Vegetation is affected by hydrological cycle components that have altered under the influence of climate change. Therefore, it is necessary to investigate the impact of hydrological cycle components on regional vegetation growth, especially in alpine regions. In this study, we employed multiple satellite observations to comprehensively investigate the spatial heterogeneity of hydrological cycle components in the Yarlung Zangbo River (YZR) basin for the period 1982–2014 and to determine the underlying mechanisms driving regional vegetation growth. Results showed that the normalized difference vegetation index (NDVI) values during May–October were high, and the NDVI values increased from the upper reaches of the YZR to its lower reaches, reflecting the enhancement of vegetation growth. Annual precipitation, precipitation-actual evapotranspiration (AET), and snow water equivalent (SWE) all affect terrestrial water storage in the YZR basin through changes in soil moisture (SM), i.e., SM is the intermediate variable. Seasonal variability of vegetation is controlled mainly by precipitation, temperature, AET, SM anomaly, and SWE. Groundwater storage anomalies (GWA) and terrestrial water storage anomalies (TWSA) were not reliable indicators of vegetation growth in the YZR basin and the midstream and downstream regions. The effects of GWA and TWSA on vegetation occurred in the upstream region.

  • The spatial heterogeneity of hydroclimatological factors in the Yarlung Zangbo River basin were analyzed based on satellite observations.

  • The interactions of hydrological factors were discussed.

  • The major controlling factors of the vegetation driving forces were examined.

Graphical Abstract

Graphical Abstract
Graphical Abstract

The Tibetan Plateau (TP) and surrounding mountains, called the third pole by scientists (Yao et al. 2012), is an alpine region widely covered by snow, glaciers, permafrost, and seasonal frost soil. According to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, the global surface air temperature has increased by 0.85 °C for the period 1880–2015, a period in which the TP has experienced rapid climate warming (IPCC 2013). Many studies have reported that the hydrological cycle of the TP has been altered substantially under the effects of climate change (Kang et al. 2010; Zhang & Li 2018; Yang et al. 2019). Vegetation is sensitive to climate change, and the normalized difference vegetation index (NDVI) is an indicator that is effective for the quantification of variations of vegetation dynamics (Ouyang et al. 2020; Yu et al. 2021). Hydrological cycle components (including temperature, precipitation, actual evapotranspiration (AET), snow water equivalent (SWE), and soil moisture (SM)) have a profound impact on vegetation on the TP (Piao et al. 2012; Zhang et al. 2016; Xu et al. 2018). Therefore, it is necessary and meaningful to investigate changes in hydrological cycle components and their effects on vegetation on the TP to support studies on both climate change in the TP region and ecological protection in other alpine areas.

Many major rivers originate on the TP, e.g., the Yarlung Zangbo River (YZR), which is one of the main rivers on the TP and the highest great river in the world (You et al. 2007). The YZR, which represents a substantial source of freshwater for the TP and downstream southern Asian countries, has attracted considerable attention in relation to global warming (Immerzeel et al. 2010; Chen et al. 2017). For example, Liu et al. (2014) revealed that both land use and climate change affect runoff in the YZR basin. Therefore, investigation on how vegetation and climate have changed, and analysis of the effects of climate change on vegetation, could prove useful to support studies on runoff in the YZR basin. Previous studies have typically used conventional indicators (i.e., temperature and precipitation) to investigate the effects of variations of hydrological cycle components on vegetation in the YZR basin (Li et al. 2015, 2019; Han et al. 2018; Liu et al. 2019; Sun et al. 2019). Li et al. (2019) detected a significant (nonsignificant) positive (negative) correlation between growing season NDVI and surface air temperature (precipitation) in the YZR basin. Sun et al. (2019) found that precipitation in both arid and semiarid regions of the YZR basin was correlated positively with NDVI, whereas temperature showed a positive (negative) correlation with NDVI in arid (semiarid) regions. All of the above studies considered only the relationship between NDVI and temperature/precipitation, whereas other hydrological cycle components such as AET, SWE, SM, groundwater (GW), and terrestrial water storage (TWS) were ignored.

In this study, multiple remote-sensing datasets (including precipitation, temperature, AET, SWE, SM, glacier, and NDVI data) were adopted to analyze the changes of hydrological cycle components and their effects on vegetation growth in the alpine basin for the period 1982–2014. The objectives of this study were to determine variations in the hydrological cycle components of the YZR basin and to reveal how regional vegetation growth responds to hydrological and climatic variability. The results have important implications to understand the response of vegetation to hydrological cycle components variation in alpine mountainous regions under the impact of climate change.

