EN中文

Journal of Arid Land >

2024 , Vol. 16 >Issue 8: 1044 - 1061

DOI: https://doi.org/10.1007/s40333-024-0059-2

  • SUN Chao 1, 2, 3 ,
  • BAI Xuelian 1 ,
  • WANG Xinping 1 ,
  • ZHAO Wenzhi , 1, * ,
  • WEI Lemin 1, 2
展开

收稿日期: 2024-01-09

  修回日期: 2024-05-25

  录用日期: 2024-05-30

  网络出版日期: 2025-08-13

收起

Response of vegetation variation to climate change and human activities in the Shiyang River Basin of China during 2001-2022

  • SUN Chao 1, 2, 3 ,
  • BAI Xuelian 1 ,
  • WANG Xinping 1 ,
  • ZHAO Wenzhi , 1, * ,
  • WEI Lemin 1, 2
Expand
  • 1Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou 730000, China
  • 2University of Chinese Academy of Sciences, Beijing 100049, China
  • 3Gansu Provincial Academy of Water Sciences, Lanzhou 730000, China
*ZHAO Wenzhi (E-mail: )

Received date: 2024-01-09

  Revised date: 2024-05-25

  Accepted date: 2024-05-30

  Online published: 2025-08-13

Fold

本文引用格式

SUN Chao , BAI Xuelian , WANG Xinping , ZHAO Wenzhi , WEI Lemin . [J]. Journal of Arid Land, 2024 , 16(8) : 1044 -1061 . DOI: 10.1007/s40333-024-0059-2

Abstract

Understanding the response of vegetation variation to climate change and human activities is critical for addressing future conflicts between humans and the environment, and maintaining ecosystem stability. Here, we aimed to identify the determining factors of vegetation variation and explore the sensitivity of vegetation to temperature (SVT) and the sensitivity of vegetation to precipitation (SVP) in the Shiyang River Basin (SYRB) of China during 2001-2022. The climate data from climatic research unit (CRU), vegetation index data from Moderate Resolution Imaging Spectroradiometer (MODIS), and land use data from Landsat images were used to analyze the spatial-temporal changes in vegetation indices, climate, and land use in the SYRB and its sub-basins (i.e., upstream, midstream, and downstream basins) during 2001-2022. Linear regression analysis and correlation analysis were used to explore the SVT and SVP, revealing the driving factors of vegetation variation. Significant increasing trends (P<0.05) were detected for the enhanced vegetation index (EVI) and normalized difference vegetation index (NDVI) in the SYRB during 2001-2022, with most regions (84%) experiencing significant variation in vegetation, and land use change was determined as the dominant factor of vegetation variation. Non-significant decreasing trends were detected in the SVT and SVP of the SYRB during 2001-2022. There were spatial differences in vegetation variation, SVT, and SVP. Although NDVI and EVI exhibited increasing trends in the upstream, midstream, and downstream basins, the change slope in the downstream basin was lower than those in the upstream and midstream basins, the SVT in the upstream basin was higher than those in the midstream and downstream basins, and the SVP in the downstream basin was lower than those in the upstream and midstream basins. Temperature and precipitation changes controlled vegetation variation in the upstream and midstream basins while human activities (land use change) dominated vegetation variation in the downstream basin. We concluded that there is a spatial heterogeneity in the response of vegetation variation to climate change and human activities across different sub-basins of the SYRB. These findings can enhance our understanding of the relationship among vegetation variation, climate change, and human activities, and provide a reference for addressing future conflicts between humans and the environment in the arid inland river basins.

Key words： vegetation variation; climate change; land use change; normalized difference vegetation index (NDVI); enhanced vegetation index (EVI); Shiyang River Basin

1 Introduction

Vegetation is one of the most important components of terrestrial ecosystems, and its growth status acts as an important indicator reflecting the response of vegetation variation to changes in environmental factors such as climate, soil, and hydrology, as well as human activities (Nemani et al., 2003; Poulter et al., 2014; Seddon et al., 2016; Reich et al., 2018). A significant body of research has been conducted on vegetation variation, including vegetation coverage, net primary productivity, and vegetation structure (Wang et al., 2019; Wei et al., 2019; Bai and Zhao, 2023), and their driving factors (Zhang et al., 2011; Zhou et al., 2014; Li et al., 2021). Climate change and anthropogenic pressures are considered to be the main reasons for vegetation variation (Lesk et al., 2016; Li et al., 2023; Yin et al., 2023a). For example, climate change can exacerbate the severity and frequency of drought events, affecting vegetation growth processes, phenology, species composition, and carbon cycling (Huang et al., 2016; Schlaepfer et al., 2017; Li et al., 2023; Zohner et al., 2023). High temperatures and drought limit vegetation growth in temperate grasslands (Yin et al., 2023b), and an increase in annual rainfall can promote vegetation greening (Ukkola et al., 2021). Intensive human activities lead to farmland expansion, exacerbating land degradation and global desertification processes (Song et al., 2018; Burrell et al., 2020; Winkler et al., 2021). The CO2 fertilization effect and temperature and precipitation changes induced by human activities are the drivers for the variation in global vegetation greenness (Smith et al., 2000; Zhang et al., 2022), and over 19% of the variation in vegetation greenness in the temperate regions of the Northern Hemisphere is mainly driven by land use change (Chen et al., 2019). Analyzing the impact of climate change and human activities on vegetation variation can help address conflicts between humans and the environment.
Many methods can be used to explore the impact of climate change and human activities on vegetation variation at different scales, including the dynamic land ecosystem model, regional climate model, biophysical model, and geographical detector (Geodetector) model (Tian et al., 2010; Ngabire et al., 2023). Although earlier studies have determined the influencing factors of vegetation variation in some regions, significantly different and even contradictory results for change trends and driving factors indicated the influence of spatial-temporal heterogeneity of the study regions (Jeong et al., 2011; Zhang et al., 2011; Zhu et al., 2016; Piao et al., 2020; Li et al., 2021). For example, some results indicated that climate change can lead to an increase in global net primary productivity (Nemani et al., 2003; Zhao and Running, 2010; Jeong et al., 2011), but other results suggested that the increase in CO2 caused by human activities may be a driving factor for global vegetation variation (Zhu et al., 2016; Piao et al., 2020). In addition, some studies have indicated that the contribution of human activities to vegetation growth and degradation was greater than 70% in the Hexi Corridor and Shiyang River Basin (SYRB), China (Zhang et al., 2011; Li et al., 2021), and that there was no significant correlation between vegetation growth and precipitation (Yang et al., 2016). However, other studies showed that climate change was the driver for vegetation increase in northern China (Liu et al., 2015; Zhou et al., 2015; Zhang et al., 2019; Chen et al., 2021). These studies have mainly focused on the response of vegetation variation to climate change and human activities at the level of the entire basin, while less information was available for different segments within the basin (Tang et al., 2017).
Sensitivity of vegetation to climate change is the degree or pattern of response of vegetation productivity in time and space (Huxman et al., 2004); it is an important indicator to reflect the degree of adaptation of vegetation to regional environments, and also a key indicator to understand the current and future variation in vegetation under climate change. Under global climate change, the sensitivity of vegetation to precipitation (SVP) showed a decreasing trend, with the highest and increased SVP in the arid regions and the decreased SVP in humid areas (Bao et al., 2021; Zeng et al., 2022; Zhang et al., 2022). During the past 40 a, the SVP on the Qinghai-Tibet Plateau of China, and in some other sparsely vegetated areas in global has increased (Scott et al., 2014; Wang et al., 2022a). Land use change and CO2 increase induced by human activities strongly affect the SVP (Ukkola et al., 2021). In addition, the spatial heterogeneity of different regions can lead to differences in the response of vegetation variation, resulting in various temporal and spatial response patterns of the sensitivity of vegetation to climate change (Cui et al., 2022; Xue and Wu, 2024). In China, the sensitivity of vegetation to climate change also has strong spatial heterogeneity, with grasslands and farmlands in northern China being more susceptible to drought than those in southern China (Ding et al., 2020; Jiao et al., 2021). Although many studies have focused on the sensitivity of vegetation to climate change, they mainly analyzed the SVP (Li et al., 2013, 2015), with less attention to the sensitivity of vegetation to temperature (SVT).
The SYRB is the third-largest inland river basin in China, and its vegetation growth and distribution are highly susceptible to the dual constraints of climate change and human activities (Chen et al., 2015; Bai et al., 2023). In the past few decades, accelerated human activities have led to the expansion of oases, significant variation in vegetation, accelerated desertification, and increasingly prominent conflicts between humans and the environment (Zhang et al., 2011; Tang et al., 2017; Wang et al., 2019). It is necessary to understand the spatial-temporal changes in the SVT and SVP to predict the response of vegetation to future climate change. Here, we systematically analyzed the response of vegetation to climate change and human activities in the SYRB. Our main goals were to: (1) understand the vegetation variation, climate change, and land use change in the SYRB during 2001-2022; (2) identify the SVT and SVP in different sub-basins of the SYRB; and (3) explore the driving factors of vegetation variation under the background of climate change and human activities in the SYRB. The findings of this study will increase the understanding of the driving mechanism of vegetation variation in the arid inland river basins, provide a theoretical basis for vegetation management, and support efforts to decelerate the process of desertification and alleviate the conflicts between humans and the environment.

