Abstract

Continental or regional water storage variations (WSVs) are crucial to regional economic development and human society and play an important role in coping with global change. Water scarcity is currently an especially key issue in Central Asia (CA), and therefore the study of WSVs can aid in the adoption of measures for mitigating pressures from contemporary environmental changes and economic development in CA. Based on Gravity Recovery and Climate Experiment (GRACE), Global Land Data Assimilation System (GLDAS), and CRU meteorological datasets and a proposed combined filter strategy, WSVs in Central Asia and its surrounding areas over 30 years are investigated in this paper. The results indicate that the WSVs derived from GRACE and GLDAS over CA generally show a decreasing tendency. CRU data demonstrated that CA has been undergoing a warming trend. The water loss in CA may be caused by warming, which will lead to the loss of soil moisture. Moreover, the water mass in the Tibetan Plateau and Tarim basin increases, which may be caused by glacier melting in the Pamirs and Himalaya. The precipitation contributed little to changes in water storage, but at the basin scale, the precipitation anomalies were very similar to the GRACE and GLDAS data, which can be viewed as an indicator of WSVs.

INTRODUCTION

Central Asia (CA) is one of most seriously ecologically and environmentally deteriorated regions in the world and now faces vegetation degradation, land desertification, soil salinization, shortages of water resources, discontinuous river flows and disappearing lakes. Water scarcity is currently an especially key issue in this region and has severely restricted economic and societal development. Water storage is a key state variable in the global water cycle and climate change and is an extremely important resource related to economic and societal development.

Water scarcity is currently a key issue, and given that the main rivers are common to state boundaries, conflicts over water use arise regularly on national and local scales in Central Asia (CA) (Horsman 2001; Bernauer & Siegfried 2012). Millions of people in the geopolitically important region of Central Asia depend on water from international rivers that are driven by snow and glacial meltwater, especially the Syr Darya and Amu Darya Rivers (Aizen et al. 2001). The riparian countries of these rivers have experienced recurring water allocation conflicts since the collapse of the Soviet Union (Siegfried et al. 2010). At the same time, economic development has increased the demand for water resources, and a large amount of terrestrial water resources have been exploited in CA (Stone 2012). This exploitation caused friction between CA countries in arid areas and affected regional stability and national security (Qi & Evered 2008). Climate change is also widely expected to exacerbate water stresses in CA (Siegfried et al. 2010). The warming trends in CA have increased the melting rates of mountain glaciers and snow over the past 50 years and have changed the water resources of major river basins (Qi & Evered 2008). Issues surrounding water resources in CA have become important topics for numerous scholars and various levels of the government (Parry 2007; Siegfried et al. 2010).

Water storage is a key state variable of the global water cycle and climate (Famiglietti 2004), and it is an extremely important resource for economic and societal development (Yang et al. 2013). The study of water storage variations (WSVs) in CA and its surrounding areas (SA) is necessary for analyses of water recycling and climate change and for estimating water resource changes in the future. The Gravity Recovery and Climate Experiment (GRACE) was launched by NASA and the German Aerospace Center in March 2002 to observe the Earth's gravitational field and energy cycle, and has provided new approaches for WSV measurements of unprecedented accuracy (Tapley et al. 2004a). Over 10 years, GRACE has collected large amounts of spatial data that can aid studies of variations in global and regional water storage, hydrology and water resources (Ramillien et al. 2008). There has recently been considerable progress in the use of GRACE datasets for monitoring WSVs at mid-latitudes in Asia. Xu et al. used GRACE data to estimate the trend in the spatiotemporal variation of water reserves in the Tianshan Mountains during a recent 8-year period (Xu et al. 2013). Yang & Chen studied terrestrial WSVs based on GRACE data and compared the results with those derived from the Tianshan Mountains and the SA using the Global Land Data Assimilation System (GLDAS) model (Yang & Chen 2015). They reached the same conclusion as Xu et al. (2013) for the respective study time frames: the interannual fluctuations in water storage tended to decrease in the Tianshan Mountains. Sun et al. (2014) analyzed the annual variations in water storage over CA using 10 years of GRACE data with support from other data, and the amplitudes of the annual variations in water storage exhibited a general E-W increasing trend.

Generally, at mid-latitudes in the area of Asia, most analyses of WSVs related to GRACE datasets have focused on the Qinghai-Tibet plateau, Tarim basin and Tianshan Mountain areas. The existing status and trends of future changes in water storage in Central Asia and its SA are still open scientific issues. The aim of this study is to estimate the characteristics of spatial-temporal variations in water storage and their relationship to climate factors that affect CA and the SA by combining GRACE, GLDAS and other datasets from 1979 to 2012, a period of over 30 years, to lay a foundation for more quantitative studies in the future and hopefully provide basic information and theoretical support for the sustainable use and development of water resources in CA and its SA.