Study area description

The YZR originates on the Southeast TP (China) from the Gemayangzong glacier at an elevation of 5,200 m (Figure 1). The YZR is the highest river in the world with an average elevation of over 4,000 m. The drainage area of the YZR basin (27°40′–31°15′N, 82°00′–97°15′E) is approximately 2.4×105km2, and the range of its elevation is 152–7,182 m. The YZR basin has a complex topography and includes several different climate zones from its lower reaches to the upper reaches. The YZR has five major tributaries, i.e., the Duoxiong Zangbo River, Nianchu River, Lhasa River, Nyang River, and Parlung Zangbo River (Ban et al. 2020). Average annual precipitation in the upper reaches to the lower reaches varies from 200 to 5,000 mm, respectively (Li et al. 2014). Rainfall during the months of the Indian summer monsoon (June–September) accounts for 60–80% of the annual total (You et al. 2007). Precipitation, snow meltwater, and glacier water are the main sources of the water resource of the YZR basin, which is vital to the socioeconomic development of the entire TP region and downstream southern Asian countries.

Dataset description

Precipitation, temperature, and AET datasets

Precipitation and temperature data from the China regional surface Meteorological Feature Dataset (CMFD) (Yang et al. 2010) were used in this study. The precipitation rate and temperature data in the CMFD have a spatial resolution of 0.1° (https://data.tpdc.ac.cn/zh-hans/data/8028b944-daaa-4511-8769-965612652c49/) and are available for the period 1979–2018. The dataset, which incorporates multiple data sources and has been validated using ground-based observational data (Yang et al. 2020), has been used widely in many research areas, e.g., hydrological models (Ma et al. 2016) and land surface process simulations (Huang et al. 2016). The AET data (Ma et al. 2019) used in this study were obtained from the terrestrial evapotranspiration dataset across China.

SWE and SM datasets

The SWE data were derived from snow depth data obtained from the National TP Data Center (http://data.tpdc.ac.cn). The raw data were used to retrial snow depth data derived from passive microwave satellite remote-sensing data (e.g., Scanning Multichannel Microwave Radiometer, Special Sensor Microwave/Imager, and Special Sensor Microwave Imager/Sounder) provided by the National Snow and Ice Data Center (Che et al. 2008). The spatial resolution of the snow depth data is 0.25°. Following the method of Han et al. (2019), SWE can be calculated as follows:
formula
(1)
where is the snow depth based on passive microwave sensor data and is the density of liquid water.

The SM data were obtained from the Global Land Data Assimilation System (GLDAS) Noah land surface model (https://disc.sci.gsfc.nasa.gov/services/grads-gds/gldas). The GLDAS dataset includes four soil layers for a soil depth of 2 m and excludes GW. Monthly time-series data for the period 1982–2014 with 0.25° spatial resolution were derived from the GLDAS. In this study, the SM of the four soil layers from the GLDAS dataset was accumulated as a new SM metric for further analysis.

Glacier data and NDVI data

In this study, the glacier data were provided by the National TP Data Center (http://data.tpdc.ac.cn), obtained using Landsat multispectral data based on the remote-sensing visual interpretation method, represented three periods: 1976, 2001, and 2013 (Ye et al. 2017). The Global Inventory Modeling and Mapping Studies (GIMMS) NDVI data were used to characterize the changes in regional vegetation (Huang et al. 2020). The GIMMS-NDVI third-generation (GIMMS-NDVI3 g) data, downloaded from the website of the National Oceanic and Atmospheric Administration, were obtained for 15-d intervals with an 8-km spatial resolution for the period 1981–2015 (Zhu et al. 2013).

Gravity recovery and climate experiment (GRACE)-retrieved TWS change

The Center for Space Research (CSR) GRACE RL06 mascon product (http://www2.csr.utexas.edu/grace/RL06_mascons.html) used in this study has the same standards applied as the CSR RL06 Spherical harmonics solutions using GRACE level-1 observations, and the C20 solutions from satellite laser ranging analysis were used to replace the C20 coefficients (Cheng & Tapley 2004). Further details regarding the processing of GRACE data can be found in Swenson et al. (2008) and Save (2019). The spatial resolution of the CSR GRACE RL06 mascon data is 0.25° (Save 2019). Monthly TWS anomaly data, derived from the CSR GRACE RL06 mascon data for the period 2003–2014, were applied in this study. There were 11 months with missing data (i.e., 2003–06, 2011–01, 2011–06, 2012–05, 2012–10, 2013–03, 2013–08, 2013–09, 2014–02, 2014–07, and 2014–12), for which substitute values were obtained using the linear interpolation method.

Linear regression method

The linear regression method was applied to analyze the trends of the hydrological cycle components factors at each pixel in the YZR basin for the period 1982–2014. The linear regression method can be expressed as follows:
formula
(2)
where y is a hydrological cycle component, t is the year, a is the slope, and b is the intercept of the regression.

