EN中文

Journal of Arid Land >

2024 , Vol. 16 >Issue 6: 779 - 797

DOI: https://doi.org/10.1007/s40333-024-0077-0

  • LI Chuanhua , 1, * ,
  • ZHANG Liang 1 ,
  • WANG Hongjie 2 ,
  • PENG Lixiao 1 ,
  • YIN Peng 1 ,
  • MIAO Peidong 1
展开

收稿日期: 2023-12-27

  修回日期: 2024-04-25

  录用日期: 2024-04-29

  网络出版日期: 2025-08-13

收起

Influence of vapor pressure deficit on vegetation growth in China

  • LI Chuanhua , 1, * ,
  • ZHANG Liang 1 ,
  • WANG Hongjie 2 ,
  • PENG Lixiao 1 ,
  • YIN Peng 1 ,
  • MIAO Peidong 1
Expand
  • 1College of Geography and Environmental Science, Northwest Normal University, Lanzhou 730070, China
  • 2Hebei First Surveying and Mapping Institute, Shijiazhuang 050000, China
*LI Chuanhua (E-mail: )

Received date: 2023-12-27

  Revised date: 2024-04-25

  Accepted date: 2024-04-29

  Online published: 2025-08-13

Fold

本文引用格式

LI Chuanhua , ZHANG Liang , WANG Hongjie , PENG Lixiao , YIN Peng , MIAO Peidong . [J]. Journal of Arid Land, 2024 , 16(6) : 779 -797 . DOI: 10.1007/s40333-024-0077-0

Abstract

Vapor pressure deficit (VPD) plays a crucial role in determining plant physiological functions and exerts a substantial influence on vegetation, second only to carbon dioxide (CO2). As a robust indicator of atmospheric water demand, VPD has implications for global water resources, and its significance extends to the structure and functioning of ecosystems. However, the influence of VPD on vegetation growth under climate change remains unclear in China. This study employed empirical equations to estimate the VPD in China from 2000 to 2020 based on meteorological reanalysis data of the Climatic Research Unit (CRU) Time-Series version 4.06 (TS4.06) and European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis 5 (ERA-5). Vegetation growth status was characterized using three vegetation indices, namely gross primary productivity (GPP), leaf area index (LAI), and near-infrared reflectance of vegetation (NIRv). The spatiotemporal dynamics of VPD and vegetation indices were analyzed using the Theil-Sen median trend analysis and Mann-Kendall test. Furthermore, the influence of VPD on vegetation growth and its relative contribution were assessed using a multiple linear regression model. The results indicated an overall negative correlation between VPD and vegetation indices. Three VPD intervals for the correlations between VPD and vegetation indices were identified: a significant positive correlation at VPD below 4.820 hPa, a significant negative correlation at VPD within 4.820-9.000 hPa, and a notable weakening of negative correlation at VPD above 9.000 hPa. VPD exhibited a pronounced negative impact on vegetation growth, surpassing those of temperature, precipitation, and solar radiation in absolute magnitude. CO2 contributed most positively to vegetation growth, with VPD offsetting approximately 30.00% of the positive effect of CO2. As the rise of VPD decelerated, its relative contribution to vegetation growth diminished. Additionally, the intensification of spatial variations in temperature and precipitation accentuated the spatial heterogeneity in the impact of VPD on vegetation growth in China. This research provides a theoretical foundation for addressing climate change in China, especially regarding the challenges posed by increasing VPD.

Key words： vapor pressure deficit (VPD); near-infrared reflectance of vegetation (NIRv); leaf area index (LAI); gross primary productivity (GPP); Climatic Research Unit (CRU) Time-Series version 4.06 (TS4.06); European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis 5 (ERA-5); climate change

1 Introduction

Since the Industrial Revolution, the global average temperature has significantly increased, leading to a continuous acceleration of global warming (Jaramillo et al., 2010). Consequently, the frequency and intensity of extreme weather events, such as droughts and heatwaves, have escalated, profoundly impacting global terrestrial ecosystems (Reichstein et al., 2013). Current research on the effects of climate change on ecosystems predominantly focuses on factors such as temperature, precipitation, and solar radiation (Kong et al., 2017; Tu et al., 2021; Li et al., 2022), paying little attention to vapor pressure deficit (VPD). VPD refers to the difference between the saturated vapor pressure and the actual vapor pressure in the air at a given temperature, serving as a robust indicator of atmospheric water demand (Park et al., 2013; Novick et al., 2016; Rigden and Salvucci, 2017). VPD affects the stomatal conductance of plant leaves, thereby influencing vegetation transpiration, with crucial implications for the structure and functioning of ecosystems. Some studies indicate that VPD-induced water stress in vegetation inhibits photosynthesis, making VPD a highly important atmospheric factor affecting vegetation, second only to carbon dioxide (CO2) (Keenan et al., 2013; Yuan et al., 2019; Mathias and Thomas, 2021). Recent research further confirms the inhibitory effect of VPD on global vegetation (Li et al., 2023). Grasslands in the United States exhibit a higher sensitivity to VPD than to precipitation (Konings et al., 2017), and the increase of VPD exerts a stronger negative impact on tree growth compared to temperature rise (Eamus et al., 2013). Although VPD plays a critical role in regulating the carbon and water cycles of terrestrial ecosystems, its impact on vegetation growth varies with vegetation types and climatic conditions, changing in response to environmental shifts (Li et al., 2023). However, the influence of VPD on vegetation growth under climate change remains unclear in China. Therefore, it is essential to conduct research on the relationship between VPD and vegetation growth in China.
At present, VPD is estimated using two primary approaches, namely spatial interpolation based on ground-based measurement data (Daly et al., 2015) and remote sensing inversion (Cai et al., 2021). However, both approaches have various challenges. The ground-based approach may not offer sufficient precision because of the sparse and uneven distribution of meteorological stations (Hashimoto et al., 2008; Şahin et al., 2013). The remote sensing inversion faces challenges such as difficulty in decoupling variables, complex atmospheric correction, and low spatiotemporal resolution. Accuracy can be improved using meteorological reanalysis data, which combine multiple influencing factors, utilize relevant algorithms for remote sensing observations, and can effectively represent measured data (Xu et al., 2001; Chen et al., 2013; Chang, 2018; Liu et al., 2019). Based on meteorological reanalysis data including temperature, humidity, and dew point temperature, Yuan et al. (2019) used the vapor pressure formulas proposed by Buck et al. (1981) to estimate VPD with high accuracy. The VPD estimation method has been widely used in meteorological and ecological studies (Willett and Sherwood, 2012; Yu et al., 2014; Cucchi et al., 2020).
Vegetation indices are important tools for monitoring and evaluating the vegetation growth status using remote sensing techniques (Cui et al., 2016). Various vegetation features are characterized by different vegetation indices, and a single index cannot comprehensively capture the vegetation growth status (Viña et al., 2011). Specifically, gross primary productivity (GPP) represents the total amount of carbon absorbed by terrestrial ecosystems through photosynthesis and is a critical factor in carbon cycling and ecosystem ecological processes (Huete, 2012). Leaf area index (LAI) indicates the total leaf area of plants per unit land area, and its magnitude is directly related to plant productivity (Wang et al., 2005). Near-infrared reflectance of vegetation (NIRv) is a novel satellite product that effectively separates vegetation signals from non-vegetative backgrounds, eliminating non-vegetative components in pixel data (Badgley et al., 2017).
Considering the abovementioned factors, this study estimated the VPD in China based on multiple meteorological reanalysis data and characterized vegetation growth status using three vegetation indices, namely GPP, LAI, and NIRv. The spatiotemporal dynamics of VPD in China from 2000 to 2020 and its impact on vegetation growth were explored. The scientific questions addressed in this study are as follows: (1) what is the influence of VPD on vegetation growth in China at different time intervals from a scientific perspective? and (2) how does the influence of VPD on vegetation growth change with global warming? The findings are expected to provide deeper insight into the impact of climate change on vegetation growth in China.