STUDY AREA AND DATASETS

Study area

The study area included five countries in Central Asia (Kazakhstan, Uzbekistan, Kyrgyzstan, Turkmenistan and Tajikistan) (Cowan 2007) and the SA (including Xinjiang and the Tibetan Plateau in China, the western part of Mongolia and southern Russia) (Figure 1, left). Central Asia is located in the interior of Eurasia, in which the southeastern edges of mountain ranges cut off warm moist air that flows from the Indian and Pacific Oceans (Mughal 2013). Hence, Central Asia has a typical temperate continental desert and grassland climate. Temperatures and precipitation follow a gradient from north to south and from the lowlands to the mountains (Gessner et al. 2013). The annual precipitation in these regions is between 100 and 400 mm (Gessner et al. 2013). The water resources in the study area are preserved in the form of glaciers, lake water, runoff, and soil moisture. The high elevation regions, Tianshan, Pamirs and Tibetan Plateau, contain abundant glaciers, which are very important sources of fresh water. The Pamirs' glaciers also make a notable contribution to the runoff in the headwater catchments of CA, including those of the Amu Darya and Syr Darya Rivers (Unger-Shayesteh et al. 2013). The Caspian Sea, which is the largest inland body of water in the world, the Aral Sea, Balkhash Lake and Issyk-Kul Lake contain large water resources that are used in CA for agriculture and industry. Agriculture is the largest water consumer in the study area, accounting for more than 90% of water withdrawal (Unger-Shayesteh et al. 2013). Agricultural water consumption has generated many serious environment problems, including the shrinkage of the Aral Sea. The area of the Aral Sea declined by 62% from 2002 to 2009 (Singh et al. 2012). To study the changes in water storage at the basin scale, we selected four typical watershed basins within the study area: the Aral drainage, Issyk-Kul Basin, Balkhash Basin and Tarim River Basin (Figure 1, right).

Figure 1

Sketch map of the study area. Left: including the five countries in Central Asia and the SA (Xinjiang and Tibetan Plateau, China). Right: four river basins (provided by DLR).

Figure 1

Sketch map of the study area. Left: including the five countries in Central Asia and the SA (Xinjiang and Tibetan Plateau, China). Right: four river basins (provided by DLR).

Datasets

In this study, GRACE Level-2 RL05 variable gravity field spherical harmonic coefficient (SHC) data released by CSR (the University of Texas Center for Space Research) agencies (http://www.csr.utexas.edu/grace/) were used to compute the terrestrial water height (Landerer & Swenson 2012) from January 2003 to December 2012. The GRACE data consist of monthly estimates of SHCs that are developed up to degree and order 60.

Hydrology model datasets from GLDAS terrestrial hydrological assimilation data (Rodell et al. 2004b) were processed to validate, explain and enhance the understanding of the GRACE-derived results. Our study first used the results from GLDAS to validate the results from the GRACE data (Rodell et al. 2004a; Tapley et al. 2004b; Chen et al. 2005), and the GLDAS dataset was then used to extend the variations in water storage beyond the GRACE period to determine the water mass changes (Rodell et al. 2004b) in the study area from 1979 to 2012.

Climate data, i.e., temperature and precipitation, released by the University of East Anglia's Climatic Research Unit (Mitchell & Jones 2005) were used to investigate the relationship between WSVs and climate factors. This dataset is provided in 0.5-degree grids every month and is available at http://www.cru.uea.ac.uk/data.

METHODOLOGY

Figure 2 shows the methodology for analyzing the WSVs in the study area using GRACE and GLDAS. Three aspects are included: (1) preprocessing the GRACE, GLDAS and CRU datasets, which includes filtering and GIA (glacial isostatic adjustment model) corrections; (2) estimating and analyzing the WSVs based on the least squares spectral analysis method (two parameters, the linear trends and the period components of the WSVs, were derived from the GRACE and GLDAS datasets); and (3) investigating the spatial-temporal characteristics of the WSVs and the relationship between the WSVs and climate factors.

Figure 2

The WSV analysis methodology for Central Asia and its SA.

Figure 2

The WSV analysis methodology for Central Asia and its SA.

Data pre-processing: combined filter strategy

A key issue of GRACE's data processing is the selection and establishment of a filter strategy (Werth et al. 2009). In this study, a combined filter strategy with a Fan filter of 300 km was proposed based on the error and time series characteristics of a monthly solution gravity model. This filter strategy included a statistical filter, a Fan filter and a de-correlation filter. A statistical filter was used to reduce unnecessary drops while processing the potential coefficient and saving the original signal. The basic steps were as follows. First, a statistical approach, i.e., the F test (Davis et al. 2008; Huang et al. 2012), was used to examine the linear and periodic components in the time series of GRACE SHCs. If the SHCs passed the test, the datasets remained unchanged; otherwise, a Fan filter (Zhang et al. 2009) was applied. All the SHCs were then filtered using the decorrelation method (Swenson & Wahr 2006) to remove the correlated errors. After filtering, the monthly SHCs were converted into equivalent water heights (EWHs) (Wahr et al. 1998) on one degree grids.

Data pre-processing: post-glacial rebound

GRACE uses satellite perturbations to determine the geoid and its changes. Changes in the geoid are not only caused by mass movements and the redistribution of the Earth but are also a consequence of the Earth as an elastic medium that is globally affected by post-glacial rebound (PGR), with changes at different rates in different regions. GRACE measurements are particularly sensitive to PGR (Velicogna & Wahr 2006a, 2006b). Therefore, this effect should be removed when GRACE is used to invert the mass change of the Earth, especially in North America, North Europe, Greenland and Antarctica, whose change rates due to PGR reach 5–11 mm/yr.

The PGR signal in the study area was investigated using the ICE-5G model, which gives global uplift rates. From Figure 3, the PGR change rate in CA and its SA increased from west to east and was less than 0.5 mm/yr, which amounted to less than a 2 mm EWH correction for GRACE. Considering that the geoid accuracy determined by GRACE is approximately 10 mm, the PGR effect can be ignored in this study area.

Figure 3

Uplift prediction map with the ICE-5G model (unit: mm/yr).

Figure 3

Uplift prediction map with the ICE-5G model (unit: mm/yr).

