. This observation suggests that grasslands can be transformed into forest land under suitable climatic conditions [62]. In contrast, the arid northwestern region, with its limited rainfall, remains dominated by grasslands [3]. These findings highlight that the GFG policy has been effectively tailored to the local ecological environment, achieving significant ecological restoration outcomes [30,63].
After 2000, NDVI values on the Loess Plateau increased markedly, which we attribute largely to the implementation of the Grain for Green (GFG) program. This national ecological restoration initiative, launched in 1999, included several large-scale management practices [43]. Cropland on steep slopes was retired and converted into grassland or forest to reduce soil erosion. Extensive afforestation projects, including both natural regeneration and artificial tree planting, significantly expanded forest cover [64]. Grassland restoration and enclosure measures were also implemented to reduce grazing pressure and promote recovery. In addition, soil and water conservation measures such as terracing and check-dams further improved soil conditions [13]. These interventions collectively created favorable ecological conditions that explain the pronounced post-2000 NDVI increase observed in our study, consistent with previous findings. However, recent studies indicate that large-scale vegetation restoration may have unintended hydrological consequences. For instance, afforestation has been shown to increase evapotranspiration, reduce deep soil moisture, and decrease streamflow in semi-arid regions of the Loess Plateau [65,66]. Similarly, analyses using hydrological models suggest that the GFG project has led to localized reductions in water yield, particularly in areas with limited precipitation, potentially exacerbating water stress for agricultural and urban use [67,68]. Moreover, the effects of restoration on water resources are spatially heterogeneous, with some regions experiencing more pronounced declines in groundwater recharge and soil moisture than others [69]. These findings underscore the necessity of integrating water resource management into ecological restoration planning and highlight the importance of future research to quantify the hydrological impacts of large-scale vegetation recovery at regional scales.

4.2.2. Climate Change

The correlation between NDVI changes and precipitation on the Loess Plateau increased significantly after 2000, driven primarily by shifts in precipitation patterns (Table 3). Specifically, regional precipitation trends reversed from a declining pattern (1982–1999) to a positive trend (2000–2015) (Figure 3), providing enhanced soil moisture necessary for vegetation growth. This increase in rainfall availability significantly promoted NDVI growth, in line with previous findings in the region [28,30,55]. Conversely, the correlation between temperature and NDVI weakened over time. Between 1982 and 1999, rising temperatures accelerated phenological development and lengthened the growing season, resulting in positive NDVI responses. However, after 2000, the rate of temperature increase stabilized, reducing its marginal benefit to vegetation growth [46,70].
Spatially, NDVI-precipitation correlations were strongest in the arid northwestern regions of the plateau (Figure 8a), where vegetation is most water-limited. These regions exhibited enhanced NDVI sensitivity due to increased precipitation post-2000, especially in Shanxi (Figure 8b), as precipitation shifts favored formerly drier areas [57].
In contrast, temperature increases led to a spatially heterogeneous NDVI response: negative correlations appeared in the northwest due to increased evapotranspiration and water stress, while positive correlations persisted in the more humid southeastern regions (Figure 8c,d). Increased temperatures will enhance vegetation transpiration, leading to water loss [71]. These spatial trends highlight that precipitation, rather than temperature, has become the dominant climatic driver of vegetation dynamics on the Loess Plateau [25,30,71].
Beyond the temporal and spatial patterns observed, the shifting dominance from temperature to precipitation highlights the critical role of water availability in mediating ecosystem resilience under climate change. The Loess Plateau is characterized by fragile soil structures and limited water-holding capacity; thus, vegetation recovery largely depends on consistent precipitation inputs rather than incremental warming [22]. The strengthened NDVI–precipitation relationship after 2000 also suggests that ecological restoration projects, such as the Grain for Green Program, may have amplified vegetation sensitivity to rainfall by introducing species with higher water demands and greater reliance on soil moisture [30,69]. This synergistic effect between human-induced land cover change and climatic precipitation shifts underscores the need to consider coupled natural–human drivers when interpreting vegetation dynamics [57]. Moreover, while precipitation emerges as the dominant factor, the interaction with temperature should not be neglected, as warming-induced evapotranspiration could offset the positive impacts of increased rainfall, particularly under future climate scenarios with intensified drought risk [9].