2 Materials and methods

2.1 Study area

China is situated in the eastern part of the Eurasian continent and along the western coast of the Pacific Ocean, with a gradual decrease in terrain from west to east (Fig. 1). Approximately two-thirds of the land area of China is comprised of mountains, highlands, and hills. The average annual precipitation in China is around 600.00 mm, and the distribution of precipitation generally decreasing from southeast to northwest. The average annual temperature is approximately 7.29°C. China exhibits a diverse range of climate types, with the eastern region experiencing a monsoon climate and the northwestern region characterized by a continental desert climate within the westerly belt.
Fig. 1 Overview of China based on the digital elevation model (DEM) and the spatial distribution of meteorological stations. DEM data were obtained from the Geospatial Data Cloud (http://www.gscloud.cn) with a spatial resolution of 90 m. Note that this map is based on the standard map (GS(2023)2767) of the Map Service System (http://bzdt.ch.mnr.gov.cn/) marked by the Ministry of Natural Resources of the People's Republic of China, and the base map has not been modified.

2.2 Data collection and preprocessing

2.2.1 Meteorological reanalysis data

Meteorological reanalysis data utilized in this study included the Climatic Research Unit (CRU) Time-Series version 4.06 (TS4.06) (Harris et al., 2020) and European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis 5 (ERA-5) (Dee et al., 2011). The CRU TS4.06 dataset, reconstructed by the CRU at the British Atmospheric Data Centre (https://crudata.uea.ac.uk/cru), integrates data from various renowned databases. Covering the period from 1901 to 2020, the dataset has a spatial resolution of 0.50°×0.50° and a temporal resolution of one month. In this study, near-surface air temperature and actual vapor pressure data were obtained from this dataset. The ERA-5 dataset, provided by the ECMWF (https://cds.climate.copernicus.eu), offers monthly averaged reanalysis data spanning from 1979 to the present. This dataset has a spatial resolution of 0.10°×0.10° and a temporal resolution of one month. This study incorporated near-surface air temperature, dew point temperature, precipitation, and solar radiation data from the ERA-5 dataset.

2.2.2 Meteorological station observational data

Meteorological station observational data were obtained from the Global Summary of the Day (GSOD) dataset provided by the National Oceanic and Atmospheric Administration (NOAA) of the United States (https://www.ncei.noaa.gov). This dataset covers approximately 27,000 stations globally, offering daily meteorological element observations from 1929 to 2018. It has been widely utilized in studies on global and regional large-scale vegetation dynamics (Goetz et al., 2006; Wang et al., 2011). In this study, a subset of stations within China was utilized, comprising average air temperature, dew point temperature, and station elevation data of 299 meteorological stations. These stations have been proven to be highly reliable (Sun et al., 1998), and their distribution is illustrated in Figure 1.

2.2.3 CO2 data

CO2 data were sourced from the Atmospheric Infrared Sounder (AIRS) and Orbiting Carbon Observatory-2 (OCO-2) datasets released by the National Aeronautics and Space Administration (NASA) (https://disc.gsfc.nasa.gov). AIRS_CO2, a high-resolution infrared spectrometer with a wavelength range of 3.7-15.4 μm and 2378 spectral channels, is considered one of the most reliable global CO2 observation products (Engelen and Mcnally, 2005). The OCO-2_CO2 data are widely applied, and their consistency and reliability in atmospheric CO2 monitoring have been proven through tests (Kiel et al., 2019). The AIRS_CO2 and OCO-2_CO2 data have spatial resolutions of 2.00°×2.50° and 2.25°×1.29°, and have temporal resolutions of one month and 16 d, respectively.

2.2.4 Vegetation indices

NIRv was obtained by multiplying near-infrared reflectance (NIR) with the normalized differential vegetation index (NDVI), through which issues arising from non-vegetation information components in pixels can be mitigated. NDVI and NIR data were obtained from the MOD13C2 dataset available on the Land Processes Distributed Active Archive Center (LP DAAC) (https://lpdaac.usgs.gov). This dataset has been extensively used for monitoring global or regional-scale vegetation dynamics (Piao and Fang, 2001). The MOD13C2 dataset has a spatial resolution of 0.05°×0.05° and a temporal resolution of one month.
LAI was derived from the Global LAnd Surface Satellite (GLASS) data product downloaded from the National Earth System Science Data Center (http://www.geodata.cn). This product is based on artificial neural networks, providing a long time series with high accuracy. It can effectively reflect the overall vegetation characteristics for long-term sequential analysis (Xiao et al., 2017; Xu et al., 2018). This product has a spatial resolution of 0.05°×0.05° and a temporal resolution of one month.
GPP was obtained from the Global OCO-2 Solar-Induced Fluorescence (SIF) (GOSIF) GPP product dataset (https://globalecology.unh.edu), which exhibits a high correlation with GPP observed at 91 global flux tower sites (Li et al., 2019). The data cover the period from 2000 to 2020, with a spatial resolution of 0.05°×0.05° and a temporal resolution of 8 d.

2.2.5 Aridity index (AI)

AI data were obtained from the National Tibetan Plateau Data Center (https://data.tpdc.ac.cn/). The data have a spatial resolution of 0.0083°×0.0083° and a monthly temporal resolution. AI serves as an indicator of the degree of dryness or wetness in a region. In general, regions can be classified as wet (AI<1.0), semi-humid (1.0≤AI<1.5), semi-arid (1.5≤AI<4.0), and arid (AI≥4.0) zones based on AI (Peng et al., 2017).

2.2.6 Data preprocessing

As the application of data with different spatial resolutions leads to various scientific problems, in this study, we employed the resampling method for data consistenty processing, which is a widely accepted method in the field (Accadia et al., 2003; Obu et al., 2019; Fan and Bai, 2021). To ensure a uniform temporal resolution, we used a synthesis method to standardize the data to a monthly temporal resolution.