WSV estimation based on least squares spectral analysis

The time series of mass anomalies were obtained on a 1° × 1° grid from the GRACE and GLDAS datasets. These mass anomalies mainly indicate a hydrological signal that has significant annual, semi-annual and linear changes. The least squares spectral analysis was applied to perform the deterministic modeling of the mass anomalies (Vaníček 1969; Wu et al. 1995; Craymer 1998). Thus, the fitting of each grid can be conducted as follows:  
formula
(1)
where H represents the equivalent water height derived from GRACE and GLDAS, t is time, A0 is a constant, B0 is the rate, and f1 and f2 are the annual and semi-annual frequencies. The annual amplitude S1, semi-annual amplitude S2 and corresponding phase φ can be obtained by  
formula
(2)

The amplitude of the fitting result indicates a strong or weak degree of periodic change, and a linear trend indicates increasing or decreasing water mass during a period (Rodell et al. 2004a).

RESULTS VALIDATION AND ANALYSIS

The cross validation of WSVs derived from GRACE and GLDAS datasets

Due to the lack of in situ measurements in Central Asia, cross validations of the WSVs derived from GRACE and GLDAS were conducted to demonstrate the effectiveness of the results.

Figure 4 shows the changes in linear trends from 2003–2012 based on the GRACE (left) and GLDAS data (right). The WSV linear trends for most of the areas showed decreasing tendencies, except for areas such as the Pamir region, Tarim basin, Tibet plateau, Balkhash Basin and parts of Russia. The difference can be mainly attributed to GRACE and GLDAS measurement differences, which will be explained later in this paper.

Figure 4

Linear trends from 2003–2012 based on the GRACE data (left) and the GLDAS data (right).

Figure 4

Linear trends from 2003–2012 based on the GRACE data (left) and the GLDAS data (right).

The correlation coefficients of the WSVs derived from GRACE and GLDAS were used to quantify the effectiveness of the WSVs. Figure 5 shows the spatial variations in the correlation coefficients between the monthly GRACE-based WSVs and the average GLDAS-based WSV.

Figure 5

The spatial pattern of the correlation coefficients between the GRACE-based WSVs and the average WSV from GLDAS simulations (time span from 2003 to 2012).

Figure 5

The spatial pattern of the correlation coefficients between the GRACE-based WSVs and the average WSV from GLDAS simulations (time span from 2003 to 2012).

As shown in Figure 5, the GRACE-born WSV time series indicated a positive correlation with the GLDAS simulations in most locations across the study area. The correlation coefficient was much higher than 0.6 in most of CA and exceeded 0.8 in the Pamirs, Tianshan and northwestern Kazakhstan. In Russia, the correlation coefficient also exceeded 0.8. The correlation coefficients indicate similar changes between the GRACE and GLDAS results. However, there was a distinct negative correlation between Xinjiang and the Tibetan Plateau in China and Mongolia, which could be attributed to the GLDAS model's inability to simulate ice sheet flow and mass balance, which was possibly related to groundwater changes. In these areas, the correlation coefficient value was low; it was reduced to 0.2 and even became negative for Kunlun Mountain, which is located on the Tibetan Plateau. The groundwater contribution to WSVs should not be neglected, but it was not included in the GLDAS simulations (Rodell et al. 2004b; Rodell et al. 2009). Thus, the poor correlation between the GRACE data and the GLDAS results was partially due to the inability of the GLDAS simulations to include surface water and groundwater components, especially in arid regions where groundwater can play a more important role than soil moisture in changes in the TWS. Bias and leakage errors due to the GRACE processing procedure may have also contributed to the poor correlation between the GRACE data and the GLDAS simulation results.

In general, the WSV series estimated from the GRACE-based products closely matched the GLDAS simulations in most locations within the study area, and demonstrated that the WSVs derived from GRACE were effective and could be used to investigate changes in WSVs in Central Asia combined with the WSVs derived from GLDAS datasets from 1979 to 2012. In addition, these results will be indirectly validated by the results of lake, soil moisture and vegetation changes over 30 years derived by our research group.

Interannual variability of water storage from GRACE- and GLDAS-based estimates

The monthly EWH from the GRACE and GLDAS datasets was obtained using Equations (1) and (2). The trend of EWH is calculated on 1-degree grids by using the GRACE and GLDAS datasets. In order to investigate the WSV in a wide range, we assigned the mean value of trends in the grids of the study area as the WSVs. Figure 6 shows the results of using the GRACE data (red) and GLDAS data (blue).

Figure 6

EWH estimated from GRACE products and GLDAS model outputs in the study area. The full color version of this figure is available in the online version of this paper, at http://dx.doi.org/10.2166/ws.2017.206.

Figure 6

EWH estimated from GRACE products and GLDAS model outputs in the study area. The full color version of this figure is available in the online version of this paper, at http://dx.doi.org/10.2166/ws.2017.206.

The series (Figure 6) estimated from the GRACE and GLDAS data show notable decreasing trends in the EWH across the study area. The amplitude of the changes in the EWH in the study area estimated by the GRACE data was approximately −80 to 90 mm over the duration of the research. The interannual variability amplitude of the GRACE data was approximately 80 mm, which was greater than that of the GLDAS data and similar to the results obtained by Yang & Chen (2015) in the Tianshan Mountains and their adjacent areas. The WSV peaks and troughs generated from the GRACE and GLDAS data generally appeared at the same times, with maxima that were typically identified in March and April and minima identified in September, October and November. However, the loss rates indicated by the GLDAS data were greater than those indicated by the GRACE data, especially during 1999–2012 (Figure 6). As shown in Figure 6, there was a notable shift in the GLDAS WSV. A possible reason may be that the land cover map derived from MODIS, which has a higher resolution, was used to generate the GLDAS datasets, and more accurate GLDAS datasets were therefore obtained (Rodell & Houser 2004).