4.3. Limitations

This study analyzed vegetation changes and their driving mechanisms on the Loess Plateau based on NDVI data from 1982 to 2015. However, several limitations remain. First, NDVI tends to saturate in areas with dense vegetation, potentially underestimating the effects of ecological restoration. Second, the spatial resolution of the NDVI data is relatively coarse, which may obscure fine-scale heterogeneity in vegetation dynamics. Third, the selection of climatic variables was limited, excluding factors such as solar radiation, which may constrain a comprehensive understanding of climate-driven impacts. Fourth, key anthropogenic drivers, including urban expansion and advances in agricultural technology, were not quantitatively assessed, limiting the accuracy of attribution. Fifth this study did not account for the cumulative and lagged effects of climate change, which may delay or amplify vegetation responses and thus bias the attribution results. Moreover, the study period does not capture vegetation responses under recent intensification of ecological policies. Future research should incorporate multi-source datasets, improve spatial resolution, and extend the temporal scope to enhance attribution accuracy and monitor ongoing ecosystem dynamics.

5. Conclusions

This study reveals that from 1982 to 2015, vegetation cover in the Loess Plateau exhibited a significant greening trend, particularly after 2000, driven primarily by human activities such as the GFG program. Quantitatively, taking 1999 as a reference, the average NDVI trend rate was 0.9 × 10−3 a−1 from 1982 to 1999; while following the implementation of the GFG program, the trend rate significantly increased to 2.8 × 10−3 a−1 from 2000 to 2015. While climate change initially contributed to NDVI increases, its influence weakened after policy implementation, as land use changes—especially afforestation and cropland abandonment—became dominant. Forests responded most strongly to human interventions, followed by croplands and grasslands. Spatial patterns highlight the effectiveness of restoration in low-elevation and semi-arid areas. The expansion of forest and grassland at the expense of farmland and barren land underscores the ecological success of the GFG policy. These findings provide important insights into sustainable land management in ecologically fragile regions worldwide. Moving forward, integrated strategies incorporating long-term monitoring, biodiversity assessment, and climate adaptation will be essential to sustain ecological gains and enhance ecosystem resilience.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/su17188233/s1: Figure S1: Proportion of area affected by different driving factors for different vegetation types; Figure S2: Area proportion of different vegetation types contributing to climate change; Figure S3: Area proportion of different vegetation types contributing to human activities; Figure S4: Land Use/Cover Changes of Loess Plateau (a): 2000 (b): 2015.

Author Contributions

J.L.: Writing—original draft, Formal analysis; H.L.: Funding acquisition, Methodology; D.C.: Conceptualization, Supervision, Writing—review & editing; H.Z.: Visualization, Writing—review & editing; G.Q.: Data curation, Software; W.W.: Project administration, Writing—review & editing. All authors have read and agreed to the published version of the manuscript.

Funding

This study was jointly supported by the National Natural Science Foundation of China (Grant Nos. 52309035, 42471028), the State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water Resources and Hydropower Research (IWHR-SKL-KF202401), and the NWU Tang Scholar. We are especially grateful to the Editor, Associate Editor, and anonymous reviewers for their helpful comments and suggestions, which have improved the quality of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

All data generated or analyzed during this study are included in this published article.