2.3 Methods

2.3.1 Calculation of VPD

VPD was estimated using the vapor pressure formulas proposed by Buck et al. (1981), incorporating terrain and enhancement factors for correction. The formulas are as follows:
V P D = S V P − A V P
SVP = 6.112 × f w × e 17.67 × T a T a + 243.5
AVP = 6.112 × f w × e 17.67 × T d T d + 243.5
f w = 1 + 7 × 10 − 4 + 3.46 × 10 − 6 × P mst
P mst = P msl × T a + 273.16 T a + 273.16 + 0 . 0 0 6 5 × z 5.625
where SVP represents the saturated vapor pressure (hPa); AVP stands for the actual vapor pressure (hPa); fw is the enhancement factor, a slight correction introduced to handle moist air rather than pure water vapor, and it is a weak function of temperature and pressure; Ta is the near-surface air temperature (°C); Td is the dew point temperature (°C); Pmst represents the atmospheric pressure (hPa); Pmsl is the mean sea level pressure fixed at 1013.250 hPa; and z is the elevation (m). Data of the mentioned indicators were derived from the CRU TS4.06 and ERA-5 datasets. Elevation data were from the digital elevation model (DEM) data of China obtained from the Geospatial Data Cloud (http://www.gscloud.cn) with a spatial resolution of 90 m.
Datasets of VPD estimated from the ERA-5 and CRU TS4.06 datasets were denoted as ERA_VPD and CRU_VPD, respectively (Fig. 2). The ranges and spatial distribution patterns of the two VPD datasets were largely consistent, with multi-year averages of 5.780 hPa (ERA_VPD) and 5.580 hPa (CRU_VPD) during 2000-2020. A high and very close correlation was observed between estimated and observed values, with ERA_VPD exhibiting slightly higher accuracy than CRU_VPD (Fig. 3). Accordingly, ERA_VPD was used for subsequent analyses in further investigation.
Fig. 2 Spatial distribution of ERA_VPD (a) and CRU_VPD (b) in China in 2020. VPD, vapor pressure deficit; ERA_VPD, vapor pressure deficit estimated from the European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis 5 (ERA-5); CRU_VPD, vapor pressure deficit estimated from the Climatic Research Unit (CRU) Time-Series version 4.06 (CRU TS4.06). Note that the figures are based on the standard map (GS(2023)2767) of the Map Service System (http://bzdt.ch.mnr.gov.cn/) marked by the Ministry of Natural Resources of the People's Republic of China, and the base map has not been modified.
Fig. 3 Accuracy validation of ERA_VPD (a) and CRU_VPD (b) using the meteorological station observational data. RMSE, root mean square error. Shaded areas indicate a 95% confidence interval.

2.3.2 Theil-Sen median trend analysis and Mann-Kendall test

The Theil-Sen median trend analysis does not require the samples to follow a specific distribution and is robust against outliers, exhibiting strong resistance to measurement errors or outlier data (Liu et al., 2015a; Guo et al., 2018). Therefore, we employed the Theil-Sen median trend analysis, coupled with the Mann-Kendall test to analyze the trends of VPD and vegetation indices as well as the significance test of the trends in this study. The calculation formula is as follows:
α = mean x a − x b a − b ∀ b > a
where α is the estimated magnitude of the trend slope in the time series of the data; and xa and xb are the ath and bth data points in the time series, respectively. α>0 indicates that the time series shows an upward trend; α<0 indicates that the time series shows a downward trend.
The Mann-Kendall test is capable of determining the significance of the trend that is not affected by outliers, making it well-suited for trend analysis and examination of long time series (Liu et al., 2015b; Lou et al., 2021). The calculation formulas are as follows:
Z = S – 1 var ( S ) S > 0 0 S = 0 S – 1 var ( S ) S < 0
var ( S ) = n × n – 1 × 2 n – 5 18
S = ∑ i = 1 n − 1 ∑ j = i + 1 n s g n ( x j – x i )
s g n ( x j – x i ) = 1 0 − 1 x j – x i > 0 x j – x i = 0 x j – x i < 0
where Z is the standardized test statistic; S is the test statistic; var(S) is the variance; n is the length of the time series; and xj and xi are the jth and ith data points in the time series, respectively. Z>0 indicates an increasing trend in the time series, Z<0 indicates a decreasing trend, and Z=0 indicates no trend in the time series.
The method of determining the significance of the trend is shown in Table 1.
Table 1 Categories of the Mann-Kendall test trend
α Z Change trend
α>0 |Z|≥2.56 Highly significant increasing
α>0 1.96≤|Z|<2.56 Significant increasing
α=0 |Z|<1.96 Not significant change
α<0 1.96≤|Z|<2.56 Significant decreasing
α<0 |Z|≥2.56 Highly significant decreasing

Note: α is estimated magnitude of the trend slope in the time series of the data; Z is the standardized test statistic.

2.3.3 Calculation of relative contribution of each meteorological factor to vegetation growth

A multiple linear regression model was established with vegetation indices (NIRv, LAI, and GPP) as dependent variables and meteorological factors (VPD, temperature, precipitation, solar radiation, and CO2) as independent variables. The impact of each meteorological factor on vegetation growth was determined by calculating their relative contribution to the change in each vegetation index.
Firstly, standard scores were applied to standardize each variable, as shown in Equation 11:
X s = X k − X ¯ σ
where Xs is the standardized value; Xk is the variable value;
X ¯
 is the mean value of the variable; and σ is the standard deviation of the variable.
Secondly, a multiple linear regression model was developed to quantify the sensitivity of the vegetation indices to the meteorological factors. The standardized annual average values of vegetation indices (NIRv, LAI, and GPP) from 2000 to 2020 were calculated using the formula as follows:
y = β 1 ( V P D ) + β 2 ( T M P ) + β 3 ( P R E ) + β 4 ( S S R ) + β 5 ( C O 2 ) + ε
where y denotes the standardized annual average value of each vegetation index; VPD, TMP, PRE, SSR, and CO2 denote the standardized annual mean values of VPD, temperature, precipitation, solar radiation, and CO2, respectively; β1-β5 are the partial regression coefficients of the meteorological factors; and ε is a constant term.
Thirdly, the relative contribution of each meteorological factor was calculated using the formula as follows (Xie et al., 2020):
η i = β i β 1 + β 2 + β 3 + β 4 + β 5
where ηi represents the relative contribution of meteorological factor i to the change in each vegetation index; and βi is the partial regression coefficient of meteorological factor i.