Spatial variabilities of the various estimates determined from GRACE and GLDAS

The least squares spectral analysis was used to capture the linear trends and amplitudes in the WSV. Figure 7 shows the linear trends in the WSV based on the GRACE and GLDAS data for most of the locations in the study area.

Figure 7

Linear trends of the WSV based on (left plot) GRACE data from 2003 to 2012 and based on (right plot) GLDAS data from 1979 to 2012 in Central Asia.

Figure 7

Linear trends of the WSV based on (left plot) GRACE data from 2003 to 2012 and based on (right plot) GLDAS data from 1979 to 2012 in Central Asia.

From Figure 7, for the five Central Asian countries, generally, the WSVs based on the GRACE and GLDAS data both indicate a decreasing trend, and the water mass loss was 6.13 mm/yr and 7.56 mm/yr from 2003 to 2012, respectively. This means that the soil land surfaces of the five Central Asian countries will have a tendency to become desiccated. Soil moisture change results derived by microwave remote sensing over 30 years in Central Asia support this supposition (Li et al. 2015). In addition, the vegetation greenness showed a significant degradation in Central Asia over 20 years from 1992 to 2012 (Zhou et al. 2015), which also might be related to decreases in water storage in Central Asia. For the SA of Central Asia, in the northwestern portion of Kazakhstan, the Caspian Sea area and north Tianshan Mountain area, the WSV showed a significantly decreasing tendency from the GRACE datasets, but increasing trends in the Tibetan Plateau, Tarim River Basin, Balkhash Lake basin, middle border between Uzbekistan and Turkmenistan and part of Russia were observed from the GRACE datasets. For example, a water mass increase of 15.45 mm/yr was detected by GRACE over the Tibetan Plateau from 2003 to 2012. In the Caspian Sea, GRACE revealed a loss rate of 25.86 mm/yr. From the GLDAS datasets, most of the area also showed a decreasing tendency, except for the Qinghai-Tibetan Plateau, Tarim River Basin and northern parts of Pakistan and India. From Figure 7, the WSVs based on the GRACE data were much more spatially heterogeneous than the WSVs based on the GLDAS data, which can be attributed to the fact that GRACE can map total water storage changes (such as soil moisture, lakes, rivers, groundwater and seasonal snow), but GLDAS can only capture soil moisture and seasonal snow changes. In addition, the linear trends of the WSVs based on the GRACE data were greater than those based on the GLDAS data. For example, in the northwestern region of Kazakhstan, the WSV based on the GRACE data decreased from 10 to 15 mm/yr (see Figure 7, left), and the WSV based on the GLDAS data decreased from 8 to 10 mm/yr (see Figure 7, right).

The results of the least squares spectral analysis indicated that the annual amplitudes of the GRACE and GLDAS data varied in space and time in different periods (see Figure 8). The annual amplitude based on the GRACE data indicates a more positive change compared to that based on the GLDAS data. For example, an annual amplitude of greater than 100 mm in the Pamirs was detected by GRACE, and GLDAS's amplitude was less than 40 mm. The GLDAS annual amplitude in the 2000s was much lower than the annual amplitude in the 1990s across the study area, especially in Kazakhstan. In northern Kazakhstan, the annual amplitude decreased by more than 30 mm of EWH (see Figure 8(a) and 8(b)). In the Caspian Sea, the annual amplitude decreased during both periods but to a lesser extent in the 2000s. The GRACE and GLDAS datasets also showed marked differences in the Pamirs from 2005 to 2008 (see Figure 8(c) and 8(d)). The annual fluctuations in the GRACE and GLDAS data were considerable, and the annual fluctuations were much higher for the GRACE data than for the GLDAS data. In a large area such as CA, hydrological signals are dominated by soil moisture. This can explain the similarity between the GRACE and GLDAS datasets. The discrepancies in the linear rates and amplitudes of the GRACE and GLDAS datasets can be attributed to changes in the water masses of lakes, rivers and groundwater.

Figure 8

The annual amplitudes of WSVs derived by least square spectral analysis for different periods. (a) GLDAS 1992–1995; (b) GLDAS 2002–2005; (c) GLDAS 2005–2008; and (d) GRACE 2005–2008.

Figure 8

The annual amplitudes of WSVs derived by least square spectral analysis for different periods. (a) GLDAS 1992–1995; (b) GLDAS 2002–2005; (c) GLDAS 2005–2008; and (d) GRACE 2005–2008.

DISCUSSION

There were notable spatial differences in the WSVs variations in CA from the above analysis. In this section, we thus focus on the discussion of why such spatial differences exist and what the main impact factors are.

Effects of glacier on WSVs in typical regions and basins

Substantial changes in the cryosphere of Central Asia and its SA due to global warming have been observed (Hagg et al. 2013; Yang et al. 2014). The glacier mass balance in the study area can be estimated using GRACE data, but the results vary depending on how the GRACE data are processed (Matsuo & Heki 2010; Jacob et al. 2012).

Based on our proposed filter strategy, the glacier mass balance was estimated in the Tianshan, Pamirs and Karakoram following Jacob et al. (2012). GLDAS's signal from GRACE was removed, as shown in Figure 9, and the melting rates of the glaciers in the Tianshan, Pamirs and Karakoram were −1.36, −3.98 and −7.32 mm/yr (see Figure 9(c)9(e)). The water mass changes on the Tibetan Plateau (TP) and in the Caspian Sea were greater than those in other regions, as shown in Figures 7 and 9. From 1979 to 2012, the WSV in the Tibetan Plateau and Karakoram increased at a rate of more than 10 mm/yr (see Figure 7, right). The Tibetan Plateau has experienced overall surface air warming and increases in humidity since the 1980s, which triggered more convective precipitation over the central TP (Yang et al. 2014). The average rate of WSV based on the GRACE data was 15.45 mm/yr in the TP (see Figure 9(a)). The rate was much higher in the northern Tibetan Plateau. The results of some studies have indicated that the water mass increase in the TP can be attributed to the increasing number and levels of glacial lakes (Song et al. 2013; Zhang et al. 2013).