Conflicts of Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

  1. Jin, K.; Wang, F.; Han, J.; Shi, S.; Ding, W. Contribution of climatic change and human activities to vegetation NDVI change over China during 1982–2015. Acta Geogr. Sin. 2020, 75, 961–974. [Google Scholar]
  2. Fang, J.Y.; Song, Y.C.; Liu, H.Y.; Piao, S.L. Vegetation-climate relationship and its application in the division of vegetation zone in China. Acta Bot. Sin. 2002, 44, 1105–1122. [Google Scholar]
  3. Wang, X.Z.; Wu, J.Z.; Liu, Y.L.; Hai, X.Y.; Shanguan, Z.P.; Deng, L. Driving factors of ecosystem services and their spatiotemporal change assessment based on land use types in the Loess Plateau. J. Environ. Manag. 2022, 311, 114835. [Google Scholar] [CrossRef]
  4. Chen, C.; Park, T.; Wang, X.H.; Piao, S.L.; Xu, B.D.; Chaturvedi, R.K.; Fuchs, R.; Brovkin, V.; Ciais, P.; Fensholt, R.; et al. China and India lead in greening of the world through land-use management. Nat. Sustain. 2019, 2, 122–129. [Google Scholar] [CrossRef]
  5. Li, P.; Wang, J.; Liu, M.; Xue, Z.; Bagherzadeh, A.; Liu, M. Spatio-temporal variation characteristics of NDVI and its response to climate on the Loess Plateau from 1985 to 2015. Catena 2021, 203, 105331. [Google Scholar] [CrossRef]
  6. Li, G.; Sun, S.; Han, J.; Yan, J.; Liu, W.; Wei, Y.; Lu, N.; Sun, Y. Impacts of Chinese Grain for Green program and climate change on vegetation in the Loess Plateau during 1982–2015. Sci. Total Environ. 2019, 660, 177–187. [Google Scholar] [CrossRef] [PubMed]
  7. Song, F.; Kang, M.; Yang, P.; Chen, Y.; Liu, Y.; Xing, K. Comparisonv and validation of GIMMS, SPOT-VGT and MODIS global NDVI products in the Loess Plateau of northern Shaanxi Province, northwestern China. J. Beijing For. Univ. 2010, 32, 72–80. [Google Scholar]
  8. Meng, Y.Y.; Hou, B.W.; Ding, C.; Huang, L.; Guo, Y.P.; Tang, Z.Y. Spatiotemporal patterns of planted forests on the Loess Plateau between 1986 and 2021 based on Landsat NDVI time-series analysis. GISci. Remote Sens. 2023, 60, 2185980. [Google Scholar] [CrossRef]
  9. 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]
  10. Yan, K.; Gao, S.; Yan, G.; Ma, X.; Chen, X.; Zhu, P.; Li, J.; Gao, S.; Gastellu-Etchegorry, J.-P.; Myneni, R.B.; et al. A global systematic review of the remote sensing vegetation indices. Int. J. Appl. Earth Obs. Geoinf. 2025, 139, 104560. [Google Scholar] [CrossRef]
  11. Nwilo, P.; Okolie, C.; Umar, A.; Akinnusi, S.; Ojegbile, B.; Olanrewaju, H. Spatial relationship between NDVI, EVI, SAVI and land cover change in the Lake Chad area from 1987 to 2017. In Proceedings of the 3rd Intercontinental Geoinformation Days (IGD), Mersin, Turkey, 17–18 November 2021. [Google Scholar]
  12. Li, Q.; Cheng, J.; Yan, J.; Zhang, G.; Ling, H. Comparison of Satellite-Derived Vegetation Indices for Assessing Vegetation Dynamics in Central Asia. Water 2025, 17, 684. [Google Scholar] [CrossRef]
  13. Huang, Y.; Jin, Y.; Chen, S. The Spatiotemporal Dynamics of Vegetation Cover and Its Response to the Grain for Green Project in the Loess Plateau of China. Forests 2024, 15, 1949. [Google Scholar] [CrossRef]
  14. Sun, W.; Song, X.; Mu, X.; Gao, P.; Wang, F.; Zhao, G. Spatiotemporal vegetation cover variations associated with climate change and ecological restoration in the Loess Plateau. Agric. For. Meteorol. 2015, 209, 87–99. [Google Scholar] [CrossRef]
  15. Wen, Y.; Liu, X.; Bai, Y.; Sun, Y.; Yang, J.; Lin, K.; Pei, F.; Yan, Y. Determining the impacts of climate change and urban expansion on terrestrial net primary production in China. J. Environ. Manag. 2019, 240, 75–83. [Google Scholar] [CrossRef] [PubMed]
  16. Wen, Y.; Yang, J.; Liao, W.; Xiao, J.; Yan, S. Refined assessment of space-time changes, influencing factors and socio-economic impacts of the terrestrial ecosystem quality: A case study of the GBA. J. Environ. Manag. 2023, 345, 118869. [Google Scholar] [CrossRef]
  17. Wang, X.; Wang, B.; Xu, X.; Liu, T.; Duan, Y.; Zhao, Y. Spatial and temporal variations in surface soil moisture and vegetation cover in the Loess Plateau from 2000 to 2015. Ecol. Indic. 2018, 95, 320–330. [Google Scholar] [CrossRef]
  18. Ren, Z.G.; Tian, Z.H.; Wei, H.T.; Liu, Y.; Yu, Y.P. Spatiotemporal evolution and driving mechanisms of vegetation in the Yellow River Basin, China during 2000–2020. Ecol. Indic. 2022, 138, 108832. [Google Scholar] [CrossRef]