2 Materials and methods

2.1 Study area

The SYRB is located in the eastern part of the Hexi Corridor in Gansu Province, China. The geographical location is between 101°07′24″-104°15′30″E and 37°07′22″N-39°27′47″N, and the area totals 4.0600×104 km2. The elevation of the SYRB is 1254-5125 m. The southern and central regions encompass the Qilian Mountains and plain area, respectively, and the northern region is covered by low mountains, hills, and deserts (Fig. 1). The SYRB has a temperate continental climate, characterized by strong solar radiation, large temperature differences, low precipitation, and strong evaporation. The average annual temperature during 1960-2022 was 8.10°C, the annual evapotranspiration varied from 1300.00 to 2600.00 mm, and the sunshine duration totaled 2700-3000 h. The annual precipitation was 50.00-200.00 mm in the SRYB, concentrated in summers. The climate exhibits significant spatial variability under the influence of geographical location, geomorphic differences, and atmospheric circulation, so the SYRB can be divided into three climate zones from south to north. The southern region is an alpine, semi-arid, and semi-humid area, with an average annual temperature of 0.30°C and an annual precipitation of 300.00-600.00 mm; the central region is a warm and cool arid area, with an average annual temperature and an annual precipitation of 7.70°C and 150.00-300.00 mm, respectively; and the northern region is a warm and arid area, with an average annual temperature of 9.30°C and an annual precipitation of <150.00 mm. Vegetation also exhibits strong north-south differentiation, with sparse forest and grassland in the northern region, agricultural vegetation in the central region, and desert vegetation in the southern region.
Fig. 1 Overview of the Shiyang River Basin (SYRB) and its three sub-basins. Note that the image was the true color composite (band 4/3/2) of Sentinel-2 data in 2023 downloaded from the European Space Agency (https://scihub.copernicus.eu/dhus/#/home).
The Shiyang River originates from the Lenglongling Mountain in the eastern part of the Qilian Mountains and is mainly composed of eight rivers. The total length of the Shiyang River is 250 km and its annual runoff is 15.6×108 m3. The SYRB is densely inhabited and has well-developed agriculture, both of which contribute to its highest utilization of water resources in the Hexi Corridor. Since the 1990s, extensive economic development and excessive groundwater extraction have resulted in severe degradation of the environment, and the problem of water resources has become increasingly prominent.

2.2 Data sources and pre-processing

Temperature, precipitation, and evapotranspiration are the three most important climate factors representing climate change, and they are used to analyze the response of vegetation to climate change (Bai and Zhao, 2023; Han et al., 2023). Land use change is the most intuitive manifestation of human activities because it can directly express changes in underlying surfaces, or indirectly express changes in population, economy, and water demand (Wang et al., 2006; Tang et al., 2013; Hu et al., 2017; Du et al., 2021). Here, land use change was used to represent human activities when analyzing the driving mechanism of vegetation variation. Climate data with 0.5°×0.5° resolution were provided by the climatic research unit (CRU) time series (TS) 4.07 dataset (http://www.cru.uea.ac.uk/), and the time coverage was 1901-2022. The runoff depth dataset was downloaded from the National Tibetan Plateau Data Center (http://data.tpdc.ac.cn) (Gou et al., 2021; Miao and Gou, 2022; Miao et al., 2022). This dataset comprised natural runoff depth data that were reconstructed from 200 hydrological stations in China for years 1961-2018, at a resolution of 0.25°. We obtained land use data through supervised classification based on Landsat images with a resolution of 1 km and an accuracy of >90%, which were provided by the Resource and Environmental Science Data Platform (https://www.resdc.cn/). Five Landsat images were used to analyze land use change, including years 2000, 2005, 2010, 2015, and 2020. To obtain a more accurate range of land use data, we used Sentinel-2 images to calibrate land use products. Then, we classified all land use types into six categories, namely forest, grassland, farmland, construction land, unused land, and water body, based on land resources and their utilization attributes. Finally, all of these data were resampled to a spatial resolution of 500 m to analyze the SVT and SVP and their determining factors.
Normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI) can effectively reflect vegetation conditions and have been widely used to represent the growth status of vegetation (Maurer et al., 2020; Piao et al., 2020). They were calculated using the Moderate Resolution Imaging Spectroradiometer (MODIS) Terra Surface Reflectance 8-day L3 Global 500 m SIN Grid (MOD09A1) product, with a time coverage of 2001-2022, a spatial resolution of 500 m, and a time resolution of 8 d. The equations were as follows:
$\mathrm{NDVI}=\frac{R_{\mathrm{NIR}}-R_{\mathrm{Red}}}{R_{\mathrm{NIR}}+R_{\text {Red }}},$
$\mathrm{EVI}=2.5 \times \frac{R_{\mathrm{NIR}}-R_{\text {Red }}}{R_{\mathrm{NIR}}+6 R_{\text {Red }}-7.5 R_{\text {Blue }}+1},$
where RNIR, RRed, and RBlue (nm) are the bands of near-infrared, red, and blue that correspond to the bands 1, 2, and 3 of MODIS data, respectively.