Figure 9

WSVs (expressed in EWH, unit: mm) on (a) the Tibetan Plateau, with a GRACE (red) rate of 15.45 mm/yr and GLDAS (blue) rate of −4.11 mm/yr; (b) the Caspian Sea, with a GRACE rate of 25.86 mm/yr; mass balance of glaciers (subtracting values based on GLDAS data from values based on GRACE data) in the (c) Tianshan, (d) Pamirs, and (e) Karakoram, with rates of −1.36, −3.98 and −7.32 mm/yr, respectively. The full color version of this figure is available in the online version of this paper, at http://dx.doi.org/10.2166/ws.2017.206.

Figure 9

WSVs (expressed in EWH, unit: mm) on (a) the Tibetan Plateau, with a GRACE (red) rate of 15.45 mm/yr and GLDAS (blue) rate of −4.11 mm/yr; (b) the Caspian Sea, with a GRACE rate of 25.86 mm/yr; mass balance of glaciers (subtracting values based on GLDAS data from values based on GRACE data) in the (c) Tianshan, (d) Pamirs, and (e) Karakoram, with rates of −1.36, −3.98 and −7.32 mm/yr, respectively. The full color version of this figure is available in the online version of this paper, at http://dx.doi.org/10.2166/ws.2017.206.

For the Caspian Sea, Swenson and Wahr concluded that the water level of the Caspian Sea increased for the period 2002–2006, as shown in the results from GRACE (RL04 data) and Jason-1. The GRACE RL05 data also indicated that the Caspian Sea gained water at a rate of 42.67 mm/yr from 2003 to 2005 (see Figure 9(b)). We found that the water level substantially retreated after 2006. The results indicate that the changes in the level of the Caspian Sea should be considered to be a multistage process that is affected by geological factors, hydro-climatic factors and water balance as well as anthropogenic factors, and the hydro-climatic changes have a dominant influence on the Caspian Sea level (Ramiz 2015).

Hydrological signals are stronger at the basin scale. The time series of the WSV in the four basins (see Figure 1) are shown in Figure 10, and the least squares spectral analysis results are summarized in Table 1. From 2003 to 2012, the linear trends derived in the four basins using the GLDAS data were negative. However, the trends derived using the GRACE data indicated that the WSV increased in the Tarim River Basin from 2003 to 2012. For the longer time series based on the GLDAS data (see Figure 10, right), the first three basins also indicated decreasing trends at rates of −3.3, −1.2 and −3.5 mm/yr. However, the GLDAS data indicated an increasing rate of 3.4 mm/yr in the Tarim River Basin, especially for the period 2000–2006 (see Figure 10, bottom right). Snow and glaciers play an important role in stream flow regimes in the Tarim River Basin, and there is minimal summer precipitation. Analyses based on MODIS data for the period 2000–2007 confirm that the decrease in snow cover duration is continuing and that snowmelt occurs earlier currently (Khalsa & Aizen 2008). Intensified snowmelt in regions surrounding the Tarim River Basin may strongly affect the quantity and seasonal distribution of local runoff, and increasing runoff has been observed in the Tarim River in past decades (Tao et al. 2011). Yang et al. (2015) concluded that increasing snow/ice melt water draining into the Tarim River Basin may be the underlying cause of the increasing WSV (Yang et al. 2015).

Table 1

The results of the least squares spectral analysis for four basins (see Figure 1, right)

Basin Result Annual Semiannual Rate (mm/yr) 
    A (mm) P (month) A (mm) P (month) 79-12 03-12 03-04 05-08 09-12 
Aral GRACE 61 −1.20 −0.04  4.3 15.7 28.1 1.9 
GLDAS 25 −0.62 −1.40 3.3 6.5 −3.9 16.3 −2.9 
Issyk-Kul GRACE 52 −1.08 −0.61  3.8 17.9 30.2 13.4 
GLDAS 33 −0.35 2.41 1.2 6.6 −5.0 20.1 0.9 
Balkhash GRACE 42 −1.04 −0.05  2.1 −11.9 18.4 17.8 
GLDAS 29 −0.10 2.30 3.5 4.8 −9.9 16.3 1.7 
Tarim GRACE 22 −2.84 0.52  2.4 −13.2 10.8 12.6 
GLDAS 24 −1.72 −2.21 3.4 3.9 18.9 15.8 −3.4 
Basin Result Annual Semiannual Rate (mm/yr) 
    A (mm) P (month) A (mm) P (month) 79-12 03-12 03-04 05-08 09-12 
Aral GRACE 61 −1.20 −0.04  4.3 15.7 28.1 1.9 
GLDAS 25 −0.62 −1.40 3.3 6.5 −3.9 16.3 −2.9 
Issyk-Kul GRACE 52 −1.08 −0.61  3.8 17.9 30.2 13.4 
GLDAS 33 −0.35 2.41 1.2 6.6 −5.0 20.1 0.9 
Balkhash GRACE 42 −1.04 −0.05  2.1 −11.9 18.4 17.8 
GLDAS 29 −0.10 2.30 3.5 4.8 −9.9 16.3 1.7 
Tarim GRACE 22 −2.84 0.52  2.4 −13.2 10.8 12.6 
GLDAS 24 −1.72 −2.21 3.4 3.9 18.9 15.8 −3.4 

Note: A is the amplitude, and P is the phase. The rates in different periods were estimated using data in the corresponding periods. The linear trends of the four basins varied in different periods (see Table 1). It can be inferred from Figure 10 and Table 1 that the GRACE and GLDAS datasets showed stronger decreasing trends from 2005 to 2008.