  19. Zhou, L.M.; Tucker, C.J.; Kaufmann, R.K.; Slayback, D.; Shabanov, N.V.; Myneni, R.B. Variations in northern vegetation activity inferred from satellite data of vegetation index during 1981 to 1999. J. Geophys. Res. Atmos. 2001, 106, 20069–20083. [Google Scholar] [CrossRef]
  20. Yang, X.; Yang, T.; Liu, H.; Ghebrezgabher Mihretab, G.; Wang, Q.; Wei, H. Vegetation Variation in the North Hemisphere under Climate Warming in the Last 30 Years. Arid Zone Res. 2016, 33, 379–391. [Google Scholar]
  21. Cao, S.P.; Zhang, L.F.; He, Y.; Zhang, Y.L.; Chen, Y.; Yao, S.; Yang, W.; Sun, Q. Effects and contributions of meteorological drought on agricultural drought under different climatic zones and vegetation types in Northwest China. Sci. Total Environ. 2022, 821, 153270. [Google Scholar] [CrossRef]
  22. Du, G.; Yan, S.; Chen, H.; Yang, J.; Wen, Y. Intra-Annual Cumulative Effects and Mechanisms of Climatic Factors on Global Vegetation Biomes’ Growth. Remote Sens. 2024, 16, 779. [Google Scholar] [CrossRef]
  23. Wang, J.; Lin, Y.; Glendinning, A.; Xu, Y. Land-use changes and land policies evolution in China’s urbanization processes. Land Use Policy 2018, 75, 375–387. [Google Scholar] [CrossRef]
  24. Yang, Y.; Zhang, P.; Wu, F.; Zhou, Y.; Song, Y.; Wang, Y.; An, S. The significance and countermeasures of vegetation construction on the Loess Plateau to carbon neutrality. Acta Ecol. Sin. 2023, 43, 9071–9081. [Google Scholar]
  25. Zhang, Y.C.; Jiang, X.H.; Lei, Y.X.; Gao, S.Q. The contributions of natural and anthropogenic factors to NDVI variations on the Loess Plateau in China during 2000–2020. Ecol. Indic. 2022, 143, 109342. [Google Scholar] [CrossRef]
  26. Naeem, S.; Zhang, Y.Q.; Zhang, X.Z.; Tian, J.; Abbas, S.; Luo, L.L.; Meresa, H.K. Both climate and socioeconomic drivers contribute to vegetation greening of the Loess Plateau. Sci. Bull. 2021, 66, 1160–1163. [Google Scholar] [CrossRef] [PubMed]
  27. Zhang, B.Q.; Tian, L.; Yang, Y.T.; He, X.G. Revegetation Does Not Decrease Water Yield in the Loess Plateau of China. Geophys. Res. Lett. 2022, 49, e2022GL098025. [Google Scholar] [CrossRef]
  28. Xin, Z.; Xu, J.; Zheng, W. Spatiotemporal variations of vegetation cover on the Chinese Loess Plateau (1981–2006): Impacts of climate changes and human activities. Sci. China Ser. D Earth Sci. 2008, 51, 67–78. [Google Scholar] [CrossRef]
  29. Wang, J.; Chen, Y.; Shao, X.; Zhang, Y.; Cao, Y. Land-use changes and policy dimension driving forces in China: Present, trend and future. Land Use Policy 2012, 29, 737–749. [Google Scholar] [CrossRef]
  30. Chen, S.F.; Zhang, Q.F.; Chen, Y.N.; Zhou, H.H.; Xiang, Y.Y.; Liu, Z.H.; Hou, Y.F. Vegetation Change and Eco-Environmental Quality Evaluation in the Loess Plateau of China from 2000 to 2020. Remote Sens. 2023, 15, 424. [Google Scholar] [CrossRef]
  31. Kong, D.; Miao, C.; Wu, J.; Zheng, H.; Wu, S. Time lag of vegetation growth on the Loess Plateau in response to climate factors: Estimation, distribution, and influence. Sci. Total Environ. 2020, 744, 140726. [Google Scholar] [CrossRef] [PubMed]
  32. Li, X.; Xu, J.; Jia, Y.; Liu, S.; Jiang, Y.; Yuan, Z.; Du, H.; Han, R.; Ye, Y. Spatio-temporal dynamics of vegetation over cloudy areas in Southwest China retrieved from four NDVI products. Ecol. Inform. 2024, 81, 102630. [Google Scholar] [CrossRef]
  33. Yang, J.; Huang, X. The 30 m annual land cover dataset and its dynamics in China from 1990 to 2019. Earth Syst. Sci. Data 2021, 13, 3907–3925. [Google Scholar] [CrossRef]
  34. Baiming, C.; Xiaoping, Z. Explanation of Current Land National Standard of the Use Condition Classification for People’s Republic of China. J. Nat. Resour. 2007, 22, 994–1003. [Google Scholar]
  35. He, W.; Xiao, Z.; Lu, Q.; Wei, L.; Liu, X. Digital Mapping of Soil Particle Size Fractions in the Loess Plateau, China, Using Environmental Variables and Multivariate Random Forest. Remote Sens. 2024, 16, 785. [Google Scholar] [CrossRef]
  36. Wang, X.; He, L.; Li, P.; Ma, J.; Shi, Y.; Tian, Q.; Zhao, G.; He, J.; Feng, H.; Shi, H.; et al. Spatially adaptive estimation of multi-layer soil temperature at a daily time-step across China during 2010–2020. Earth Syst. Sci. Data Discuss. 2025; in review. [Google Scholar] [CrossRef]