2.3 Methods

Linear regression can to help establish relationships between vegetation status and climate factors. Compared with other methods, this approach is easy to implement in some basins with low data requirements, and the regression's structure can effectively reduce the uncertainty (Zhang et al., 2022; Bai and Zhao, 2023). Here, we used linear regression to determine the changes in NDVI and EVI during 2001-2022, with slope representing the change trend. There is an increasing trend in NDVI (or EVI) when the slope is positive, and a decreasing trend in NDVI (or EVI) when the slope is negative. Linear regression was also used with annual NDVI (or EVI) and precipitation (or temperature) for each pixel during 2001-2022, and the slope was used to explore the SVT and SVP. A positive slope indicates that an increase in annual precipitation (or temperature) will promote vegetation growth, while a negative slope implies that a decrease in annual precipitation (or temperature) will hinder vegetation growth. The slopes of linear regression between the growth status of vegetation (annual mean NDVI and EVI) and climate factors (precipitation and temperature) were calculated as follows for sensitivity analysis:
$\mathrm{SVT}=\frac{n \sum_{i=1}^{n} T_{i} Y_{i}-\sum_{i=1}^{n} T_{i} \sum_{i=1}^{n} Y_{i}}{n \sum_{i=1}^{n} T_{i}^{2}-\left(\sum_{i=1}^{n} T_{i}\right)^{2}},$
$\mathrm{SVP}=\frac{n \sum_{i=1}^{n} P_{i} Y_{i}-\sum_{i=1}^{n} P_{i} \sum_{i=1}^{n} Y_{i}}{n \sum_{i=1}^{n} P_{i}^{2}-\left(\sum_{i=1}^{n} P_{i}\right)^{2}},$
where SVT is the sensitivity of vegetation to temperature; n is the number of years; i is the ith year; Ti is the temperature in the ith year (°C); Yi is the NDVI (or EVI) in the ith year; SVP is the sensitivity of vegetation to precipitation; and Pi is the precipitation in the ith year (mm).
In this study, we used a 5-a moving average to analyze the temporal trends in the SVT and SVP. The SVT and SVP of NDVI and EVI were calculated across all grids of satellite images for a 5-a window during 2001-2022, and then the average SVT and SVP were calculated to explore the overall temporal dynamics of the SVT and SVP in the study area. To identify the determining factors of the temporal variations in NDVI and EVI, we used multiple correlation analysis to explore the relationships between several factors. Then, correlation coefficient values of all factors were ranked, and the factor with the highest value was taken as the dominant factor. In general, land use change was represented by area change that could be considered as numerical variable. Therefore, we used the area of each land use type to represent the land use change over time when analyzing the correlation coefficient between several factors.
In this study, the classification and area statistics of land use type, processing of climate data and runoff depth, calculation of vegetation indices, and analysis of the SVT and SVP were conducted on the ArcGIS 10.3 platform, and the exploration on the drivers of vegetation variation was completed through multivariate correlation analysis on the Statistical Package for the Social Sciences (SPSS) software produced by the International Business Machines Corporation (IBM) in New York, USA. Furthermore, the spatial distribution maps were drawn using ArcGIS 10.3, and the temporal change figures were plotted using Origin 2021 produced by OriginLab in Northampton, Massachusetts, USA.

3 Results

3.1 Spatial-temporal variation in vegetation

Significant increasing trends were detected for NDVI and EVI in the SYRB during 2001-2022, and the same trends were also found in each sub-basin (Fig. 2). The increasing trends in NDVI were relatively large in the upstream and midstream basins of the SYRB (with the slope values of 0.0028/a and 0.0021/a, respectively), while the increasing trend was relatively small in the downstream basin (with the slope value of 0.0007/a). The change trend in EVI in each sub-basin was consistent with that in NDVI. The average annual NDVI and EVI were 0.16 and 0.11 in the SYRB during 2001-2022, respectively. NDVI and EVI decreased overall from the upstream to downstream basin, with the average NDVI of 0.25, 0.16, and 0.10, and average EVI of 0.15, 0.11, and 0.07 in the upstream, midstream, and downstream basins, respectively (Fig. 2). Regions with increasing trends in NDVI and EVI had an advantage over regions with decreasing trends in these vegetation indices, accounting for 95% and 5% of the total SYRB area, respectively, and regions with decreasing trends in NDVI and EVI were mainly distributed in the midstream and downstream basins, including the oasis areas. More specifically, regions (area of 3.4000×104 km2) with a significant change in NDVI accounted for 84% of the total SYRB area, and regions with a significant increase in NDVI accounted for 82% of the total area. Regions with a significant increase in NDVI were larger in the upstream basin than in the midstream and downstream basins, with areas accounting for 90%, 83%, and 76% of the total area of the upstream, midstream, and downstream basins, respectively.
Fig. 2 Spatial (a1-a6) and temporal (b1 and b2) distribution of normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI) in the SYRB and its sub-basins during 2001-2022. (a1), spatial distribution of average NDVI; (a2), spatial distribution of the change slope of NDVI; (a3), spatial distribution of the correlation coefficient between time and NDVI; (a4), spatial distribution of average EVI; (a5), spatial distribution of the change slope of EVI; (a6), spatial distribution of the correlation coefficient between time and EVI; (b1), temporal distribution of NDVI; (b2), temporal distribution of EVI.

3.2 Climate change and land use change

3.2.1 Climatic and hydrological variations

The average annual temperature and precipitation were 6.08°C and 247.89 mm during 2001-2022 in the SYRB, respectively. Climate and hydrology varied in different sub-basins of the SYRB, with increasing trends in temperature and potential evapotranspiration, and decreasing trends in precipitation and runoff depth from the upstream to downstream basin (Fig. 3). Temperature was respectively 0.97°C, 6.38°C, and 9.26°C, and precipitation was respectively 420.79, 244.84, and 135.03 mm, in the upstream, midstream, and downstream basins. The average potential evapotranspiration and runoff depth were 31.31 mm/d and 93.55 mm, respectively, in the SYRB. Specifically, the average potential evapotranspiration was 26.71, 31.14, and 34.50 mm/d, respectively, and the average runoff depth was 199.09, 54.85, and 49.58 mm, respectively, for the upstream, midstream, and downstream basins. A significant increasing trend was found in temperature in the SYRB during 2001-2022, and the same change trend was also found in each sub-basin. The change trend of temperature was relatively large in the downstream basin, with a slope of 0.06°C/a. Although a non-significant change was found in precipitation, an increasing trend was detected in each sub-basin in the past two decades, especially in the upstream basin where the change trend of precipitation was relatively large (with an increasing slope of 2.77 mm/a). In the SYRB as well as in the downstream and midstream basins, there was an increasing trend in potential evapotranspiration, and in the upstream basin, there was a decreasing trend in potential evapotranspiration. A decreasing trend in runoff depth was detected in the downstream basin, while there was an increasing trend in runoff depth in the other sub-basins.
Fig. 3 Temporal changes of climatic and hydrological factors in the SYRB and its sub-basins during 2001-2022. (a), temperature; (b), precipitation; (c), potential evapotranspiration; (d), runoff depth.