Change-point detection method

A nonparametric method developed by Pettitt (1979) is widely applied to detect abrupt changes of hydrological cycle components (Martinez et al. 2010; Zuo et al. 2014, 2016). The Pettitt test uses a version of the Mann–Whitney statistic, which can be expressed as follows:
formula
(3)
where
formula
formula
(4)
formula
The breakpoint is defined as where reaches its maximum value:
formula
(5)
formula
(6)
where P is the significance level. A significant change point exists when < 0.05.

Correlation analysis

In this study, the Pearson correlation coefficient was applied to decide the quantitative contribution of many hydrological cycle components to the NDVI change (Pearson 1895):
formula
(7)
where x and y represent the NDVI and hydrological cycle components, respectively, R is the correlation coefficient, and n is the number of data pairs.

GW calculated from water balance equation

As GW data are not included in the GLDAS dataset, GW data can be expressed based on the water balance equation. Following the work of Jia et al. (2020), the GW anomaly can be expressed as follows:
formula
(8)
where , , , and are the anomalies of GW storage, TWS, snow water, and SM, respectively.

Intra-annual variabilities of hydrological cycle components

Multiple remote-sensing datasets were used to investigate the intra-annual variabilities of hydrological cycle components. The intra-annual variabilities of precipitation, temperature, AET, SWE, SM, and NDVI in the YZR basin for the period 1982–2014 are illustrated in Figure 2. It can be seen that June–September precipitation (402.7 mm) in the YZR basin accounts for 70% of the annual total precipitation (577.9 mm). The average air temperature in May–September is over 3 °C, whereas it is generally below 0 °C in other months. Similar to temperature, the highest average AET (over 50 mm) occurs in May–September, accounting for approximately 77% of the average annual total AET (465.2 mm). It shows that AET increases with the increase of temperature in summer. In contrast to precipitation, the average SWE in June–September reaches its lowest value (0.01 mm), whereas that in December–February is the highest (4.8 mm). Snowfall in winter is usually heavy, which leads to the relatively high SWE, whereas the occurrence of snow cover remaining throughout summer is rare for the high temperature, which produces the low SWE. The average SM in all months is generally over 500 mm, reaching a maximum in August (646.4 mm), which might be related to the high level of precipitation and high volumes of snow meltwater and glacier water in summer. The average NDVI value in June–September is generally over 0.24, reaching its maximum in August with a value over 0.32. Vegetation grows as the temperature rises; therefore, vegetation growth is maximum (minimum) in summer (winter) with high (low) temperatures.

Spatial distributions of annual hydrological cycle components

The spatial distributions of annual precipitation, temperature, AET, SWE, SM, and NDVI in the YZR basin during the period 1982–2014 are shown in Figure 3. The average annual precipitation in the YZR basin is in the range of 192.1–2023.8 mm with a mean value of approximately 577.9 mm, showing a trend of increase from the upper reaches of the river to its lower reaches. Annual precipitation in the upper reaches is approximately 200 mm, while that in the middle reaches is nearly 800 mm; the maximum value of approximately 2000 mm is found in the southeast of the region. The average annual temperature across the YZR basin ranges from −14.7 to 23.4 °C with a mean value of 0.1 °C; the areas of low temperature are distributed mainly in the upper reaches of the river and in the northeast region of the YZR basin. Along the main channel of the YZR, the temperature is higher than in other regions. The annual AET of the upper reaches of the YZR basin is in the range of 1.9–285.9 mm, while that of the middle and lower reaches is in the range of 570.0–854.1 mm; the mean value for the entire YZR basin is 465.2 mm. The spatial distribution of the average annual AET is related to that of both precipitation and temperature in the YZR basin. The average annual SWE varies from 0.1 mm in the upper reaches of the YZR to 10.5 mm in the lower reaches; the SWE in the middle and lower reaches is generally in the range of 3.4–10.5 mm. The average annual SM in the entire basin ranges from 462.2 to 1,184.5 mm. The SM values in most parts of the middle-lower reaches of the YZR basin are in the range of 462.2–649.4 mm, although the values at certain grids in the upper and middle reaches are occasionally over 836.5 mm. The average annual NDVI across the entire YZR basin ranges from −0.3 to 0.7. The values in the upper and middle reaches are over 0, while the values in the lower reaches are mainly in the range of 0.4–0.7, reflecting the dominant control of precipitation and temperature on the spatial distribution of vegetation. Vegetation cover in the upper reaches is low owing to the lack of precipitation for supporting growth, whereas vegetation grows well in the lower reaches of the YZR basin owing to abundant precipitation and relatively high temperatures.