Figure 10

WSVs (expressed in EWH, unit: mm) in four basins (see Figure 1, right).

Figure 10

WSVs (expressed in EWH, unit: mm) in four basins (see Figure 1, right).

As shown in Table 1, the GRACE and GLDAS datasets varied in their annual amplitudes, with the exception of the linear trends. In general, the GRACE data's annual amplitudes were much greater than those of the GLDAS data, except for the annual amplitude for the Tarim River Basin. This is easily understood, because the total water mass signal based on the GRACE data was stronger than the soil moisture data from the GLDAS dataset. For example, the annual amplitude for the Aral drainage based on the GRACE data was almost 2.5 times higher than that based on the GLDAS data (see Table 1). This can be explained by runoff and lake water, which were not included in the GLDAS simulations.

Effects of climate change on WSVs

According to high resolution regional climate models implemented for CA, temperatures are projected to increase up to 7 °C until the end of the 21st century (Mannig et al. 2013). The effects of human activities are excluded in the consideration of these data. The warming trend will aggravate the retreat of glaciers and the loss of surface waters over CA. Precipitation is another important contributing factor for WSVs. Many drought events in the basins have been triggered by extreme reductions in precipitation (Chen et al. 2009). We have analyzed the interannual variabilities in the precipitation and temperature for our research period to evaluate the causes of the interannual variability in water storage based on the GLDAS data.

Effects of temperature on WSVs

Similar to the meteorological station and reanalysis data, the CRU temperature data for 1979–2012 also demonstrated that the five countries in CA experienced warming trends (see the red line in Figure 11). In Figure 11, the temperature represents the average of grids in the five countries in CA, and the temperature data for 2004 meet the Mann-Kendall test for trends, which indicates that the warming trend was significant. The time series based on the GLDAS data for the five countries in CA in 2002 meets the Mann-Kendall test for trends (see Figure 11). The significance statistic for temperature in 2002 was 1.94, which approximated the significance level (see Figure 11). Moreover, except some years (1984–1987), the statistical values of the temperature have been greater than zero since 1980 (see Figure 11), which indicate that the temperature has an obvious rising trend. By contrast, the statistical values of water reserve change have been less than zero except in 1988, which indicates the water reserve has been reducing all the time. This relationship between temperature and the GLDAS data implied that warming trends had a potential influence on WSVs.

Figure 11

The Mann-Kendall test for trends for the GLDAS data (blue), temperature (red) and precipitation (green); the significance level was 0.05, and the critical value was ±1.96. The statistic value of temperature in 2002 was 1.94 and meets the test for trends in 2004; the GLDAS data meet the test for trends in 2002. The full color version of this figure is available in the online version of this paper, at http://dx.doi.org/10.2166/ws.2017.206.

Figure 11

The Mann-Kendall test for trends for the GLDAS data (blue), temperature (red) and precipitation (green); the significance level was 0.05, and the critical value was ±1.96. The statistic value of temperature in 2002 was 1.94 and meets the test for trends in 2004; the GLDAS data meet the test for trends in 2002. The full color version of this figure is available in the online version of this paper, at http://dx.doi.org/10.2166/ws.2017.206.

The changes in the glaciers in CA were influenced more by climatic factors, especially temperature (Hagg et al. 2013; Yang et al. 2014). To demonstrate the effect of temperature here, two regions, Kunlun Mountain (R01, 37–39N, 88–89E) and the eastern Pamirs (R02, 36–38N, 74–76E), which are adjacent to the TP and the Tarim River Basin, respectively, were selected for this study. These two selected regions are covered by ice and snow, which supply water resources to the SA. The CRU data confirmed that the temperatures of R01 and R02 increased at rates of 0.46 °C and 0.39 °C/decade, respectively (see Figure 12). Although the warming trend was notable, the highest annual mean temperature during 1979–2012 was below the freezing point. This warming trend may not accelerate glacial melting. Actually, the shrinkage of glaciers is highly related to the trend of increasing temperatures in the summer (Kriegel et al. 2013). The CRU also revealed that R01 and R02 have experienced increasing trends in June, July and August at rates of 0.49 °C and 0.29 °C/decade, respectively (see Figure 12).

Figure 12

The annual mean value (left) and summer (June, July and August) mean value (right) for the CRU temperatures over Kunlun Mountain (R01, 37–39N, 88–89E, top), which is adjacent to the northern part of the Tibetan Plateau, and the eastern Pamirs (R02, 36–38N, 74–76E, bottom), which are adjacent to the Tarim River Basin.

Figure 12

The annual mean value (left) and summer (June, July and August) mean value (right) for the CRU temperatures over Kunlun Mountain (R01, 37–39N, 88–89E, top), which is adjacent to the northern part of the Tibetan Plateau, and the eastern Pamirs (R02, 36–38N, 74–76E, bottom), which are adjacent to the Tarim River Basin.

Effects of precipitation on WSVs

Precipitation has an important effect on WSVs (Chen et al. 2009). As indicated by the CRU data, the precipitation over the five countries in CA was stable during the period of the study (see Figure 11). No abnormal change was detected by the Mann-Kendall test. However, we cannot conclude that precipitation did not contribute to the WSVs over the study area.