3.2.2 Spatial-temporal change in land use

The SYRB was mainly occupied by unused land, farmland, and grassland (Fig. 4), totaling over 92% of the total basin area. The area of unused land made up the largest proportion of the total SYRB area in 2000 and 2020, at 49% and 47%, respectively. Farmland area occupied 17% and 18% of the total SYRB area in 2000 and 2020, respectively. The changes in the areas of unused land, farmland, grassland, water body, and construction land were significant (P<0.05) between years 2000 and 2020, while the change in the area of forest was not significant. A decrease in the areas of grassland and unused land, and an increase in the areas of farmland and construction land were the main land use change trends in the SYRB during 2001-2022. Specifically, during 2001-2022, farmland and construction land increased by 0.0490×104 and 0.0228×104 km2, respectively, while unused land and grassland decreased by 0.0596×104 and 0.0150×104 km2, respectively. From 2000 to 2020, the land conversion area of the SYRB was 0.9199×104 km2, accounting for 23% of the total basin area (Fig. 4). The reciprocal conversion between grassland and unused land was the main land use conversion, of which the grassland conversion was the largest at 0.3513×104 km2, accounting for 38% of the total conversion area. Among them, grassland was the dominant type of land use in the upstream basin, with an area proportion of 53% of the total upstream basin area, and the area proportions of farmland and forest were 16% and 21% of the total upstream basin area, respectively. Unused land, farmland, and grassland were the dominant land use types in the midstream basin, with area proportions of 44%, 29%, and 23% of the total midstream basin area, respectively. In the downstream basin, unused land was by far the main land use type, with an area proportion of 77%, while the area proportions of farmland and grassland were 10% and 12%, respectively. The land conversion areas decreased from the upstream to downstream basin, with area proportions of 29%, 26%, and 15% of the total upstream, midstream, and downstream basin area, respectively. The reciprocal conversion between grassland and forest was the main land use change in the upstream basin, while the conversions among unused land, grassland, and farmland were the dominant land use conversions in the midstream and downstream basins.
Fig. 4 Spatial-temporal distribution of land use types in 2000 (a1), 2005 (a2), 2010 (a3), 2015 (a4), and 2020 (a5), spatial distribution of major land use conversions during 2001-2022 (b), chord diagram of the proportion of land use conversion area during 2001-2022 (c), and area proportions of different land use types during 2001-2022 (d1-d4) in the SYRB and its sub-basins

3.3 Spatial-temporal variations in the SVT and SVP, and drivers of vegetation variation

A non-significant decreasing trend was detected in the SVT in the SYRB during 2001-2022, with averages of -1.20×10-3/°C and -0.30×10-3/°C, respectively, for NDVI and EVI (Fig. 5). Spatial differences in the SVT were also observed from the upstream to downstream basin, with relatively high SVT of -3.10×10-3/°C in the upstream basin and 0.60×10-3/°C in the downstream basin for NDVI, and high SVT of -3.40×10-3/°C in the upstream basin and 1.60×10-3/°C in the midstream basin for EVI. The SVT of NDVI increased rapidly and that of EVI decreased rapidly in the upstream basin during 2001-2022, with slope values of 0.68×10-3/(°C•a) and 0.40×10-3/(°C•a), respectively. The SVT changed slowly in the midstream basin for NDVI and in the downstream basin for EVI. The average annual SVP in the SYRB during 2001-2022 was 7.30×10-5/mm for NDVI and 4.80×10-5/mm for EVI, and there was a non-significant decreasing trend in the upstream and downstream basins. Vegetation was more sensitive to precipitation in the upstream and midstream basins, with the SVP of NDVI at 10.20×10-5/mm and 9.70×10-5/mm, respectively.
Fig. 5 Temporal changes in the sensitivity of vegetation to temperature (SVT) and sensitivity of vegetation to precipitation (SVP) in the SYRB and its sub-basins in a 5-a moving window during 2001-2022. (a), SVT of NDVI; (b), SVT of EVI; (c), SVP of NDVI; (d), SVP of EVI.
In the SYRB, there was a relatively small area with a significant correlation between vegetation indices and climate factors (Fig. 6). Regions with significant changes in the SVT and SVP of NDVI accounted for 23% and 31% of the total SYRB area, respectively. Regions with significant change in the SVT of NDVI were mainly distributed in the upstream basin and regions with significant change in the SVP of NDVI were mainly located in the midstream basin. The spatial differences in regions with significant changes in the SVT and SVP were also found from the upstream to downstream basin, with a relatively larger area showing significant change in the SVT of NDVI in the midstream basin than in the upstream and downstream basins. The area proportions with a significant change in the SVT of NDVI were 17%, 36%, and 16% of the upstream, midstream, and downstream basin area, respectively. Compared with the SVT of NDVI, there was a smaller area where the SVT of EVI significantly changed, with area proportions of 9%, 38%, and 21% of each basin area from the upstream to downstream basin, respectively. The area with a significant change in the SVP of NDVI was lower in the downstream basin than in the upstream and midstream basins, with area proportions of 50%, 44%, and 7% of each basin area from the upstream to downstream basin, respectively.
Fig. 6 Spatial distribution of the SVP and SVT of NDVI and EVI (a1-a4) and their correlations with time (a5-a8), and area proportions of regions with different changes of the SVP and SVT of NDVI and EVI (b1-b4) in the SYRB and its sub-basins during 2001-2022
In the SYRB, land use change was the determining factor in vegetation variation, and the effect of climate change on vegetation variation was smaller than that of land use change. There was a significant correlation between land use change and vegetation indices (Fig. 7), while there was no significant correlation of vegetation indices with temperature, precipitation, and potential evapotranspiration changes. Vegetation indices were significantly negatively correlated with water body, grassland, unused land, and construction land changes (P<0.05), and grassland change was the dominant factor of vegetation variation, with the highest correlation coefficient of 0.98. The dominant factors of vegetation change varied in different sub-basins. Specifically, in the upstream basin, there was a significant negative correlation of vegetation indices with water body, grassland, and unused land, and water body change was found to be the dominant factor of vegetation variation, with the highest correlation coefficient of 0.93. The dominant factors of vegetation variation were farmland and construction land changes in the midstream basin, and water body, unused land, and forest changes in the downstream basin.
Fig. 7 Person's correlation coefficients among vegetation indices, climate change, runoff depth change, and land use change in the SYRB and its sub-basins. (a), SYRB; (b), upstream basin; (c), midstream basin; (d), downstream basin. The direction of ellipse represents the positive or negative correlation coefficient, with negative value to the left direction and positive value to the right direction. The size of ellipse represents the magnitude of the correlation coefficient, with a larger ellipse indicating a smaller correlation coefficient. * represents a significant correlation at the 0.05 level.