Inter-annual variations of annual hydrological cycle components

The inter-annual variations of annual hydrological cycle components in the YZR basin for the period 1982–2014 are displayed in Figure 4. Annual precipitation shows a highly fluctuating trend of increase with an average rate of increase of approximately 26.43 mm/10a. The highest (lowest) value of annual precipitation is 650–700 (below 450) mm. Annual temperature increases significantly at a rate of 0.65 °C /10a for the period 1982–2014; however, the rate of increase during 1982–1997 is slower than during 1998–2014. The annual AET and the NDVI both show trends of increase for the period 1982–2014 with rates of 8.73 mm/10a and 0.002/10a, respectively. The annual SWE and SM show rates of decrease of 0.26 and 4.00 mm/10a, respectively. Rising temperatures induce increases in AET and snowmelt, which result in reduced SWE. With increasing precipitation and temperature, regional vegetation grows well. These results reveal that the YZR basin is experiencing climate warming and wetting, which is consistent with the conclusions of Zhang et al. (2020).

Spatiotemporal variations of TWS

The spatial pattern of the trend (2003–2014) of annual TWS in the YZR basin, derived from GRACE CSR RL06 data, is illustrated in Figure 5. There is obvious spatial heterogeneity in the trend of annual TWS in the YZR basin, with significant losses of TWS in the YZR basin and surrounding areas. The trend of decline in TWS is most (least) severe in the lower (upper) reaches of the YZR basin with a maximum (minimum) value of −11.90 (−0.12) mm/a.

The variations in the trends of annual TWS anomalies (TWSA) and groundwater storage anomalies (GWA) in the YZR basin for the period 2003–2014 are presented in Figure 6. The trends of TWSA and GWA in the YZR basin during 2003–2014, calculated based on GRACE data, show rates of decrease of −15.45 and −14.52 mm/a, respectively, consistent with the trend of TWS shown in Figure 5. The results show that TWSA and GWA were in a continual state of decline during the period 2003–2014.

Abrupt change points of hydrological cycle components

Abrupt change points of hydrological cycle components in the YZR basin for the period 1982–2014 were detected using the nonparametric Pettitt test (Figure 7). The year 1995 was detected as an abrupt change point in annual precipitation at the 0.05 significance level. Similarly, an abrupt change point in 1997 was detected at the 0.001 significance level for annual temperature. Abrupt change points in the years 2000, 2000, and 2005 were also found for annual AET, SWE, and SM, respectively, at the 0.05 significance level. No abrupt change point was detected in the NDVI. The results indicate that abrupt change points for most hydrological cycle components occurred near the year 2000. Liu (2015) and Wang (2016) demonstrated that the precipitation, temperature, and potential evapotranspiration changed significantly in the late 1990s. Liu et al. (2018) decided that the year 2000 was the turning point to investigate the variation characteristics of the standardized precipitation evapotranspiration index in the YZR basin. In this study, the year 2000 was chosen as the abrupt change point to investigate the changes of hydrology cycle components.

Interactive relationships of hydrological cycle components

Precipitation is the major factor affecting the interactions within the hydrological system (Gao et al. 2014; Xu et al. 2018; Yao et al. 2019). Monthly precipitation, SMA, and TWSA were compared in Figure 8. The results showed that there was a good consistency between precipitation and SMA for the period 2003–2014, more precipitation in summer infiltrated into the soil to form soil water, and vice versa in winter. Monthly TWSA showed a significant decreasing trend during 2003–2014, with the most rapid decreasing trend from 2008 to 2014. The reduction in TWSA was induced by both decreasing precipitation and increasing AET, with AET having a more important role in water storage changes (Meng et al. 2018).

Interactive relationships between annual precipitation, TWS change, SM, Precipitation-AET (P-E), and SWE in the YZR basin were investigated (Figure 9). Changes in precipitation are correlated more with changes in SM than with changes in TWS, e.g., correlation coefficients of 0.70 and 0.10, respectively. It means that SM changes are driven by precipitation and that additional precipitation leads to increased SM. Similarly, changes in P-E are correlated more with changes in SM than with changes in TWS. The annual SM is correlated most with the annual TWS with a correlation coefficient of 0.55, which means that changes in SM are the driving factor of changes in TWS. In comparison with the positive correlation between SWE and SM, changes in precipitation, P-E, and TWS have a weaker positive correlation with SWE, indicating that changes in SWE are more closely related to SM than to precipitation, P-E, and TWS. Annual precipitation, P-E, and SWE have considerable influence on the changes in SM, and SM is also closely correlated with the changes in the TWS. Therefore, annual precipitation, P-E, and SWE have an impact on TWS throughout the YZR basin via changes in SM, i.e., SM is the intermediate variable.