  37. Nie, T.; Dong, G.; Jiang, X.; Lei, Y. Spatio-Temporal Changes and Driving Forces of Vegetation Coverage on the Loess Plateau of Northern Shaanxi. Remote Sens. 2021, 13, 613. [Google Scholar] [CrossRef]
  38. Gao, W.; Zheng, C.; Liu, X.; Lu, Y.; Chen, Y.; Wei, Y.; Ma, Y. NDVI-based vegetation dynamics and their responses to climate change and human activities from 1982 to 2020: A case study in the Mu Us Sandy Land, China. Ecol. Indic. 2022, 137, 108745. [Google Scholar] [CrossRef]
  39. Jiang, W.; Yuan, L.; Wang, W.; Cao, R.; Zhang, Y.; Shen, W. Spatio-temporal analysis of vegetation variation in the Yellow River Basin. Ecol. Indic. 2015, 51, 117–126. [Google Scholar] [CrossRef]
  40. Piao, S.; Yin, G.; Tan, J.; Cheng, L.; Huang, M.; Li, Y.; Liu, R.; Mao, J.; Myneni, R.B.; Peng, S.; et al. Detection and attribution of vegetation greening trend in China over the last 30 years. Glob. Change Biol. 2015, 21, 1601–1609. [Google Scholar] [CrossRef]
  41. Lin, M.; Hou, L.; Qi, Z.; Wan, L. Impacts of climate change and human activities on vegetation NDVI in China’s Mu Us Sandy Land during 2000–2019. Ecol. Indic. 2022, 142, 109164. [Google Scholar] [CrossRef]
  42. Sun, Y.-L.; Shan, M.; Pei, X.-R.; Zhang, X.-K.; Yang, Y.-L. Assessment of the impacts of climate change and human activities on vegetation cover change in the Haihe River basin, China. Phys. Chem. Earth Parts A/B/C 2020, 115, 102834. [Google Scholar] [CrossRef]
  43. Yin, X.; Cao, D. Spatiotemporal variation and driving mechanisms of vegetation net primary productivity in Hunan Province from 2001 to 2023. Sci. Remote Sens. 2025, 12, 100269. [Google Scholar] [CrossRef]
  44. Hu, R.; Fang, F.; Salinas, P.; Pain, C.C.; Domingo, N.S.; Mark, O. Numerical simulation of floods from multiple sources using an adaptive anisotropic unstructured mesh method. Adv. Water Resour. 2019, 123, 173–188. [Google Scholar] [CrossRef]
  45. Zhang, B.; Wu, P.; Zhao, X.; Wang, Y.; Gao, X. Changes in vegetation condition in areas with different gradients (1980–2010) on the Loess Plateau, China. Environ. Earth Sci. 2012, 68, 2427–2438. [Google Scholar] [CrossRef]
  46. Piao, S.; Friedlingstein, P.; Ciais, P.; Zhou, L.; Chen, A. Effect of climate and CO2 changes on the greening of the Northern Hemisphere over the past two decades. Geophys. Res. Lett. 2006, 33, L23402. [Google Scholar] [CrossRef]
  47. Li, J.; Peng, S.; Li, Z. Detecting and attributing vegetation changes on China’s Loess Plateau. Agric. For. Meteorol. 2017, 247, 260–270. [Google Scholar] [CrossRef]
  48. Jin, K.; Wang, F.; Li, P. Responses of Vegetation Cover to Environmental Change in Large Cities of China. Sustainability 2018, 10, 270. [Google Scholar] [CrossRef]
  49. Deng, C.; Bai, H.; Gao, S.; Liu, R.; Ma, X.; Huang, X.; Meng, Q. Spatial-temporal Variation of the Vegetation Coverage in Qinling Mountains and Its Dual Response to Climate Change and Human Activities. J. Nat. Resour. 2018, 33, 425–438. [Google Scholar]
  50. Jin, F.; Yang, W.; Fu, J.; Li, Z. Effects of vegetation and climate on the changes of soil erosion in the Loess Plateau of China. Sci. Total Environ. 2021, 773, 145514. [Google Scholar] [CrossRef]
  51. Piao, S.; Ciais, P.; Huang, Y.; Shen, Z.; Peng, S.; Li, J.; Zhou, L.; Liu, H.; Ma, Y.; Ding, Y.; et al. The impacts of climate change on water resources and agriculture in China. Nature 2010, 467, 43–51. [Google Scholar] [CrossRef]
  52. Evans, J.; Geerken, R. Discrimination between climate and human-induced dryland degradation. J. Arid Environ. 2004, 57, 535–554. [Google Scholar] [CrossRef]
  53. Zhao, G.; Mu, X.; Wen, Z.; Wang, F.; Gao, P. Soil Erosion, Conservation, and Eco-Environment Changes in the Loess Plateau of China. Land Degrad. Dev. 2013, 24, 499–510. [Google Scholar] [CrossRef]
  54. Ding, Y.B.; Wang, F.Z.; Mu, Q.; Sun, Y.A.; Cai, H.J.; Zhou, Z.Q.; Xu, J.T.; Shi, H.Y. Estimating land use/land cover change impacts on vegetation response to drought under ‘Grain for Green’ in the Loess Plateau. Land Degrad. Dev. 2021, 32, 5083–5098. [Google Scholar] [CrossRef]
  55. Zhao, A.; Zhang, A.; Lu, C.; Wang, D.; Wang, H.; Liu, H. Spatiotemporal variation of vegetation coverage before and after implementation of Grain for Green Program in Loess Plateau, China. Ecol. Eng. 2017, 104, 13–22. [Google Scholar] [CrossRef]