The precipitation anomalies were obtained by subtracting the mean precipitation value from 2003 to 2012. The precipitation anomalies in the four basins (see Figure 1, right) were averaged and were determined to be consistent with the GRACE and GLDAS data in the Aral drainage, Issyk-Kul and Balkhash basins (see Figure 13). In these three basins, the CRU precipitation decreased during 2005–2008, similar to the GRACE and GLDAS data. The precipitation anomaly, GRACE data and GLDAS data all indicated minimum precipitation in 2008. A possible reason was that a drought event might have occurred during that year. The change pattern in the Tarim River Basin was different from that in other basins; the precipitation anomaly reached a minimum in 2009. The fluctuations in the GRACE data, GLDAS data and precipitation anomaly were also different. A possible reason is that the geographical environment in the Tarim River Basin is more complicated than the environments in other basins because the large Taklimakan Desert is located in this basin. It is noteworthy that the magnitude of the precipitation anomaly was lower than the magnitudes indicated by the GRACE data and the GLDAS data. In addition, other contributions from runoff, lake water and groundwater cannot be neglected. Generally, at the basin scale, the precipitation anomalies were very similar to the GRACE and GLDAS data, which can be viewed as an indicator of WSVs.

Figure 13

The changes in the GRACE data (blue), GLDAS data (green) and precipitation anomalies (red) in the four basins; the precipitation anomaly was calculated by subtracting the mean precipitation value from 2003 to 2012. The full color version of this figure is available in the online version of this paper, at http://dx.doi.org/10.2166/ws.2017.206.

Figure 13

The changes in the GRACE data (blue), GLDAS data (green) and precipitation anomalies (red) in the four basins; the precipitation anomaly was calculated by subtracting the mean precipitation value from 2003 to 2012. The full color version of this figure is available in the online version of this paper, at http://dx.doi.org/10.2166/ws.2017.206.

CONCLUSIONS

Based on GRACE, GLDAS and CRU meteorological datasets, variations in water storage in Central Asia and its SA over 30 years were investigated in this paper.

The result indicates the following. (1) The water storage values obtained from GRACE and GLDAS over CA generally exhibited a decreasing tendency, and the results from GRACE over CA were spatially more heterogeneous than those from GLDAS. (2) For CA's SA, the water storage presented a different change tendency due to the effects of natural factors such as snow or glacier melting. (3) The water losses in the five CA countries may be caused by warming, which will lead to the loss of soil moisture, and furthermore, the water increases in the TP and Tarim basin may be caused by glacier melting around the Pamirs and Himalaya. At the basin scale, the precipitation anomalies were very similar to the GRACE and GLDAS data, which can be viewed as an indicator of WSVs.

Due to a lack of in situ measurements related to water resource changes, a full quantitative analysis and assessment of the change tendency of WSVs was not conducted in this paper. In addition, because human and social economic data were not included in this study, the mechanisms of changes in water storage were not fully explained and understood. In the future, to further investigate water storage changes under a background of global change, an observational network needs to be constructed, and more datasets need to be collected to fully understand tendencies in water storage changes and their mechanisms and how these changes affect the ecological environment of Central Asia and its surroundings.

ACKNOWLEDGEMENTS

This work was supported by the 13th Five Year Plan Project of the Chinese Academy of Sciences (Grant No. Y6SG040030 and Y6SG0400CX), Bureau of International Co-operation, Chinese Academy of Sciences (Grant No. 131C11KYSB20160061), National Key Research and Development Program of China (Grant No. 2016YFA0600304) and Key Laboratory of Tibetan Environment Changes and Land Surface Processes, Chinese Academy of Sciences. We would like to thank CSR, NOAA and CRU for providing the free data, and we are grateful to Dr Haoming Yan from the Institution of Geodesy and Geophysics, Chinese Academy of Sciences, for helping us process the GLDAS data.