3 Results

3.1 Spatiotemporal dynamics of VPD

From 2000 to 2020, the VPD in China exhibited a fluctuating upward trend at a rate of 0.012 hPa/a (P<0.01), as shown in Figure 4a. The most rapid growth occurred between 2003 and 2009, reaching approximately 9.6 times the average growth rate during 2000-2020. The highest VPD value was recorded in 2007, at around 6.010 hPa, while the lowest was observed in 2003, at approximately 5.160 hPa. Additionally, temperature and precipitation increased from 2000 to 2020 at rates of 0.021°C/a and 2.572 mm/a (P<0.01), respectively (Fig. 4b and c). Specifically, precipitation in China experienced a significant increase trend at a rate of 6.203 mm/a after 2010.
Fig. 4 Trends in annual average VPD (a), annual average temperature (b), and annual precipitation (c) in China during 2000-2020. Shaded areas indicate a 95% confidence interval.
Significant variations were observed in the spatial distribution of the multi-year average VPD in China from 2000 to 2020, ranging from 0.000 to 21.241 hPa (Fig. 5a). Low VPD values were predominantly distributed in the Tibetan Plateau and Northeast China, where annual average temperature and annual evaporation are low, and the lowest value was observed in the Yushu Tibetan Autonomous Prefecture of Qinghai Province. High VPD values were concentrated in the arid areas of Northwest China, which are characterized by low precipitation and high evaporation, with the highest value observed in the Turpan region of Xinjiang Uygur Autonomous Region. From north to south, VPD exhibited an initial gradual increase, reaching its peak of approximately 9.030 hPa around 40°N, corresponding to the arid areas of China. Subsequently, VPD gradually decreased with decreasing latitude. VPD remained low and stable within the region between 28°N and 36°N, which includes the Tibetan Plateau. Beyond this range (28°-36°N), VPD showed further increase. From west to east, VPD reached its highest value at approximately 8.130 hPa around 78°E, corresponding to the heart of the Taklimakan Desert, which is the largest desert in China. VPD showed a decrease with increasing longitude under the influence of the low-value zone of VPD in the Tibetan Plateau. Near 95°E, VPD increased due to the high VPD in the Qaidam Basin. Approaching 115°E, closer to the ocean, VPD gradually decreased because atmospheric vapor content gradually increases.
Fig. 5 Spatial variations of multi-year average VPD (a), rate of change (indicated by α-value) in annual average VPD (b), and change trend of annual average VPD (c) in China during 2000-2020. Figure 5a2 and a3 show variations in the mean values of multi-year average VPD along latitude and longitude. Note that the figures are based on the standard map (GS(2023)2767) of the Map Service System (http://bzdt.ch.mnr.gov.cn/) marked by the Ministry of Natural Resources of the People's Republic of China, and the base map has not been modified.
Between 2000 and 2020, VPD exhibited an increasing trend in approximately 68.80% of the regions in China, as shown in Figure 5b. Regions with a higher rate of change (α>0.006 hPa/a) accounted for about 53.71%, primarily concentrated in the North China Plain, Yungui Plateau, and the northern part of Xinjiang Uygur Autonomous Region. Among them, the western part of the North China Plain exhibited the fastest growth, with an average α-value of approximately 0.014 hPa/a, and the multi-year average annual VPD in this region ranged from 7.000 to 20.000 hPa. Approximately 31.20% of the regions showed a downward trend of VPD, with regions having a higher decline rate (α< -0.003 hPa/a) accounting for approximately 25.23%. These regions were mainly located in the southern and northeastern parts of China, where VPD ranged from 3.000 to 7.500 hPa. In particular, the most rapid decline was observed in Guangxi Zhuang Autonomous Region and Guangdong Province, with an average α-value of approximately -0.008 hPa/a.
Variations in VPD were not statistically significant in approximately 87.49% of the regions in China, and only 12.51% of the regions passed the significance test (P<0.05; Fig. 5c). Among them, regions with highly significant increasing in VPD accounted for approximately 4.34%, and regions with significant decreasing in VPD accounted for approximately 7.25%. These significant trends were mainly observed in the North China Plain, the western part of the Tibetan Plateau, and the eastern part of Yunnan Province. Very few regions showed highly significant decreasing in VPD, and regions with significant decreasing in VPD accounted for approximately 0.90%. These significant downward trends were primarily distributed in the central part of Northeast China, the northeastern part of the Tibetan Plateau, and the southern part of Guangxi Zhuang Autonomous Region and Guangdong Province.

3.2 Spatiotemporal dynamics of vegetation indices

The three vegetation indices (NIRv, LAI, and GPP) exhibited highly consistent spatial patterns, all showing a decreasing distribution pattern from southeast to northwest (Fig. 6). During the study period, all the three vegetation indices showed a significant upward trend (Fig. 7), indicating a continuous improvement in vegetation coverage and good vegetation growth in China. The multi-year average values were 0.090 for NIRv, 0.790 for LAI, and 787.210 g C/(m2•a) for GPP. The average growth rates were 0.008/a for NIRv, 0.007/a for LAI, and 9.720 g C/(m2•a) for GPP. When using their standardized values, the three indices exhibited relatively similar growth rates, with GPP showing the fastest increase rate and NIRv exhibiting the slowest rate.
Fig. 6 Spatial distributions of multi-year average NIRv (a), multi-year average LAI (b), and multi-year average GPP (c) in China during 2000-2020. NIRv, near-infrared reflectance of vegetation; LAI, leaf area index; GPP, gross primary productivity. Note that the figures are based on the standard map (GS(2023)2767) of the Map Service System (http://bzdt.ch.mnr.gov.cn/) marked by the Ministry of Natural Resources of the People's Republic of China, and the base map has not been modified.
Fig. 7 Trends in NIRv (a), LAI (b), and GPP (c) in China during 2000-2020. Shaded areas indicate a 95% confidence interval.
From 2000 to 2020, the trends in the three vegetation indices showed generally consistent spatial distributions (Fig. 8). The area proportions of regions with increases in NIRv, LAI, and GPP were 90.29%, 80.01%, and 83.68%, respectively. The area proportions of regions with significant increasing in NIRv, LAI, and GPP were 64.09%, 49.03%, and 53.96%, respectively. The fastest increases were observed in the hilly areas of the Guangxi Zhuang Autonomous Region and Guangdong Province, with average α-values of 0.004/a, 0.035/a, and 47.240 g C/(m2•a) in terms of NIRv, LAI, and GPP, respectively. Regions with significant decreasing in NIRv, LAI, and GPP accounted for 1.16%, 0.83%, and 3.42%, respectively, mainly concentrated in the Qaidam Basin, Tibetan Plateau, and some parts of Northeast China. The Qaidam Basin experienced the fastest decrease in NIRv, LAI, and GPP, with average α-values of -0.000/a, -0.009/a, and -12.150 g C/(m2•a), respectively.
Fig. 8 Spatial variations in rates of change (indicated by α-values) of NIRv (a), LAI (b), and GPP (c) in China during 2000-2020. Black markers indicate the regions with statistically significant trends (P<0.05). Note that the figures are based on the standard map (GS(2023)2767) of the Map Service System (http://bzdt.ch.mnr.gov.cn/) marked by the Ministry of Natural Resources of the People's Republic of China, and the base map has not been modified.

3.3 Influence of VPD on vegetation growth

3.3.1 Correlations between VPD and vegetation indices

From 2000 to 2020, the relationship between VPD and vegetation indices displayed an initial significant positive correlation, and shifted to a significant negative correlation beyond a threshold (4.820 hPa). Subsequently, after surpassing another threshold (9.000 hPa), the correlation became non-significantly negative (Fig. 9a). The relationship between VPD and rates of change in vegetation indices from 2000 to 2020 (Fig. 9b) was analogous to that between VPD and vegetation indices (Fig. 9a), albeit with differing thresholds. When VPD was below 4.820 hPa, vegetation indices exhibited a rapid increase trend with increasing VPD, indicating a significant positive correlation (Fig. 9a). In the VPD range of 4.820-9.000 hPa, vegetation indices markedly decreased with increasing VPD (Fig. 9a). After VPD exceeded 9.000 hPa, vegetation indices did not exhibit a significant decrease trend with increasing VPD, and the correlation between VPD and rates of change in vegetation indices was noticeably attenuated. It should be noted that this analysis did not account for the impact of human activities.
Fig. 9 Correlations between VPD and vegetation indices (a) as well as between VPD and rates of change (indicated by α-values) of vegetation indices (b). VPD, NIRv, LAI, and GPP values in the figures were multi-year average values during 2000-2020.
China can be divided into four climatic zones according to the AI: wet, semi-humid, semi-arid, and arid zones, with VPD values of 4.280, 3.331, 4.830, and 9.480 hPa, respectively. Regions with VPD exceeding 8.000 hPa primarily corresponded to arid zone. Figure 10 illustrates the correlation between VPD and vegetation indices in each climatic zone. In wet zone, the correlations between VPD and vegetation indices shifted from positive to negative beyond VPD of 4.320 hPa. In semi-humid zone, this shift occurred at 4.630 hPa. In semi-arid zone, the correlation shifted from positive to negative beyond VPD of 3.963 hPa, but reversed beyond VPD of 7.910 hPa, with only 8.16% of the semi-arid zone exceeding this threshold, primarily in the North China Plain. In arid zone, the correlation shifted at 6.071 hPa.
Fig. 10 Correlations between VPD and vegetation indices in wet zone (a), semi-humid zone (b), semi-arid zone (c), and arid zone (d) in China. VPD, NIRv, LAI, and GPP values in the figures were multi-year average values during 2000-2020.