Effects of hydrological cycle components on the NDVI

Changes in vegetation are strongly controlled by variations in hydrological cycle components (e.g., precipitation, temperature, AET, SMA, SWE, GWA, and TWSA) (Chen et al. 2015). To evaluate the effects of the variations of hydrological cycle components on vegetation cover, correlation coefficients were determined to explore the relationships between the NDVI and the hydrological cycle components in the YZR basin for the period 1982–2014. Based on the abrupt change points found in most annual time series of the hydrological cycle components, the correlation coefficients were calculated for two periods: 1982–1997 and 2003–2014. To detect the relationships between the NDVI and the hydrological cycle components, correlation coefficients were calculated on the annual basis and for the four seasons (Table 1). The four seasons were defined as follows: spring (March–May), summer (June–August), autumn (September–November), and winter (December–February).

The correlation coefficients between the NDVI and the hydrological cycle components in the YZR basin for the period 1982–2014 are listed in Table 1. The relationship between precipitation and the NDVI is indistinct throughout the entire study period; however, the NDVI is correlated significantly with temperature in spring, autumn, winter, and on the annual scale. The NDVI is also correlated significantly with AET in all seasons and on the annual scale. The above results indicate that the effects of temperature and AET on vegetation are intense under the conditions of climate warming. The SWE has no obvious relationship with the NDVI in four seasons and on the annual scale. Additionally, the NDVI is correlated significantly with SMA in summer but has an indistinct relationship in other seasons, indicating that the roots of vegetation in the YZR basin are more dependent on surface SM conditions that can provide sufficient water for plant growth. For the period 1982–2000, precipitation is correlated significantly with the NDVI in autumn and winter, and temperature is correlated significantly with the NDVI annually. AET and SMA did not show an obvious relationship with the NDVI, while SWE is correlated significantly with the NDVI in autumn, winter, and annually. For the period 2003–2014, precipitation, temperature, SMA, GWA, and TWSA show no obvious relationship with the NDVI, while AET shows an obvious relationship with the NDVI in spring, summer, and on the annual scale. SWE is correlated significantly with the NDVI in winter.

To investigate the impact of elevation on vegetation, we studied the effects of hydrological cycle components on NDVI in the upstream, midstream, and downstream regions of the YRZ basin (Tables 24). In the upstream region (Table 2), for the period 1982–2014, precipitation is correlated significantly with the NDVI in spring and winter. Temperature is correlated significantly with the NDVI in autumn and winter, while AET is correlated significantly with the NDVI in summer and winter. SMA shows no obvious relationship with the NDVI and SWE is correlated significantly with the NDVI on the annual scale while having an indistinct relationship in the four seasons. For the period 1982–2000, precipitation is correlated significantly with the NDVI in spring and winter, while temperature and SWE have an obvious relationship with the NDVI in the four seasons (except autumn) and on the annual scale. AET and SMA show a significant correlation with the NDVI in winter and on the annual scale, respectively. During 2003–2014, precipitation is correlated significantly with the NDVI in winter, while temperature and AET are correlated significantly with the NDVI in the two seasons. No obvious correlation was found between SMA and the NDVI. SWE was the same as SMA. GWA and TWSA are correlated significantly with the NDVI in winter.

In the midstream region (Table 3), during 1982–2014 precipitation and AET are correlated significantly with the NDVI on the annual scale, while temperature has an obvious relationship with the NDVI in all seasons (except summer) and on the annual scale. SMA has an indistinct relationship with the NDVI except for summer, and SWE shows no obvious relationship with the NDVI. For the period 1982–2000, the relationship between NDVI and hydroclimatological factors is the same as that in 1982–2014. During 2003–2014, precipitation and temperature are correlated significantly with the NDVI on the annual scale and spring, respectively. SWE is correlated significantly with the NDVI in winter, while other hydroclimatological factors have no obvious relationship with the NDVI. In the downstream region (Table 3), for the period 1982–2014, precipitation and SWE are correlated significantly with the NDVI in summer and winter, respectively, while other hydroclimatological factors show no obvious relationship with the NDVI in the four seasons and on the annual scale. During 1982–2000, precipitation has an indistinct relationship with the NDVI, while temperature and AET show a significant correlation with the NDVI in winter, and SWE is correlated significantly with the NDVI in winter and on the annual scale. For the period 2003–2014, precipitation and AET are correlated significantly with the NDVI in summer and spring, respectively, while other hydroclimatological factors show no obvious relationship with the NDVI.