4 Discussion

4.1 Driving mechanism of vegetation variation

Vegetation variation is influenced by various factors, such as land use change, geomorphic properties, soil properties, hydrological characteristics, and climate change, but mainly by climate change and human activities (Nemani et al., 2003; Poulter et al., 2014; Seddon et al., 2016; Reich et al., 2018). The significant increasing trends in the EVI and NDVI in the SYRB during 2001-2022 indicated that the vegetation status has been continuously improving since 2001, with an increase of 84% in area, which was consistent with the overall trend of global large-scale vegetation greening (Piao et al., 2020; Yang et al., 2023). Research has reported that the CO2-driven fertilization affected global greening (Smith et al., 2000; Piao et al., 2020), and climate change dominated vegetation growth in arid and semi-arid areas of China (Chen et al., 2021). There were temporal differences in vegetation variation in the SYRB, where human activities were the main drivers of vegetation increase, as shown in our study. The reason was that the implementation of ecological restoration projects, such as planting of trees and grasses and returning farmland to forest, has led to significant change in land use in the SYRB since 2001, with a corresponding reduction of 0.0560×104 km2 in desert area (Fig. 4). Regions showing the increasing trend in vegetation area were largest in the downstream basin (18% of the downstream basin area), but the rate of increase in the downstream basin was slower than those in the midstream and downstream basins. This may be due to different vegetation types. The downstream basin is mainly dominated by unused land (77% of the total downstream basin area), and natural vegetation is composed of dry or super dry shrubs and semi-shrubs, such as Artemisia desertorum, Nitraria tangutorum, and Zygophyllum xanthoxylum. These species have low leaf area index, productivity, and vegetation coverage, resulting in a lower rate of change. In the downstream basin, the survival rate of artificial vegetation is relatively low due to limitations in water resources and high temperatures, which also leads to a lower rate of change (Bai et al., 2008; Zhang et al., 2011, 2019; Zhou et al., 2014; Gherardi and Sala, 2019). Furthermore, the downstream basin of the SYRB is located between the Tengger and Badain Jaran deserts and is influenced by strong interference from sandstorms and land desertification, which all affect vegetation survival (Zhang et al., 2011; Zhou et al., 2014; Ngabire et al., 2023).

4.2 Spatial-temporal change in the SVT

The non-significant downward trends in the SVT of NDVI and EVI suggested that the dependence of vegetation variation on temperature was continuously weakening in the SYRB, which is consistent with results obtained in the country (China) and North America (Bi et al., 2013; He et al., 2017). In different sub-basins of the SYRB, there were differences in the interannual change in the SVT; an upward trend in the SVT of NDVI was found in all three sub-basins, while a downward trend in the SVT of EVI was detected in the upstream and midstream basins. Moreover, the SVT of NDVI increased at a faster rate in the upstream basin than in the midstream and downstream basins, indicating that areas with high vegetation coverage may decrease, while areas with medium and low vegetation coverage may increase. First, climate change may lead to variation in dominant plant species or ecosystem types, which may play an important role in the temporal trend in the SVT (Zeng et al., 2022). In the upstream basin, the SVT of the ecosystem may exhibit an increasing trend with an increase in temperature, such as from needle-leaf to broad-leaf forests and shrublands. The slowest rate of variation in the SVT in the midstream and downstream basins indicated that vegetation growth was not sensitive to temperature change. The reason may be that the increase in temperature promoted vegetation growth in the midstream and downstream basins, leading to a comprehensive relief of heat stress (Jeong et al., 2009; Wang et al., 2022b). Furthermore, the increase in temperature had an important impact on the functional traits of plant leaves and roots (Wei et al., 2023), thereby reducing the dependence of plant growth on heat conditions. In the SYRB, the SVT was greater in the upstream basin than in the midstream and downstream basins, indicating that vegetation was more susceptible to temperature effects in the upstream basin. This difference is related to regional dry and wet climate conditions and temperature variability (Wu et al., 2017; Li et al., 2022).
In arid climate zones, vegetation variation is sensitive to precipitation and evapotranspiration, while in cold climate zones, vegetation is sensitive to temperature (Li et al., 2022). In this study, precipitation and runoff depth in the upstream basin were greater than those in the other two sub-basins, and vegetation growth was less limited by water content in these sub-basins. The upstream basin is located at a higher altitude and is covered by alpine vegetation, which is more sensitive to temperature change than to precipitation and evapotranspiration changes. In addition, the interannual SVT was significantly influenced by temperature change, and higher interannual temperature variation has led to a sustained decrease in the interannual SVT, and vice versa (Wu et al., 2017). In the SYRB, the upstream basin had the smallest interannual temperature change, at 0.02°C/a, resulting in a higher SVT.