3.3.2 Relative contribution of VPD and other meteorological factors to vegetation growth

The spatial distributions of the relative contribution of meteorological factors (VPD, temperature, precipitation, solar radiation, and CO2) to vegetation growth were determined using a multiple linear regression model (Fig. 11). VPD exhibited negative relative contribution to changes in NIRv, LAI, and GPP in most regions, accounting for 57.65%, 60.38%, and 58.38%, respectively.
Fig. 11 Relative contributions of VPD (a-c), temperature (d-f), precipitation (g-i), solar radiation (j-l), and CO2 (m-o) to changes in NIRv, LAI, and GPP in China during 2000-2020. Note that the figures are based on the standard map (GS(2023)2767) of the Map Service System (http://bzdt.ch.mnr.gov.cn/) marked by the Ministry of Natural Resources of the People's Republic of China, and the base map has not been modified.
Regions with significantly negative relative contribution (< -0.300) of VPD to changes in NIRv, LAI, and GPP accounted for 14.34%, 15.25%, and 15.58%, respectively. Regions with positive relative contribution of VPD to changes in NIRv, LAI, and GPP accounted for 42.35%, 39.62%, and 41.62%, respectively; regions with significantly positive relative contribution (>0.300) of VPD to changes in NIRv, LAI, and GPP accounted for 7.67%, 7.27%, and 9.69%, respectively. It can be observed that the three vegetation indices showed generally the same spatial distributions in their responses to various meteorological factors.
Temperature, precipitation, solar radiation, and CO2 all positively contributed to changes in NIRv, LAI, and GPP (Fig. 12a). CO2 exhibited the largest relative contribution, with mean relative contribution values of 0.321, 0.301, and 0.296 to changes in NIRv, LAI, and GPP, respectively. The relative contribution of the remaining meteorological factors could be ranked as follows: solar radiation>temperature>precipitation. Notably, VPD showed a distinct negative relative contribution, with mean relative contribution values of -0.101, -0.084, and -0.089 to changes in NIRv, LAI, and GPP, respectively. The absolute value of the relative contribution of VPD was approximately 30.00% that of CO2.
Fig. 12 Relative contributions of various meteorological factors to vegetation indices during 2000-2020 (a), 2000-2010 (b), and 2011-2020 (c). Solid circles represent mean values. Upper and lower whiskers indicate the confidence intervals of 5% and 95%, respectively.
By dividing the study period into two phases: 2000-2010 (Phase 1) and 2011-2020 (Phase 2), we calculated the relative contribution values of each meteorological factor to vegetation growth separately (Fig. 12b and c). In Phase 1, the mean relative contribution values of VPD to changes in NIRv, LAI, and GPP were -0.074, -0.087, and -0.082, respectively. In Phase 2, these values became -0.042, -0.048, and -0.058, respectively, indicating significant reductions in negative relative contributions. Positive relative contributions from CO2, temperature, and solar radiation decreased, while positive contribution from precipitation increased.

4 Discussion

4.1 Spatiotemporal dynamics of VPD

The spatial distribution of VPD is primarily influenced by climate types and topography (Zhang et al., 2017). VPD reached its maximum in arid areas, such as the Northwest China, where precipitation is scarce and evaporation is high. Conversely, in high-altitude areas such as the Tibetan Plateau and mountainous regions, VPD reached the minimum values owing to low temperature and evapotranspiration. Despite higher temperature in the eastern monsoon region of China, VPD remained at an intermediate level due to ample precipitation. The simulated spatial distribution of VPD in this study aligns with the fundamental spatial characteristics of VPD presented by Yuan et al. (2019) and Liu et al. (2024).
Between 2000 and 2020, the noticeable increase in VPD in China can be attributed to two primary factors. Firstly, global warming has led to a significant increase in air temperature and a consequent increase in VPD (Gamelin et al., 2022). Secondly, the decrease in actual vapor pressure has also led to an increase in VPD by a consequence of various hydrological processes, including diminished oceanic evaporation and reduced precipitation (Ficklin and Novick, 2017). The decline in oceanic evaporation, especially after 1998, has been a crucial mechanism contributing to the reduction in actual vapor pressure over land, exerting a considerable influence on precipitation patterns (Fu and Feng, 2014). Global oceanic evaporation has markedly decreased since 1998, diminishing by approximately 2.08 mm on an annual scale and contributing to the continuous increase of VPD (Yuan et al., 2019). During the period of 2000-2009, China generally experienced a slow increase in annual precipitation (Liu et al., 2020), and notable decreases were observed in arid areas, particularly in the Northwest China (Xu et al., 2020). Due to this decrease in precipitation, atmospheric vapor content decreased at a rate of 0.120 g/(m3•10a), directly leading to a rapid increase in VPD (Zhang et al., 2018). Subsequently, after 2010, there was a significant increase in annual precipitation (Liu et al., 2020), contributing to a deceleration in the increase of VPD. This trend aligns with the interannual variability of VPD in China simulated by Yuan et al. (2021).

4.2 Correlation between VPD and vegetation indices

In general, VPD showed distinct correlations with vegetation indices across three intervals: <4.820, 4.820-9.000, and >9.000 hPa, which was similarly manifested across different climatic zones. Regarding the first interval, the positive correlations were predominantly observed in the southeastern part of the Tibetan Plateau, the Three-river Source area, and Northeast China. These regions are characterized by low temperature and evapotranspiration. An increase in VPD is usually accompanied by high temperature, leading to a significant rise in the vegetation indices with increasing VPD (Ding et al., 2018). Regarding the second interval, the negative correlations were mainly observed in the North China Plain and Southeast China, which are characterized by flat terrain, abundant water sources, and higher vegetation productivity (Ma et al., 2020). The hydrothermal characteristics of the North China Plain and Southeast China are also the reason why the VPD and vegetation indices in semi-arid zone changed from negative correlation to positive correlation at the threshold of 7.910 hPa (Fig. 10c). With increasing VPD, atmospheric aridity becomes more severe, triggering stomatal closure in vegetation to prevent excessive water loss due to the elevated atmospheric demand for evaporation. Consequently, the photosynthetic rates of leaves and canopies decline (Bourbia et al., 2023), resulting in the negative carbon balance, depletion of carbohydrate reserves, and carbohydrate starvation at the tissue level. Additionally, reduced soil moisture supply, combined with high evaporative demand, causes xylem vessel cavitation and rhizosphere desiccation, hindering water flow and causing plant tissues to dry, ultimately leading to plant death (McDowell et al., 2008). Regarding the third interval, the insignificant correlations were predominantly observed in the arid desert regions of Northwest China (with high VPD values), which experience a hot and arid climate. Vegetation in this area has adapted to high VPD, resulting in lower sensitivity to the increase of VPD (Grossiord et al., 2020). Accordingly, the increase in VPD has a relatively small influence on vegetation growth in this region. This pattern agrees with the conclusion that increased VPD can promote stomatal opening on leaf surfaces, facilitating better water and CO2 absorption for photosynthesis. However, when VPD exceeds a certain threshold, it leads to a reduction in stomatal conductance to prevent excessive water loss, reducing photosynthesis and causing a decline in vegetation indices (Jarvis, 1976; Leonardi et al., 2000; Williams and Baeza, 2007).