In summary, in the whole basin, during the three studied periods (i.e., 1982–2014, 1982–2000, and 2003–2014), GWA and TWSA in the YZR basin are not correlated significantly with the NDVI, whereas precipitation, temperature, AET, SMA, and SWE have obvious significant correlation with the NDVI in certain seasons. Among the upstream, midstream, and downstream regions, precipitation, temperature, AET, and SWE are correlated significantly with the NDVI, while GWA and TWSA are correlated significantly with the NDVI in the upstream region. The results indicate that the seasonal variability of regional vegetation is controlled mainly by precipitation, temperature, AET, SMA, and SWE. Thus, precipitation, temperature, AET, SMA, and SWE appear to be the hydrological cycle components best able to explain the variability of vegetation dynamics in the YZR basin and upstream, midstream, and downstream regions, while the effects of GWA and TWSA on vegetation exist in the upstream region. In the YZR basin, precipitation is the indispensable factor for vegetation growth, while higher temperature will produce more snowmelt water for vegetation grow and increase AET to a certain extent. Soil water deficit is also a major element restricting plant growth in this region (Zhong et al. 2014; Sun et al. 2019). Higher temperatures will increase the evaporation amount from the soil layer, remaining lowering soil water amount for vegetation growth (Piao et al. 2006). Sun et al. (2019) and Liu et al. (2019) also found that precipitation and temperature were the driving forces of vegetation in the YZR basin.

This study comprehensively investigated the changes in hydrological cycle components in the YZR basin for the period 1982–2014 and analyzed their impacts on vegetation using multiple remote-sensing datasets (i.e., precipitation, temperature, AET, SWE, SM, glacier, and NDVI data) and various statistical methods. The main conclusions derived were as follows:

  1. Precipitation, temperature, and AET in the YZR basin were high in June–September and low in other months. SWE displayed low values in June–August and reached high values in March–April. SMA was generally stable with only relatively high values occurring in July–September. The mean annual precipitation, temperature, and AET in the study region were 577.92 mm, 0.09 °C, and 465.20 mm, respectively. The average annual SM was in the range of 462.20–1184.45 mm.

  2. Precipitation showed an increasing trend with a rate of increase of 26.43 mm/10a. During the study period, temperature had a rate of increase with the value of 0.65 °C/10a. The values of AET and the NDVI showed obvious trends of increase during 1982–2014 at rates of 8.73 and 0.002 mm/10a, respectively. The values of SWE and SM had rates of decrease with values of −0.26 and −4.00 mm/10a, respectively.

  3. The NDVI showed a slight greening trend during 1982–2014. The NDVI values during May–October were high, and the NDVI values increased from the upper reaches of the YZR to its lower reaches, reflecting the enhancement of vegetation growth. The TWSA and GWA values in the YZR basin, calculated based on GRACE data, showed trends of decrease with rates of −15.45 and −14.18 mm/year, respectively, for the period 2003–2014.

  4. Annual precipitation, AET, and SWE have considerable influence on changes in SM, and changes in SM were closely correlated with changes in TWS. Thus, annual precipitation, P-E, and SWE had an impact on TWS in the YZR basin through changes in SM, i.e., SM was the intermediate variable.

  5. Results showed that the seasonal variability of vegetation in the YZR basin was controlled mainly by precipitation, temperature, AET, SMA, and SWE. GWA and TWSA were not reliable indicators of vegetation dynamics in the YZR basin and upstream, midstream, and downstream regions. The effects of GWA and TWSA on vegetation existed in the upstream region.

Knowledge of hydrological cycle components such as precipitation, temperature, AET, SWE, SM, and glaciers is essential for understanding various hydrological cycles and ecological processes. A comprehensive understanding of these processes is vital for agricultural water resource management and ecological protection in the alpine basin.

This project is funded by the National Natural Science Foundation of China (No. 91647202).

All relevant data are available from an online repository or repositories (http://data.tpdc.ac.cn, https://disc.sci.gsfc.nasa.gov/services/grads-gds/gldas and http://www2.csr.utexas.edu/grace/RL06_mascons.html). All data employed or analysed during this study are described in this article.