4.3 Spatial-temporal change in the SVP

A non-significant downward trend in the SVP indicated that the dependence of vegetation on precipitation is decreasing in the SYRB, which is consistent with previous research results (Zeng et al., 2022). Many studies have reported that the decline in the global SVP is mainly caused by the fertilization effect of CO2 (Smith et al., 2000; Zhang et al., 2022), but some studies suggested that this is due to climate change such as increased precipitation and vapor pressure differences (Hsu et al., 2021; Zeng et al., 2022). However, the driving factors of vegetation variation in the SYRB differed at the sub-basin scale in this study. Precipitation, evapotranspiration, and runoff depth in the SYRB did not significantly differ during 2001-2022, while land use patterns of farmland, unused land, and grassland changed significantly and were significantly correlated with vegetation indices, indicating that land use change caused by the intensified human activities may be the main driver for the decline in the SVP in the SYRB. In addition, there was a difference in the decreased slope of the SVP in different sub-basins; our results showed a significant downward trend and the fastest decline in the slope of the SVP in the midstream basin. The reason may be that human activities were more intensive in the midstream basin than in the other two sub-basins, and improvements in irrigation, biotechnology, and breeding technologies have enhanced the ability of vegetation to resist drought and heat waves (Gupta et al., 2020).
The SVP in the upstream and midstream basins was greater than that in the downstream basin, which is inconsistent with the results obtained at a global scale (Huxman et al., 2004; Zeng et al., 2022). First, it may be that vegetation types influenced the SVP; a large desert area with low vegetation coverage and productivity in the downstream basin of the SYRB resulted in a low SVP. Second, vegetation growth was influenced by other factors in addition to precipitation, such as runoff depth, groundwater, and temperature. Our results indicated that temperature, evapotranspiration, and runoff depth in the downstream basin of the SYRB have increased since 2001, and the combined effect of high temperature and drought hindered vegetation growth, which may also affect vegetation response to precipitation (Li et al., 2023). However, vegetation growth may be limited by a single factor only; for example, in humid areas, plant growth was usually not limited by water but heat conditions, so the SVP was relatively small (Zeng et al., 2022). In the upstream basin of the SYRB, precipitation and runoff depth were highest, but the average annual temperature was only 0.97°C and evapotranspiration was the lowest among sub-basins. Therefore, vegetation sensitivity in the upstream basin was mainly affected by temperature. In the midstream basin, precipitation was relatively low, and a significant increase in temperature has led to an increase in evapotranspiration, which exacerbated water pressure and rendered water as the dominant factor in vegetation growth. Additionally, we found that the peak SVP corresponded to the lowest SVT in some periods, such as 2001-2005, 2008-2012, and 2015-2019. This indicated that the limitations on vegetation growth decreased with the increase of temperature or precipitation, and multiple environmental stress factors may shift to a single stress factor. This may be due to the interaction between temperature and precipitation, in which increased temperature influenced daily precipitation, thereby affecting the SVP (Ham et al., 2023). If the rising temperature leads to a continuous increase in precipitation, the decline in the SVP will continue, and vegetation productivity will be more stable (Donat et al., 2016).
Overall, temperature and precipitation are generally interrelated, and the sensitivity of regional vegetation to temperature and precipitation may be influenced by the comprehensive effect of the interaction between them (Bai et al., 2024). The SVT and SVP are also influenced by the differences in soil physical-chemical properties and vegetation physiological characteristics, especially in arid areas where water-bearing capacity of soil is low, and vegetation recovery ability is relatively weak after being affected by high temperatures and drought. In addition, with the intensification of human activities, grazing and the expansion of oases have also increased the level of interference with vegetation growth and climate characteristics, thereby affecting the sensitivity of vegetation to climate change (Zhang et al., 2011; Tang et al., 2017; Wang et al., 2019). In the future, the impact of human activities needs to be reconciled with regional climate characteristics and vegetation growth environment such that human activities can be controlled to reduce the negative impacts on natural vegetation. Different ecological measures should be implemented in different small basins based on the soil and water carrying capacity and corresponding mechanisms for vegetation growth to achieve vegetation restoration and maintain ecosystem stability.

5 Conclusions

In this study, the response of vegetation variation to climate change and human activities was analyzed in the SYRB in Northwest China. Vegetation has constantly increased from 2001 to 2022, with significant increases in NDVI and EVI; these vegetation indices were mainly driven by land use change, while temperature and precipitation had no significant impact on them in the SYRB as a whole. The dependence of vegetation variation on temperature and precipitation decreased over time with downward trends in the SVT and SVP. There were spatial differences in vegetation indices, SVT, and SVP in sub-basins. NDVI and EVI showed increasing trends in the upstream, midstream, and downstream basins of the SYRB, but the change slope in the downstream basin was lower than those in the upstream and midstream basins. The SVT in the upstream basin was higher than those in the midstream and downstream basins, and the SVP in the downstream basin was lower than those in the upstream and midstream basins. The drivers of vegetation variation in the sub-basins were different from those in the entire SYRB. Precipitation and temperature changes controlled vegetation variation in the upstream and midstream basins while human activities (land use change) dominated vegetation variation in the downstream basin.
In future research, a more specific and quantitative analysis is recommended to focus on the relationship between human activities and vegetation variation in the inland river basins, and concentrate on the responses of different vegetation types to soil and water carrying capacity and ecological measures, to provide a scientific basis for promoting vegetation restoration and maintaining ecosystem stability.

Conflict of interest

WANG Xinping and ZHAO Wenzhi are editorial board members of Journal of Arid Land and were not involved in the editorial review or the decision to publish this article. All authors declare that there are no competing interests.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (42230720).

Author contributions

Conceptualization: SUN Chao, ZHAO Wenzhi; Methodology: BAI Xuelian; Formal analysis: WANG Xinping, WEI Lemin; Writing - original draft preparation: SUN Chao; Writing - review and editing: BAI Xuelian, ZHAO Wenzhi; Funding acquisition: ZHAO Wenzhi; Resources: ZHAO Wenzhi; Supervision: WANG Xinping, WEI Lemin. All authors approved the manuscript.

参考文献

原文顺序 | 文献年度倒序 | 文中引用次数倒序
[1]
Bai X L, Zhao W Z. 2023. Impacts of climate change and anthropogenic stressors on runoff variations in major river basins in China since 1950. Science of the Total Environment, 898: 165349, doi: 10.1016/j.scitotenv.2023.165349.

[2]
Bai X L, Zhao W Z, Liu H, et al. 2023. Effects of precipitation changes and land-use alteration on streamflow: A comparative analysis from two adjacent catchments in the Qilian Mountains, arid northwestern China. Frontiers in Environmental Science, 11: 1097049, doi: 10.3389/fenvs.2023.1097049.

[3]
Bai X L, Zhao W Z, Luo W C, et al. 2024. Effect of climate change on the seasonal variation in photosynthetic and non-photosynthetic vegetation coverage in desert areas, Northwest China. Catena, 239: 107954, doi: 10.1016/j.catena.2024.107954.

[4]
Bai Y F, Wu J G, Xing Q, et al. 2008. Primary production and rain use efficiency across a precipitation gradient on the Mongolia Plateau. Ecology, 89(8): 2140-2153.

[5]
Bao Z X, Zhang J Y, Wang G Q, et al. 2021. The sensitivity of vegetation cover to climate change in multiple climatic zones using machine learning algorithms. Ecological Indicators, 124: 107443, doi: 10.1016/j.ecolind.2021.107443.

[6]
Bi J, Xu L, Samanta A, et al. 2013. Divergent Arctic-Boreal vegetation changes between North America and Eurasia over the past 30 years. Remote Sensing, 5(5): 2093-2112.

[7]
Burrell A L, Evans J P, De Kauwe M G. 2020. Anthropogenic climate change has driven over 5 million km2 of drylands towards desertification. Nature Communications, 11(1): 3853, doi: 10.1038/s41467-020-17710-7.

[8]
Chen C, Park T, Wang X H, et al. 2019. China and India lead in greening of the world through land-use management. Nature Sustainability, 2(2): 122-129.

[9]
Chen Y N, Zhi L, Fan Y T, et al. 2015. Progress and prospects of climate change impacts on hydrology in the arid region of Northwest China. Environmental Research, 139: 11-19.

[10]
Chen Y Z, Feng X M, Tian H Q, et al. 2021. Accelerated increase in vegetation carbon sequestration in China after 2010: A turning point resulting from climate and human interaction. Global Change Biology, 27(22): 5848-5864.