  56. Zhang, B.; Wu, P.; Zhao, X. Detecting and analysis of spatial and temporal variation of vegetation cover in the Loess Plateau during 1982–2009. Trans. Chin. Soc. Agric. Eng. 2011, 27, 287–293. [Google Scholar]
  57. Mao, R.; Xing, L.; Wu, Q.; Song, J.; Li, Q.; Long, Y.; Shi, Y.; Huang, P.; Zhang, Q. Evaluating net primary productivity dynamics and their response to land-use change in the loess plateau after the ‘Grain for Green’ program. J. Environ. Manag. 2024, 360, 121112. [Google Scholar] [CrossRef]
  58. Kou, P.; Xu, Q.; Jin, Z.; Yunus, A.P.; Luo, X.; Liu, M. Complex anthropogenic interaction on vegetation greening in the Chinese Loess Plateau. Sci. Total Environ. 2021, 778, 146065. [Google Scholar] [CrossRef]
  59. Li, G.; Sun, S.; Lu, N.; Huang, R.; Yan, J.; Song, F.; Han, J.; Wang, Y. Changes in soil organic carbon stocks of forestlands and grasslands on the Loess Plateau, 1980–2015. J. Clean. Prod. 2023, 428, 139463. [Google Scholar] [CrossRef]
  60. Fu, B.J.; Wang, S.; Liu, Y.; Liu, J.B.; Liang, W.; Miao, C.Y. Hydrogeomorphic Ecosystem Responses to Natural and Anthropogenic Changes in the Loess Plateau of China. Annu. Rev. Earth Planet. Sci. 2017, 45, 223–243. [Google Scholar] [CrossRef]
  61. Su, B.; Wang, Y.; Shangguan, Z. Analysis on the Scale of A New Period of Returning Farmland to Forestland and Grassland in Northwest China. Res. Soil Water Conserv. 2017, 24, 59–65. [Google Scholar]
  62. Zheng, K.; Wei, J.Z.; Pei, J.Y.; Cheng, H.; Zhang, X.L.; Huang, F.Q.; Li, F.M.; Ye, J.S. Impacts of climate change and human activities on grassland vegetation variation in the Chinese Loess Plateau. Sci. Total Environ. 2019, 660, 236–244. [Google Scholar] [CrossRef] [PubMed]
  63. Yan, Y.; Wu, C.; Wen, Y. Determining the impacts of climate change and urban expansion on net primary productivity using the spatio-temporal fusion of remote sensing data. Ecol. Indic. 2021, 127, 107737. [Google Scholar] [CrossRef]
  64. Feng, X.M.; Sun, G.; Fu, B.J.; Su, C.H.; Liu, Y.; Lamparski, H. Regional effects of vegetation restoration on water yield across the Loess Plateau, China. Hydrol. Earth Syst. Sci. 2012, 16, 2617–2628. [Google Scholar] [CrossRef]
  65. Ge, F.; Xu, M.; Li, B.; Gong, C.; Zhang, J. Afforestation reduced the deep profile soil water sustainability on the semiarid Loess Plateau. For. Ecol. Manag. 2023, 544, 121240. [Google Scholar] [CrossRef]
  66. Liu, Y.; Kong, C.; Zhang, Y.; Liu, G.; Huang, J.; Li, G.; Du, S. Monitoring and evaluation of the effects of Grain for Green Project on the Loess Plateau: A case study of Wuqi County in China. Int. J. Appl. Earth Obs. Geoinf. 2024, 132, 104006. [Google Scholar] [CrossRef]
  67. Jia, X.; Shao, M.a.; Zhu, Y.; Luo, Y. Soil moisture decline due to afforestation across the Loess Plateau, China. J. Hydrol. 2017, 546, 113–122. [Google Scholar] [CrossRef]
  68. Zhao, H.; Dong, J.; Yang, Y.; Zhao, J.; He, J.; Yue, C. Vegetation Restoration Increases the Drought Risk on the Loess Plateau. Plants 2024, 13, 2735. [Google Scholar] [CrossRef]
  69. Zhang, Y.C.; Piao, S.L.; Sun, Y.; Rogers, B.M.; Li, X.Y.; Lian, X.; Liu, Z.H.; Chen, A.P.; Peñuelas, J. Future reversal of warming-enhanced vegetation productivity in the Northern Hemisphere. Nat. Clim. Chang. 2022, 12, 581–586. [Google Scholar] [CrossRef]
  70. Guo, W.; Huang, S.; Huang, Q.; She, D.; Shi, H.; Leng, G.; Li, J.; Cheng, L.; Gao, Y.; Peng, J. Precipitation and vegetation transpiration variations dominate the dynamics of agricultural drought characteristics in China. Sci. Total Environ. 2023, 898, 165480. [Google Scholar] [CrossRef]
  71. Zhang, B.; Liu, C.; Wang, X. Spatio-temporal changes of vegetation coverage in the Loess Plateau of northern Shaanxi and its attribution analysis. Bull. Surv. Mapp. 2022, 8, 22. [Google Scholar]
Sustainability 17 08233 g001
Figure 1. Technology roadmap.
Figure 1. Technology roadmap.
Sustainability 17 08233 g001
Sustainability 17 08233 g002
Figure 2. Location of the Loess Plateau.
Figure 2. Location of the Loess Plateau.
Sustainability 17 08233 g002
Sustainability 17 08233 g003
Figure 3. (a) NDVI, (b) Temperature, and (c) Precipitation spatial distribution characteristics, (d) temporal distribution characteristics of NDVI, temperature, and precipitation in the Loess Plateau from 1982 to 2015.