4.3 Relative contribution of VPD to vegetation growth

In general, VPD showed negative influence on vegetation growth in China, which may be associated with the spatial distribution of vegetation. Regions with notably negative relative contribution of VPD to vegetation growth were primarily concentrated in the Three-North Region and the northern part of the Tibetan Plateau. The Three-North Region, consisting largely of artificially planted forests and experiencing a dry climate with prolonged sunlight exposure (Williams et al., 2007), exhibits poor adaptability to high VPD, and the primary constraints on vegetation growth in this region are insufficient precipitation and excessive evaporation (Wang and Zhaxiyangzong, 2016). In the northern part of the Tibetan Plateau, which is characterized by cold and dry climates with high solar radiation, the predominant vegetation is alpine grassland. An increase in VPD may intensify atmospheric aridity, which restricts photosynthesis and causes a sustained decline in vegetation productivity (Yuan et al., 2019). Regions with significantly positive relative contribution of VPD to vegetation growth were mainly found in the southeastern part of the Tibetan Plateau and the northeastern part of the Sanjiang Plain. The southeastern part of the Tibetan Plateau experiences abundant precipitation, low temperature, and minimal evaporation. The northeastern part of the Sanjiang Plain, with flat terrain and good drainage conditions, exhibits low evaporation and cold and humid climate, resulting in relatively low VPD. The increase in VPD in the two regions promotes water absorption and photosynthesis in vegetation. Moreover, the elevated VPD, accompanied by higher temperature, further stimulates vegetation growth in the two regions (Ding et al., 2018). The significant fertilization effect of atmospheric CO2 significantly enhances vegetation growth in China (Wang et al., 2021). In the absence of other factors, the negative impact of increasing VPD on vegetation growth can offset approximately 30.00% of the fertilization effect of CO2. This is in agreement with the results of previous studies by Yuan et al. (2019). The absolute value of the relative contribution of VPD was higher than those of temperature, precipitation, and solar radiation, indicating that the influence of VPD on vegetation growth cannot be ignored in China. During 2011-2020, the negative impact of VPD on vegetation growth weakened, which can be associated with a slowdown in VPD increase. During this period, precipitation significantly increased (Liu et al., 2020), and the relative contribution of precipitation noticeably increased, which may also lead to a reduction in the relative contribution of VPD.

5 Conclusions

This study utilized remote sensing products to estimate VPD in China from 2000 to 2020. Focusing on the spatiotemporal dynamics of VPD and their relationship with vegetation indices, the relative contribution of VPD to vegetation growth was quantitatively analyzed. Regarding the spatial distribution, the arid areas of Northwest China exhibited the highest VPD, followed by the eastern China, while the Tibetan Plateau, Northeast China, and high-altitude regions exhibited lower VPD. Without considering the impact of human activities, VPD showed a generally negative correlation with vegetation growth. Nevertheless, this relationship was not a simple negative correlation; it can be categorized into three intervals. Specifically, VPD showed a significant positive correlation with vegetation growth below 4.820 hPa. In the range of 4.820-9.000 hPa, VPD showed a significant negative correlation with vegetation growth. Beyond 9.000 hPa, the negative correlation between VPD and vegetation growth diminished significantly. VPD had a significant negative relative contribution to vegetation growth, with the absolute contribution value being higher than those of temperature, precipitation, and solar radiation, but lower than the fertilization effect of CO2. The negative impact of VPD on vegetation growth can offset approximately 30.00% of the positive effect of CO2. With ongoing global warming and increased precipitation, the increase rate of VPD is slowing down, resulting in reduced adverse effects on vegetation growth. As global warming intensifies and precipitation disparities increase in the future, the negative impact of VPD on vegetation growth will likely undergo further differentiated changes. The limitations of this study lie in that it did not consider additional factors such as the interaction between factors, soil moisture, and human activities. In addition, variations across vegetation types were not analyzed. In the future, further research is needed to investigate the influence mechanisms of VPD on vegetation growth in different time intervals to provide a more precise theoretical basis for addressing climate change in China.

Conflict 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.

Acknowledgements

This research was supported by the National Natural Science Foundation of China (42161058). We also thank the editors and anonymous reviewers for their constructive comments.

Author contributions

Conceptualization: LI Chuanhua, ZHANG Liang; Methodology: ZHANG Liang, PENG Lixiao; Formal analysis: LI Chuanhua, ZHANG Liang; Writing - original draft preparation: ZHANG Liang, PENG Lixiao; Writing - review and editing: LI Chuanhua, ZHANG Liang; Funding acquisition: LI Chuanhua; Supervision: WANG Hongjie, YIN Peng, MIAO Peidong. 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]
Accadia C, Mariani S, Casaioli M, et al. 2003. Sensitivity of precipitation forecast skill scores to bilinear interpolation and a simple nearest-neighbor average method on high-resolution verification grids. Weather and Forecasting, 18(5): 918-932.

[2]
Badgley G, Field C B, Berry J A. 2017. Canopy near-infrared reflectance and terrestrial photosynthesis. Science Advances, 3(3): e1602244, doi: 10.1126/sciadv.1602244.

[3]
Bourbia I, Lucani C, Carins-Murphy M R, et al. 2023. In situ characterisation of whole-plant stomatal responses to VPD using leaf optical dendrometry. Plant, Cell and Environment, 46(11): 3273-3286.

[4]
Buck A L. 1981. New equations for computing vapor pressure and enhancement factor. Journal of Applied Meteorology, 20(12): 1527-1532.

[5]
Cai W Y, Ullah S, Yan L, et al. 2021. Remote sensing of ecosystem water use efficiency: A review of direct and indirect estimation methods. Remote Sensing, 13(12): 2393, doi: 10.3390/rs13122393.

[6]
Chang S T. 2018. Characteristics and impact on regional climate of water vapor over the Tibetan Plateau in summer under global warming. MSc Thesis. Lanzhou: Lanzhou University. (in Chinese)

[7]
Chen Z, Gao J, Yang X. 2013. Introduction of IGRA dataset and analysis of its data quality. Journal of Meteorology and Environment, 29(5): 106-111. (in Chinese)

[8]
Cucchi M, Weedon G P, Amici A, et al. 2020. WFDE5: Bias-adjusted ERA5 reanalysis data for impact studies. Earth System Science Data, 12(3): 2097-2120.