Ban
C.
,
Xu
Z.
,
Zuo
D.
,
Liu
X.
,
Zhang
R.
&
Wang
J.
2020
Vertical influence of temperature and precipitation on snow cover variability in the Yarlung Zangbo River Basin, China
.
International Journal of Climatology
41
(
2
),
1148
1161
.
Che
T.
,
Li
X.
,
Jin
R.
,
Armstrong
R.
&
Zhang
T.
2008
Snow depth derived from passive microwave remote-sensing data in China
.
Annals of Glaciology
49
(
1
),
145
154
.
Cheng
M.
&
Tapley
B. D.
2004
Variations in the Earth's oblateness during the past 28 years
.
Journal of Geophysical Research: Solid Earth
109
(
B9
),
1
9
.
Han
X.
,
Zuo
D.
,
Xu
Z.
,
Cai
S.
&
Gao
X.
2018
Analysis of vegetation condition and its relationship with meteorological variables in the Yarlung Zangbo River Basin of China
. In:
Innovative Water Resources Management – Understanding and Balancing Interactions Between Humankind and Nature
(
Xu
Z.
, ed.).
Proceedings of the International Association of Hydrological Sciences (IAHS). Copernicus Gesellschaft Mbh
,
Gottingen
, pp.
105
112
.
Huang
Y.
,
Salama
M. S.
,
Su
Z.
,
Van
D. R.
,
Zheng
D.
,
Krol
M.
&
Zhou
Y.
2016
Effects of roughness length parameterizations on regional-scale land surface modeling of alpine grasslands in the Yangtze River Basin
.
Journal of Hydrometeorology
17
(
4
),
1069
1085
.
Immerzeel
W. W.
,
van Beek
L. P. H.
&
Bierkens
M. F. P.
2010
Climate change will affect the Asian water towers
.
Science
328
(
5984
),
1382
1385
.
IPCC
2013
Climate Change 2013: The Physical Science Basis
.
Cambridge University Press
,
Cambridge
.
Jia
B.
,
Cai
X.
,
Zhao
F.
,
Liu
J.
,
Chen
S.
,
Luo
X.
&
Xu
J.
2020
Potential future changes of terrestrial water storage based on climate projections by ensemble model simulations
.
Advances in Water Resources
142
,
1036535
.
Kang
S.
,
Xu
Y.
,
You
Q.
,
Flügel
W.
,
Pepin
N.
&
Yao
T.
2010
Review of climate and cryospheric change in the Tibetan Plateau
.
Environmental Research Letters
5
(
1
),
015101
.
Li
F.
,
Zhang
Y.
,
Xu
Z.
,
Liu
C.
,
Zhou
Y.
&
Liu
W.
2014
Runoff predictions in ungauged catchments in southeast Tibetan Plateau
.
Journal of Hydrology
511
,
28
38
.
Li
H.
,
Li
Y.
,
Shen
W.
,
Li
Y.
,
Lin
J.
,
Lu
X.
&
Jiang
J.
2015
Elevation-dependent vegetation greening of the Yarlung Zangbo River Basin in the Southern Tibetan Plateau, 1999–2013
.
Remote Sensing
7
(
12
),
16672
16687
.
Li
H.
,
Liu
L.
,
Liu
X.
,
Li
X.
&
Xu
Z.
2019
Greening implication inferred from vegetation dynamics interacted with climate change and human activities over the Southeast Qinghai-Tibet Plateau
.
Remote Sensing
11
(
20
),
23
.
Liu
X.
2015
Analysis of the Meteorological and Hydrological Characteristics in the Yarlung Zangbo River Basins
.
Master's Thesis
,
Tsinghua University
,
Beijing
,
China
.
Liu
Z.
,
Yao
Z.
,
Huang
H.
,
Wu
S.
&
Liu
G.
2014
Land use and climate changes and their impacts on runoff in the Yarlung Zangbo River Basin, China
.
Land Degradation & Development
25
(
3
),
203
215
.
Liu
L.
,
Niu
Q.
,
Heng
J.
,
Li
H.
&
Xu
Z.
2018
Transition characteristics of the dry-wet regime and vegetation dynamic responses over the Yarlung Zangbo River Basin, Southeast Qinghai-Tibet Plateau
.
Remote Sensing
11
(
10
),
1
26
.
Ma
N.
,
Szilagyi
J.
,
Niu
G.
,
Zhang
Y.
,
Zhang
T.
,
Wang
B.
&
Wu
Y.
2016
Evaporation variability of Nam Co Lake in the Tibetan Plateau and its role in recent rapid lake expansion
.
Journal of Hydrology
537
,
27
35
.
Martinez
M. D.
,
Serra
C.
,
Burgueno
A.
&
Lana
X.
2010
Time trends of daily maximum and minimum temperatures in Catalonia (NE Spain) for the period 1975–2004
.
International Journal of Climatology
30
(
2
),
267
290
.
Meng
F.
,
Su
F.
,
Li
Y.
&
Tong
K.
2018
Changes in terrestrial water storage during 2003–2014 and possible causes in Tibetan Plateau
.
Journal of Geophysical Research: Atmospheres
124
,
2909
2931
.
Pearson
K.
1895