Figure 3. (a) NDVI, (b) Temperature, and (c) Precipitation spatial distribution characteristics, (d) temporal distribution characteristics of NDVI, temperature, and precipitation in the Loess Plateau from 1982 to 2015.
Sustainability 17 08233 g003
Sustainability 17 08233 g004
Figure 4. NDVI trends on the Loess Plateau, (a) 1982–1999; (b) 2000–2015.
Figure 4. NDVI trends on the Loess Plateau, (a) 1982–1999; (b) 2000–2015.
Sustainability 17 08233 g004
Sustainability 17 08233 g005
Figure 5. Impacts of influences factors on NDVI change, (a,c) for climate change and human activities on NDVI change, 1982–1999, and (b,d) for climate change and human activities on NDVI change, 2000–2015.
Figure 5. Impacts of influences factors on NDVI change, (a,c) for climate change and human activities on NDVI change, 1982–1999, and (b,d) for climate change and human activities on NDVI change, 2000–2015.
Sustainability 17 08233 g005
Sustainability 17 08233 g006
Figure 6. Spatial distribution of drivers of NDVI change, (a) 1982–1999, (b) 2000–2015.
Figure 6. Spatial distribution of drivers of NDVI change, (a) 1982–1999, (b) 2000–2015.
Sustainability 17 08233 g006
Sustainability 17 08233 g007
Figure 7. Spatial distribution of the contribution of drivers to changes in vegetation cover on the Loess Plateau, 1982–2015.
Figure 7. Spatial distribution of the contribution of drivers to changes in vegetation cover on the Loess Plateau, 1982–2015.
Sustainability 17 08233 g007
Sustainability 17 08233 g008
Figure 8. Spatial distributions of the partial correlation coefficients between NDVI and precipitation-temperature, (a) precipitation, (c) temperature, during 1982–1999 over the Loess Plateau; (b) precipitation, (d) temperature, during 2000–2015 over the Loess Plateau.
Figure 8. Spatial distributions of the partial correlation coefficients between NDVI and precipitation-temperature, (a) precipitation, (c) temperature, during 1982–1999 over the Loess Plateau; (b) precipitation, (d) temperature, during 2000–2015 over the Loess Plateau.
Sustainability 17 08233 g008
Sustainability 17 08233 g009
Figure 9. Area of Land Use/Cover Changes change in the Loess Plateau, 2000–2015.
Figure 9. Area of Land Use/Cover Changes change in the Loess Plateau, 2000–2015.
Sustainability 17 08233 g009
Table 1. Impact classification of climate change and human activity on vegetation recovery (10−3 a−1).
Table 1. Impact classification of climate change and human activity on vegetation recovery (10−3 a−1).
Slope (NDVI)Influence Degree
<−2.0Obvious inhibition
−2.0~−1.0Moderate inhibition
−1.0~−0.2Slight inhibition
−0.2~0.2Little influence
0.2~1.0Slight promotion
1.0~2.0Moderate promotion
≥2.0Obvious promotion
Table 2. The criteria for the determination of driving factors of vegetation NDVI change and the calculation method of contribution rate.
Table 2. The criteria for the determination of driving factors of vegetation NDVI change and the calculation method of contribution rate.
Slope (NDVIobs)Driving FactorsClassification Criteria for the DriversThe Contribution Rate of the Drivers(%)
Slope (NDVICC)Slope (NDVIHA)Climate ChangeHuman Activities
>0CC & HA>0>0 Slope NDVI cc Slope NDVI obs Slope NDVI HA Slope NDVI obs
CC>0<01000
HA<0>00100
<0CC & HA<0<0 Slope NDVI cc Slope NDVI obs Slope ( NDVI HA ) Slope ( NDVI obs )
CC<0>01000
HA>0<00100
Table 3. Area distribution of correlation between temperature, precipitation and NDVI in the Loess Plateau, 1982–2015.
Table 3. Area distribution of correlation between temperature, precipitation and NDVI in the Loess Plateau, 1982–2015.
Impact FactorPrecipitationTemperature
Year1982–19992000–20151982–19992000–2015
Significant positive correlation7.44%11.42%20.71%1.92%
Positive correlation68.74%73.24%58.25%78.94%
Significant negative correlation0.62%0.43%0.24%0.16%
Negative correlation23.19%14.92%20.81%18.98%
Table 4. Transition matrix of Land Use/Cover Changes in Loess Plateau between 2000 and 2015 (unit: Km2).
Table 4. Transition matrix of Land Use/Cover Changes in Loess Plateau between 2000 and 2015 (unit: Km2).
2015GrasslandsFarmlandShrubForestsNon-Vegetated
2000
Grasslands261,988.16 26,196.41 429.40 9565.55 5776.47
Farmland41,133.41 148,377.88 8.48 3505.79 6073.42
Shrub1251.27 17.77 1256.95 1278.65 0.09
Forests2632.42 1286.52 318.27 73,243.17 27.54
Non-vegetated14,010.18 2708.50 0.00 20.56 24,446.97
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