[9]
Cui T X, Wang Y J, Sun R, et al. 2016. Estimating vegetation primary production in the Heihe River Basin of China with multi-source and multi-scale data. PLoS ONE, 11(4): e0153971, doi: 10.1371/journal.pone.0153971.

[10]
Daly C, Smith J I, Olson K V. 2015. Mapping atmospheric moisture climatologies across the conterminous United States. PLoS One, 10(10): e0141140, doi: 10.1371/journal.pone.0141140.

[11]
Dee D P, Uppala S M, Simmons A J, et al. 2011. The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quarterly Journal of the Royal Meteorological Society, 137(656): 553-597.

[12]
Ding J Z, Yang T, Zhao Y T, et al. 2018. Increasingly important role of atmospheric aridity on Tibetan alpine grasslands. Geophysical Research Letters, 45(6): 2852-2859.

[13]
Eamus D, Boulain N, Cleverly J, et al. 2013. Global change-type drought-induced tree mortality: Vapor pressure deficit is more important than temperature per se in causing decline in tree health. Ecology and Evolution, 3(8): 2711-2729.

[14]
Engelen R J, Mcnally A P. 2005. Estimating atmospheric CO2 from advanced infrared satellite radiances within an operational four-dimensional variational (4D-var) data assimilation system: Results and validation. Journal of Geophysical Research: Atmospheres, 110(D18): 305, doi: 10.1029/2005JD005982.

[15]
Fan Z M, Bai X Y. 2021. Scenarios of potential vegetation distribution in the different gradient zones of Qinghai-Tibet Plateau under future climate change. Science of the Total Environment, 796: 148918, doi: 10.1016/j.scitotenv.2021.148918.

[16]
Ficklin D L, Novick K A. 2017. Historic and projected changes in vapor pressure deficit suggest a continental-scale drying of the United States atmosphere. Journal of Geophysical Research: Atmospheres, 122(4): 2061-2079.

[17]
Fu Q, Feng S. 2014. Responses of terrestrial aridity to global warming. Journal of Geophysical Research: Atmospheres, 119(13): 7863-7875.

[18]
Gamelin B L, Feinstein J, Wang J L. et al. 2022. Projected U.S. drought extremes through the twenty-first century with vapor pressure deficit. Scientific Reports, 12: 8615, doi: 10.1038/s41598-022-12516-7.

[19]
Goetz S J, Fiske G J, Bunn A G. 2006. Using satellite time-series data sets to analyze fire disturbance and forest recovery across Canada. Remote Sensing of Environment, 101(3): 352-365.

[20]
Grossiord C, Buckley T N, Cernusak L A, et al. 2020. Plant responses to rising vapor pressure deficit. New Phytologist, 226(6): 1550-1566.

[21]
Guo M, Li J, He H S, et al. 2018. Detecting global vegetation changes using Mann-Kendal (MK) trend test for 1982-2015 time period. Chinese Geographical Science, 28(6): 907-919.

[22]
Harris I, Osborn T J, Jones P, et al. 2020. Version 4 of the CRU TS monthly high-resolution gridded multivariate climate dataset. Scientific Data, 7: 109, doi: 10.1038/s41597-020-0453-3.

[23]
Hashimoto H, Dungan J L, White M A, et al. 2008. Satellite-based estimation of surface vapor pressure deficits using MODIS land surface temperature data. Remote Sensing of Environment, 112(1): 142-155.

[24]
Huete A R. 2012. Vegetation indices, remote sensing and forest monitoring. Geography Compass, 6(9): 513-532.

[25]
Jaramillo C, Ochoa D, Contreras L, et al. 2010. Effects of rapid global warming at the Paleocene-Eocene boundary on neotropical vegetation. Science, 330(6006): 957-961.

[26]
Jarvis P G. 1976. The interpretation of the variations in leaf water potential and stomatal conductance found in canopies in the field. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 273(927): 593-610.

[27]
Keenan T F, Hollinger D Y, Bohrer G, et al. 2013. Increase in forest water-use efficiency as atmospheric carbon dioxide concentrations rise. Nature, 499(7458): 324-327.

[28]
Kiel M, O'dell C W, Fisher B, et al. 2019. How bias correction goes wrong: measurement of XCO2 affected by erroneous surface pressure estimates. Atmospheric Measurement Techniques, 12(4): 2241-2259.

[29]
Kong D D, Zhang Q, Huang W L, et al. 2017. Vegetation phenology change in Tibetan Plateau from 1982 to 2013 and its related meteorological factors. Acta Geographica Sinica, 72(1): 39-52. (in Chinese)

[30]
Konings A G, Williams A P, Gentine P. 2017. Sensitivity of grassland productivity to aridity controlled by stomatal and xylem regulation. Nature Geoscience, 10(4): 284-288.

[31]
Leonardi C, Guichard S, Bertin N. 2000. High vapour pressure deficit influences growth, transpiration and quality of tomato fruits. Scientia Horticulturae, 84(3-4): 285-296.

[32]
Li F, Xiao J F, Chen J Q, et al. 2023. Global water use efficiency saturation due to increased vapor pressure deficit. Science, 381(6658): 672-677.

[33]
Li Q Q, Cao Y P, Miao S L. 2022. Spatio-temporal variation in vegetation coverage and its response to climate factors in the Yellow River Basin, China. Acta Ecologica Sinica, 42(10): 4041-4054. (in Chinese)

[34]
Li X, Xiao J F. 2019. A global, 0.05-degree product of solar-induced chlorophyll fluorescence derived from OCO-2, MODIS, and reanalysis data. Remote Sensing, 11(5): 517, doi: 10.3390/rs11050517.

[35]
Liu B, Wang Y, Lou Z S, et al. 2019. The MODIS PWV correction based on CMONOC in Chinese Mainland. Acta Geodaetica et Cartographica Sinica, 48(10): 1207-1215. (in Chinese)

[36]
Liu K, Nie G G, Zhang S. 2020. Study on the spatiotemporal evolution of temperature and precipitation in China from 1951 to 2018. Advances in Earth Science, 35(11): 1113-1126. (in Chinese)

[37]
Liu M, Yang G J, Yuan W P, et al. 2024. Overridingly increasing vegetation sensitivity to vapor pressure deficit over the recent two decades in China. Ecological Indicators, 161: 111977, doi: 10.1016/j.ecolind.2024.111977.

[38]
Liu Y X, Liu X F, Hu Y N, et al. 2015a. Analyzing nonlinear variations in terrestrial vegetation in China during 1982-2012. Environmental Monitoring and Assessment, 187(11): 722, doi: 10.1007/s10661-015-4922-7.

[39]
Liu Y X, Wang Y L, Peng J, et al. 2015b. Correlations between urbanization and vegetation degradation across the world's metropolises using DMSP/OLS nighttime light data. Remote Sensing, 7(2): 2067-2088.

[40]
Lou J P, Xu G Y, Wang Z J, et al. 2021. Multi-year NDVI values as indicator of the relationship between spatiotemporal vegetation dynamics and environmental factors in the Qaidam Basin, China. Remote Sensing, 13(7): 1240, doi: 10.3390/rs13071240.