[11]
Cui L Z, Pang B, Zhao G, et al. 2022. Assessing the sensitivity of vegetation cover to climate change in the Yarlung Zangbo River Basin using machine learning algorithms. Remote Sensing, 14(7): 1556, doi: 10.3390/rs14071556.

[12]
Ding Y B, Xu J T, Wang X W, et al. 2020. Spatial and temporal effects of drought on Chinese vegetation under different coverage levels. Science of the Total Environment, 716: 137166, doi: 10.1016/j.scitotenv.2020.137166.

[13]
Donat M G, Lowry A L, Alexander L V, et al. 2016. More extreme precipitation in the world's dry and wet regions. Nature Climate Change, 7(2): 154-158.

[14]
Du Y, Li J Z, Niu K J, et al. 2021. Analysis of vegetation change and its impact on natural runoff in the mountain area of the Yongding River Basin from 1982 to 2015. Journal of Hydraulic Engineering, 52(11): 1309-1323. (in Chinese)

[15]
Gherardi L A, Sala O E. 2019. Effect of interannual precipitation variability on dryland productivity: A global synthesis. Global Change Biology, 25(1): 269-276.

[16]
Gou J J, Miao C Y, Samaniego L, et al. 2021. CNRD v1.0: A high-quality natural runoff dataset for hydrological and climate studies in China. Bulletin of American Meteorological Society, 102(5): E929-E947.

[17]
Gupta A, Rico-Medina A, Cano-Delgado A I. 2020. The physiology of plant responses to drought. Science, 368(6488): 266-269.

[18]
Ham Y G, Kim J H, Min S K, et al. 2023. Anthropogenic fingerprints in daily precipitation revealed by deep learning. Nature, 622(7982): 301-307.

[19]
Han Z Q, Zuo Q T, Wang C Q, et al. 2023. Impacts of climate change on natural runoff in the Yellow River Basin of China during 1961-2020. Water, 15(5): 929, doi: 10.3390/w15050929.

[20]
He B, Chen A F, Jiang W G, et al. 2017. The response of vegetation growth to shifts in trend of temperature in China. Journal of Geographical Sciences, 27(7): 801-816.

[21]
Hsu P, Takahashi Y, Merilo E, et al. 2021. Raf-like kinases and receptor-like (pseudo)kinase GHR1 are required for stomatal vapor pressure difference response. Proceedings of the National Academy of Sciences of the United States of America, 118(47): e2107280118, doi: 10.1073/pnas.2107280118.

[22]
Hu J, Lu Y H, Fu B J, et al. 2017. Quantifying the effect of ecological restoration on runoff and sediment yields: A meta-analysis for the Loess Plateau of China. Progress in Physical Geography-Earth and Environment, 41(6): 753-774.

[23]
Huang J P, Yu H P, Guan X D, et al. 2016. Accelerated dryland expansion under climate change. Nature Climate Change, 6(2): 166-171.

[24]
Huxman T E, Smith M D, Fay P A, et al. 2004. Convergence across biomes to a common rain-use efficiency. Nature, 429(6992): 651-654.

[25]
Jeong S J, Ho C H, Jeong J H. 2009. Increase in vegetation greenness and decrease in springtime warming over East Asia. Geophysical Research Letters, 36(2): L02710, doi: 10.1029/2008gl036583.

[26]
Jeong S J, Ho C H, Gim H J, et al. 2011. Phenology shifts at start vs. end of growing season in temperate vegetation over the Northern Hemisphere for the period 1982-2008. Global Change Biology, 17(7): 2385-2399.

[27]
Jiao K W, Gao J B, Liu Z H, et al. 2021. Revealing climatic impacts on the temporal and spatial variation in vegetation activity across China: Sensitivity and contribution. Advances in Climate Change Research, 12(3): 409-420.

[28]
Lesk C, Rowhani P, Ramankutty N. 2016. Influence of extreme weather disasters on global crop production. Nature, 529(7584): 84-87.

[29]
Li C H, Wang Y T, Wu X D, et al. 2021. Reducing human activity promotes environmental restoration in arid and semi-arid regions: A case study in Northwest China. Science of the Total Environment, 768: 144525, doi: 10.1016/j.scitotenv.2020.144525.

[30]
Li F, Zhao W Z, Liu H. 2013. The response of aboveground net primary productivity of desert vegetation to rainfall pulse in the temperate desert region of Northwest China. PLoS ONE, 8(9): e73003, doi: 10.1371/journal.pone.0073003.

[31]
Li F, Zhao W Z, Liu H. 2015. Productivity responses of desert vegetation to precipitation patterns across a rainfall gradient. Journal of Plant Research, 128(2): 283-294.

[32]
Li G C, Chen W, Zhang X P, et al. 2022. Spatiotemporal changes and driving factors of vegetation in 14 different climatic regions in the global from 1981 to 2018. Environmental Science and Pollution Research, 29(50): 75322-75337.

[33]
Li W T, Pacheco-Labrador J, Migliavacca M, et al. 2023. Widespread and complex drought effects on vegetation physiology inferred from space. Nature Communications, 14(1): 4640, doi: 10.1038/s41467-023-40226-9.

[34]
Liu C Y, Dong X F, Liu Y Y. 2015. Changes of NPP and their relationship to climate factors based on the transformation of different scales in Gansu, China. Catena, 125: 190-199.

[35]
Maurer G E, Hallmark A J, Brown R F, et al. 2020. Sensitivity of primary production to precipitation across the United States. Ecology Letters, 23(3): 527-536.

[36]
Miao C, Gou J. 2022. CNRDv1.0: the China natural runoff dataset version 1.0 (1961-2018). National Tibetan Plateau/Third Pole Environment Data Center. [2023-10-20]. https://doi.org/10.11888/Atmos.tpdc.272864.

[37]
Miao C Y, Gou J J, Fu B J, et al. 2022. High-quality reconstruction of China's natural streamflow. Science Bulletin, 67(5): 547-556.

[38]
Nemani R R, Keeling C D, Hashimoto H, et al. 2003. Climate-driven increases in global terrestrial net primary production from 1982 to 1999. Science, 300(5625): 1560-1563.

[39]
Ngabire M, Wang T, Liao J, et al. 2023. Quantitative analysis of desertification-driving mechanisms in the Shiyang River Basin: Examining interactive effects of key factors through the geographic detector model. Remote Sensing, 15(12): 2960, doi: 10.3390/rs15122960.

[40]
Piao S L, Wang X H, Park T, et al. 2020. Characteristics, drivers and feedbacks of global greening. Nature Reviews Earth & Environment, 1(1): 14-27.

[41]
Poulter B, Frank D, Ciais P, et al. 2014. Contribution of semi-arid ecosystems to interannual variability of the global carbon cycle. Nature, 509(7502): 600-603.

[42]
Reich P B, Sendall K M, Stefanski A, et al. 2018. Effects of climate warming on photosynthesis in boreal tree species depend on soil moisture. Nature, 562(7726): 263-267.