Li, J.; Liu, H.; Cheng, D.; Zhang, H.; Qi, G.; Wang, W. Vegetation Response to Climate and Human Interventions on the Loess Plateau: Trends, Variability, and the Influence of the Grain for Green Program. Sustainability 2025, 17, 8233. https://doi.org/10.3390/su17188233

AMA Style

Li J, Liu H, Cheng D, Zhang H, Qi G, Wang W. Vegetation Response to Climate and Human Interventions on the Loess Plateau: Trends, Variability, and the Influence of the Grain for Green Program. Sustainability. 2025; 17(18):8233. https://doi.org/10.3390/su17188233

Chicago/Turabian Style

Li, Jiangbo, Huan Liu, Dandong Cheng, Hangzhen Zhang, Guizeng Qi, and Weize Wang. 2025. "Vegetation Response to Climate and Human Interventions on the Loess Plateau: Trends, Variability, and the Influence of the Grain for Green Program" Sustainability 17, no. 18: 8233. https://doi.org/10.3390/su17188233

APA Style

Li, J., Liu, H., Cheng, D., Zhang, H., Qi, G., & Wang, W. (2025). Vegetation Response to Climate and Human Interventions on the Loess Plateau: Trends, Variability, and the Influence of the Grain for Green Program. Sustainability, 17(18), 8233. https://doi.org/10.3390/su17188233

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