[41]
Ma Z C, Yu H B, Cao C M, et al. 2020. Spatiotemporal characteristics of fractional vegetation coverage and its influencing factors in China. Resources and Environment in the Yangtze Basin, 29(6): 1310-1321. (in Chinese)

[42]
Mathias J M, Thomas R B. 2021. Global tree intrinsic water use efficiency is enhanced by increased atmospheric CO2 and modulated by climate and plant functional types. Proceedings of the National Academy of Sciences, 118(7): e2014286118, doi: 10.1073/pnas.2014286118.

[43]
Mcdowell N, Pockman W T, Allen C D, et al. 2008. Mechanisms of plant survival and mortality during drought: Why do some plants survive while others succumb to drought? New Phytologist, 178(4): 719-739.

[44]
Novick K A, Ficklin D L, Stoy P C, et al. 2016. The increasing importance of atmospheric demand for ecosystem water and carbon fluxes. Nature Climate Change, 6(11): 1023-1027.

[45]
Obu J, Westermann S, Bartsch A, et al. 2019. Northern hemisphere permafrost map based on TTOP modelling for 2000-2016 at 1 km2 scale. Earth-Science Reviews, 193: 299-316.

[46]
Park Williams A, Allen C D, Macalady A K, et al. 2013. Temperature as a potent driver of regional forest drought stress and tree mortality. Nature Climate Change, 3(3): 292-297.

[47]
Peng S Z, Ding Y X, Wen Z M, et al. 2017. Spatiotemporal change and trend analysis of potential evapotranspiration over the Loess Plateau of China during 2011-2100. Agricultural and Forest Meteorology, 233: 183-194.

[48]
Piao S L, Fang J Y. 2001. Dynamic vegetation cover change over the last 18 years in China. Quaternary Sciences, 21(4): 294-302. (in Chinese)

[49]
Reichstein M, Bahn M, Ciais P, et al. 2013. Climate extremes and the carbon cycle. Nature, 500(7462): 287-295.

[50]
Rigden A J, Salvucci G D. 2017. Stomatal response to humidity and CO2 implicated in recent decline in US evaporation. Global Change Biology, 23(3): 1140-1151.

[51]
Şahin M, Yıldız B Y, Şenkal O, et al. 2013. Estimation of the vapour pressure deficit using NOAA-AVHRR data. International Journal of Remote Sensing, 34(8): 2714-2729.

[52]
Sun H Y, Wang C Y, Niu Z, et al. 1998. Analysis of the vegetation cover change and the relationship between NDVI and environmental factors by using NOAA time series data. National Remote Sensing Bulletin, 2(3): 204-210. (in Chinese)

[53]
Tu Y, Jiang L L, Liu R, et al. 2021. Spatiotemporal changes of vegetation NDVI and its driving forces in China during 1982-2015. Transactions of the Chinese Society of Agricultural Engineering, 37(22): 75-84. (in Chinese)

[54]
Viña A, Gitelson A A, Nguy-Robertson A L, et al. 2011. Comparison of different vegetation indices for the remote assessment of green leaf area index of crops. Remote Sensing of Environment, 115(12): 3468-3478.

[55]
Wang Q, Zhang B, Dai S P, et al. 2012. Analysis of the vegetation cover change and its relationship with factors in the Three-North Shelter Forest Program. China Environmental Science, 32(7): 1302-1308. (in Chinese)

[56]
Wang S H, Zhang Y G, Ju W M, et al. 2021. Response to comments on "Recent global decline of CO2 fertilization effects on vegetation photosynthesis". Science, 373(6562): abg7484, doi: 10.1126/science.abg7484.

[57]
Wang X H, Piao S L, Ciais P, et al. 2011. Spring temperature change and its implication in the change of vegetation growth in North America from 1982 to 2006. Proceedings of the National Academy of Sciences, 108(4): 1240-1245.

[58]
Wang X Q, Ma L Y, Jia Z K, et al. 2005. Research and application advances in leaf area index (LAI). Chinese Journal of Ecology, 24(5): 537-541. (in Chinese)

[59]
Wang Y Y, Zhaxiyangzong. 2016. Assessing vegetation response to meteorological drought in Tibet Autonomous Region using vegetation condition index. Journal of Applied Meteorological Science, 27(4): 435-444. (in Chinese)

[60]
Willett K M, Sherwood S. 2012. Exceedance of heat index thresholds for 15 regions under a warming climate using the wet-bulb globe temperature. International Journal of Climatology, 32(2): 161-177.

[61]
Williams L E, Baeza P. 2007. Relationships among ambient temperature and vapor pressure deficit and leaf and stem water potentials of fully irrigated, field-grown grapevines. American Journal of Enology and Viticulture, 58(2): 173-181.

[62]
Xiao Z Q, Liang S L, Jiang B. 2017. Evaluation of four long time-series global leaf area index products. Agricultural and Forest Meteorology, 246: 218-230.

[63]
Xie S D, Mo X G, Hu S, et al. 2020. Contributions of climate change, elevated atmospheric CO2 and human activities to ET and GPP trends in the Three-North Region of China. Agricultural and Forest Meteorology, 295: 108183, doi: 10.1016/j.agrformet.2020.108183.

[64]
Xu B D, Li J, Park T, et al. 2018. An integrated method for validating long-term leaf area index products using global networks of site-based measurements. Remote Sensing of Environment, 209: 134-151.

[65]
Xu R L, Li B F, Lian L S. 2020. Quantitative relationship between the spatiotemporal change of relative humidity and climatic factors in the arid region of Northwest China from 1960 to 2015. Research of Soil and Water Conservation, 27(6): 233-239, 246. (in Chinese)

[66]
Xu Y, Ding Y H, Zhao Z C. 2001. Confidence analysis of NCEP/NCAR 50-year global reanalyzed data in climate change research in China. Journal of Applied Meteorological Science, 12(3): 337-347. (in Chinese)

[67]
Yu X L, Guo X L, Wu Z C. 2014. Land surface temperature retrieval from Landsat 8 TIRS—comparison between radiative transfer equation-based method, split window algorithm and single channel method. Remote Sensing, 6(10): 9829-9852.

[68]
Yuan R R, Huang X L, Hao L. 2021. Spatio-temporal variation of vapor pressure deficit and impact factors in China in the past 40 years. Climatic and Environmental Research, 26(4): 413-424. (in Chinese)

[69]
Yuan W P, Zheng Y, Piao S L. et al. 2019. Increased atmospheric vapor pressure deficit reduces global vegetation growth. Science Advances, 5(8): eaax1396, doi: 10.1126/sciadv.aax1396.

[70]
Zhang S, Tao F L, Zhang Z. 2017. Spatial and temporal changes in vapor pressure deficit and their impacts on crop yields in China during 1980-2008. Journal of Meteorological Research, 31(4): 800-808.

[71]
Zhang Y, Li B F, Chen Y N. 2018. The temporal and spatial variation of water vapor content and its relationship with precipitation in the arid region of Northwest China from 1970 to 2013. Journal of Natural Resources, 33(6): 1043-1055. (in Chinese)

Options
文章导航

/

〈 〉