[43]
Schlaepfer D R, Bradford J B, Lauenroth W K, et al. 2017. Climate change reduces extent of temperate drylands and intensifies drought in deep soils. Nature Communications, 8: 14196, doi: 10.1038/ncomms14196.

[44]
Scott R L, Huxman T E, Barron-Gafford G A, et al. 2014. When vegetation change alters ecosystem water availability. Global Change Biology, 20(7): 2198-2210.

[45]
Seddon A W, Macias-Fauria M, Long P R, et al. 2016. Sensitivity of global terrestrial ecosystems to climate variability. Nature, 531(7593): 229-232.

[46]
Smith S D, Huxman T E, Zitzer S F, et al. 2000. Elevated CO2 increases productivity and invasive species success in an arid ecosystem. Nature, 408(6808): 79-82.

[47]
Song X P, Hansen M C, Stehman S V, et al. 2018. Global land change from 1982 to 2016. Nature, 560(7720): 639-643.

[48]
Tang Y, Tang Q, Tian F, et al. 2013. Responses of natural runoff to recent climatic variations in the Yellow River basin, China. Hydrology and Earth System Sciences, 17(11): 4471-4480.

[49]
Tang Z G, Ma J H, Peng H H, et al. 2017. Spatiotemporal changes of vegetation and their responses to temperature and precipitation in upper Shiyang River basin. Advances in Space Research, 60(5): 969-979.

[50]
Tian H Q, Liu M L, Zhang C, et al. 2010. The dynamic land ecosystem model (DLEM) for simulating terrestrial processes and interactions in the context of multifactor global change. Acta Geographica Sinica, 65(9): 1027-1047. (in Chinese)

[51]
Ukkola A M, De Kauwe M G, Roderick M L, et al. 2021. Annual precipitation explains variability in dryland vegetation greenness globally but not locally. Global Change Biology, 27(18): 4367-4380.

[52]
Wang C, Vera-Velez R, Lamb E G, et al. 2022a. Global pattern and associated drivers of grassland productivity sensitivity to precipitation change. Science of the Total Environment, 806: 151224, doi: 10.1016/j.scitotenv.2021.151224.

[53]
Wang J, Xie Y W, Wang X Y, et al. 2019. Detecting patterns of vegetation gradual changes (2001-2017) in Shiyang River Basin, based on a novel framework. Remote Sensing, 11(21): 2475, doi: 10.3390/rs11212475.

[54]
Wang L D, Hu F, Zhang C Y, et al. 2022b. Response of vegetation to different climate extremes on a monthly scale in Guangdong, China. Remote Sensing, 14(21): 5369, doi: 10.3390/rs14215369.

[55]
Wang X Q, Zhang Y, Zhang Y H. 2006. Change of natural runoff and contribution of the natural and artifical factors to the upper reaches of Wei River. Journal of Natural Resources, 21(6): 981-990. (in Chinese)

[56]
Wei B, Zhang D Y, Wang G Q, et al. 2023. Experimental warming altered plant functional traits and their coordination in a permafrost ecosystem. New Phytologist, 240(5): 1802-1816.

[57]
Wei W, Guo Z C, Xie B B, et al. 2019. Spatiotemporal evolution of environment based on integrated remote sensing indexes in arid inland river basin in Northwest China. Environmental Science and Pollution Research, 26(13): 13062-13084.

[58]
Winkler K, Fuchs R, Rounsevell M, et al. 2021. Global land use changes are four times greater than previously estimated. Nature Communications, 12(1): 2501, doi: 10.1038/s41467-021-22702-2.

[59]
Wu X C, Liu H Y, Li X Y, et al. 2017. Higher temperature variability reduces temperature sensitivity of vegetation growth in Northern Hemisphere. Geophysical Research Letters, 44(12): 6173-6181.

[60]
Xue S Y, Wu G C. 2024. Sensitivities of vegetation gross primary production to precipitation frequency in the Northern Hemisphere from 1982 to 2015. Remote Sensing, 16(1): 21, doi: 10.3390/rs16010021.

[61]
Yang X M, Liu S Z, Yang T B, et al. 2016. Spatial-temporal dynamics of desert vegetation and its responses to climatic variations over the last three decades: a case study of Hexi region in Northwest China. Journal of Arid Land, 8(4): 556-568.

[62]
Yang Y T, Roderick M L, Guo H, et al. 2023. Evapotranspiration on a greening Earth. Nature Reviews Earth & Environment, 4(9): 626-641.

[63]
Yin J, Gao Y, Chen R S, et al. 2023a. Flash floods: why are more of them devastating the world’s driest regions? Nature, 615(7951): 212-215.

[64]
Yin J B, Gentine P, Slater L, et al. 2023b. Future socio-ecosystem productivity threatened by compound drought-heatwave events. Nature Sustainability, 6(3): 259-272.

[65]
Zeng X, Hu Z M, Chen A P, et al. 2022. The global decline in the sensitivity of vegetation productivity to precipitation from 2001 to 2018. Global Change Biology, 28(22): 6823-6833.

[66]
Zhang C X, Wang X M, Li J C, et al. 2011. Roles of climate changes and human interventions in land degradation: a case study by net primary productivity analysis in China's Shiyanghe Basin. Environmental Earth Sciences, 64(8): 2183-2193.

[67]
Zhang X L, Xiao W H, Wang Y C, et al. 2019. Spatial-temporal changes in NPP and its relationship with climate factors based on sensitivity analysis in the Shiyang River Basin. Journal of Earth System Science, 129(1): 24, doi: 10.1007/s12040-019-1267-6.

[68]
Zhang Y, Gentine P, Luo X Z, et al. 2022. Increasing sensitivity of dryland vegetation greenness to precipitation due to rising atmospheric CO2. Nature Communications, 13(1): 4875, doi: 10.1038/s41467-022-32631-3.

[69]
Zhao M S, Running S W. 2010. Drought-induced reduction in global terrestrial net primary production from 2000 through 2009. Science, 329(5994): 940-943.

[70]
Zhou W, Li J L, Mu S J, et al. 2014. Effects of ecological restoration-induced land-use change and improved management on grassland net primary productivity in the Shiyanghe River Basin, north-west China. Grass and Forage Science, 69(4): 596-610.

[71]
Zhou W, Gang C C, Zhou F C, et al. 2015. Quantitative assessment of the individual contribution of climate and human factors to desertification in northwest China using net primary productivity as an indicator. Ecological Indicators, 48: 560-569.

[72]
Zhu Z C, Piao S L, Myneni R B, et al. 2016. Greening of the Earth and its drivers. Nature Climate Change, 6(8): 791-795.

[73]
Zohner C M, Mirzagholi L, Renner S S, et al. 2023. Effect of climate warming on the timing of autumn leaf senescence reverses after the summer solstice. Science, 381(6653): eadf5098, doi: 10.1126/science.adf509.

Options
文章导航

/

〈 〉