Contributions to the mathematical theory of evolution. II. Skew variation in homogeneous material
.
Philosophical Transactions of the Royal Society of London A: Mathematical, Physical and Engineering Sciences
186
,
343
414
.
Pettitt
A. N.
1979
A non-parametric approach to the change-point problem
.
Applied Statistics
28
(
2
),
126
135
.
Piao
S.
,
Mohammat
A.
,
Fang
J.
,
Cai
Q.
&
Feng
J.
2006
NDVI-based increase in growth of temperate grasslands and its responses to climate changes in China
.
Global Environmental Change
16
(
4
),
340
348
.
Piao
S.
,
Tan
K.
,
Nan
H.
,
Ciais
P.
,
Fang
J.
,
Wang
T.
&
Zhu
B.
, et al
2012
Impacts of climate and CO2 changes on the vegetation growth and carbon balance of Qinghai-Tibetan grasslands over the past five decades
.
Global and Planetary Change
98–99
,
73
80
.
Save
H.
2019
CSR GRACE RL06 Mascon Solutions. https://doi.org/10.18738/T8/UN91VR, Texas Data Repository Dataverse, V1
.
Swenson
S.
,
Chambers
D.
&
Wahr
J.
2008
Estimating geocenter variations from a combination of GRACE and ocean model output
.
Journal of Geophysical Research: Solid Earth
113
(
B8
),
12
.
Wang
L.
2016
Study on Hydrochemical Characteristics and Its Influencing Factors in Yarlung Tsangpo River Basin
.
PhD Thesis
,
Institute of Geographic Sciences and Natural Resources Research, CAS
,
Beijing
,
China
.
Yang
K.
,
He
J.
,
Tang
W.
,
Qin
J.
&
Cheng
C. C. K.
2010
On downward shortwave and longwave radiations over high altitude regions: observation and modeling in the Tibetan Plateau
.
Agricultural and Forest Meteorology
150
(
1
),
38
46
.
Yang
W.
,
Wang
Y.
,
Liu
X.
,
Zhao
H.
,
Shao
R.
&
Wang
G.
2020
Evaluation of the rescaled complementary principle in the estimation of evaporation on the Tibetan Plateau
.
Science of the Total Environment
699
,
134367
.
Yao
T.
,
Thompson
L.
,
Yang
W.
,
Gao
B.
,
Guo
X.
,
Yang
X.
&
Joswiak
D.
2012
Different glacier status with atmospheric circulations in Tibetan Plateau and surroundings
.
Nature Climate Change
2
(
9
),
663
667
.
Yao
J.
,
Hu
W.
,
Chen
Y.
,
Huo
W.
,
Zhao
Y.
,
Mao
W.
&
Yang
Q.
2019
Hydro-climatic changes and their impacts on vegetation in Xinjiang, Central Asia
.
Science of the Total Environment
660
,
724
732
.
Ye
Q.
,
Zong
J. B. A.
,
Tian
L.
,
Cogley
J. G.
,
Song
C.
&
Guo
W.
2017
Glacier changes on the Tibetan Plateau derived from Landsat imagery: mid-1970s-2000–13
.
Journal of Glaciology
63
(
238
),
273
287
.
You
Q.
,
Kang
S.
,
Wu
Y.
&
Yan
Y.
2007
Climate change over the Yarlung Zangbo River Basin during 1961–2005
.
Journal of Geographical Sciences
17
(
4
),
409
420
.
Yu
Y.
,
Zhou
Y.
,
Xiao
W.
,
Ruan
B.
,
Lu
F.
,
Hou
B.
&
Cui
H.
, et al
2021
Impacts of climate and vegetation on actual evapotranspiration in typical arid mountainous regions using a Budyko-based framework
.
Hydrology Research
52
(
1
),
212
228
.
Zhang
Y.
,
Zhang
C.
,
Wang
Z.
,
Chen
Y.
,
Gang
C.
,
An
R.
&
Li
J.
2016
Vegetation dynamics and its driving forces from climate change and human activities in the Three-River Source Region, China from 1982 to 2012
.
Science of the Total Environment
563
,
210
220
.
Zhang
R.
,
Xu
Z.
,
Zuo
D.
&
Ban
C.
2020
Hydro-Meteorological trends in the Yarlung Zangbo River Basin and possible associations with large-scale circulation
.
Water
12
(
1
),
20
.
Zhong
L.
,
Ma
Y.
,
Fu
Y.
,
Pan
X.
,
Hu
W.
,
Su
Z.
&
Feng
L.
, et al
2014
Assessment of soil water deficit for the middle reaches of Yarlung-Zangbo River from optical and passive microwave images
.
Remote Sensing of Environment
142
(
3
),
1
8
.
Zuo
D.
,
Xu
Z.
,
Wu
W.
,
Zhao
J.
&
Zhao
F.
2014
Identification of streamflow response to climate change and human activities in the Wei River Basin, China
.
Water Resources Management
28
(
3
),
833
851
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial purposes with no derivatives, provided the original work is properly cited (http://creativecommons.org/licenses/by-nc-nd/4.0/).