REFERENCES

REFERENCES
Aizen
E. M.
,
Aizen
V. B.
,
Melack
J. M.
,
Nakamura
T.
&
Ohta
T.
2001
Precipitation and atmospheric circulation patterns at mid-latitudes of Asia
.
J. Climatol.
21
(
5
),
535
556
.
Bernauer
T.
&
Siegfried
T.
2012
Climate change and international water conflict in Central Asia
.
J. Peace Res
49
(
1
),
227
239
.
Chen
J.
,
Wilson
C.
,
Tapley
B.
,
Yang
Z.
&
Niu
G.
2009
2005 drought event in the Amazon River basin as measured by GRACE and estimated by climate models
.
J. Geophys. Res.
114
(
B5
).
Craymer
M.
1998
The Least Squares Spectrum, Its Inverse Transform and Autocorrelation Function: Theory and Applicaitons in Geodesy
.
PhD Thesis
,
Graduate Department of Civil Engineering, University of Toronto
.
Davis
J. L.
,
Tamisiea
M. E.
,
Elosegui
P.
,
Mitrovica
J. X.
&
Hill
E. M.
2008
A statistical filtering approach for Gravity Recovery and Climate Experiment (GRACE) gravity data
.
J. Geophys. Res.
113
,
B04410
.
Famiglietti
J. S.
2004
Remote sensing of terrestrial water storage, soil moisture and surface waters
. In:
The State of the Planet: Frontiers and Challenges in Geophysics
(
Sparks
R. S. J.
&
Hawkesworth
C. L.
, eds).
American Geophysical Union
,
Washington, DC, USA
, pp.
197
207
.
Gessner
U.
,
Naeimi
V.
,
Klein
I.
,
Kuenzer
C.
,
Klein
D.
&
Dech
S.
2013
The relationship between precipitation anomalies and satellite-derived vegetation activity in Central Asia
.
Global Planet Change
110
,
74
87
.
Hagg
W.
,
Mayer
C.
,
Lambrecht
A.
,
Kriegel
D.
&
Azizov
E.
2013
Glacier changes in the Big Naryn basin, Central Tian Shan
.
Global Planet Change
110
,
40
50
.
Horsman
S.
2001
Water in Central Asia: Regional Cooperation or Conflict
.
Central Asian Security
,
Brookings Institution Press and Chatham House
,
Washington, DC, USA
, pp.
70
94
.
Huang
J.
,
Halpenny
J.
,
van der Wal
W.
,
Klatt
C.
,
James
T. S.
&
Rivera
A.
2012
Detect-ability of groundwater storage change within the Great Lakes Water Basin using GRACE
.
J. Geophys. Res.
117
,
B08401
.
Jacob
T.
,
Wahr
J.
,
Pfeffer
W. T.
&
Swenson
S.
2012
Recent contributions of glaciers and ice caps to sea level rise
.
Nature
482
(
7386
),
514
518
.
Khalsa
S. J. S.
&
Aizen
V. B.
2008
Variability in Central Asia seasonal snow cover during the MODIS period of record
. In:
Proceedings of 2008 IEEE International Geoscience & Remote Sensing Symposium
,
vol. 10
, pp.
1
2
.
Kriegel
D.
,
Mayer
C.
,
Hagg
W.
,
Vorogushyn
S.
,
Duethmann
D.
,
Gafurov
A.
&
Farinotti
D.
2013
Changes in glacierisation, climate and runoff in the second half of the 20th century in the Naryn basin, Central Asia
.
Global Planet. Change
110
,
51
61
.
Landerer
F.
&
Swenson
S.
2012
Accuracy of scaled GRACE terrestrial water storage estimates
.
Water. Resour. Res.
48
(
4
),
1
11
.
Li
X.
,
Gao
X.
,
Wang
J.
&
Guo
H.
2015
Microwave soil moisture dynamics and response to climate change in Central Asia and Xinjiang Province, China, over the last 30 years
.
Journal of Applied Remote Sensing
9
(
1
),
096012
.
Mannig
B.
,
Müller
M.
,
Starke
E.
,
Merkenschlager
C.
,
Mao
W.
,
Zhi
X.
,
Podzun
R.
,
Jacob
D.
&
Paeth
H.
2013
Dynamical downscaling of climate change in Central Asia
.
Global Planet. Change
110
,
26
39
.
Matsuo
K.
&
Heki
K.
2010
Time-variable ice loss in Asian high mountains from satellite gravimetry
.
Earth Planet. Sc. Lett.
290
(
1
),
30
36
.
Mughal
M. A. Z.
2013
Persian gulf desert and semi-desert
.
Biomes & Ecosystems
3
,
1000
1002
.
Parry
M. L.
2007
Climate Change 2007: Impacts, Adaptation and Vulnerability: Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change
,
vol. 4
.
Cambridge University Press
,
Cambridge, UK
.
Qi
J.
&
Evered
K.
2008
Environmental Problems of Central Asia and Their Economic, Social and Security Impacts
.
Springer Science & Business Media
,
New York, NY
.
Ramillien
G.
,
Famiglietti
J. S.
&
Wahr
J.
2008
Detection of continental hydrology and glaciology signals from GRACE: a review
.
Surv. Geophys.
29
(
4–5
),
361
374
.
Ramiz
M.
2015
Impact of climate changes on the Caspian sea level
.
Journal of Resources and Ecology
6
(
2
),
87
92
.
Rodell
M.
&
Houser
P. R.
2004
Updating a land surface model with MODIS derived snow cover
.
Journal of Hydrometeorology
5
(
6
),
1064
1075
.
Rodell
M.
,
Famiglietti
J.
,
Chen
J.
,
Seneviratne
S.
,
Viterbo
P.
,
Holl
S.
&
Wilson
C.
2004a
Basin scale estimates of evapotranspiration using GRACE and other observations
.
Geophys. Res. Lett.
31
(
20
),
183
213
.
Rodell
M.
,
Houser
P.
,
Jambor
U. e. a.
,
Gottschalck
J.
,
Mitchell
K.
,
Meng
C.
,
Arsenault
K.
,
Cosgrove
B.
,
Radakovich
J.
&
Bosilovich
M.
2004b
The global land data assimilation system
.
B. Am. Meteorol. Soc.
85
(
3
),
381
394
.
Rodell
M.
,
Velicogna
I.
&
Famiglietti
J. S.
2009
Satellite-based estimates of groundwater depletion in India
.
Nature
460
(
7258
),
999
1002
.
Siegfried
T.
,
Bernauer
T.
,
Guiennet
R.
,
Sellars
S.
,
Robertson
A. W.
,
Mankin
J.
&
Bauer–Gottwein
P.
2010
Will climate change exacerbate or mitigate water stress in Central Asia?
In:
AGU Fall Meeting Abstracts
,
vol. 1
, pp.
1250
.
Sun
Q.
,
Tashpolat
T.
,
Ding
J.
,
Zhang
F.
&
Mamat
S.
2014
GRACE data-based estimation of spatial variations in water storage over the Central Asia during 2003–2013
.
Acta Astronomica Sinica
55
,
498
511
.
Swenson
S.
&
Wahr
J.
2006
Post-processing removal of correlated errors in GRACE data
.
Geophys. Res. Lett.
33
(
8
),
1
4
.
Tapley
B. D.
,
Bettadpur
S.
,
Ries
J. C.
,
Thompson
P. F.
&
Watkins
M. M.
2004a
GRACE measurements of mass variability in the earth system
.
Science
305
,
503
505
.
Tapley
B. D.
,
Bettadpur
S.
,
Watkins
M.
&
Reigber
C.
2004b
The gravity recovery and climate experiment: mission overview and early results
.
Geophys. Res. Lett.
31
(
9
),
L09607
.
Unger-Shayesteh
K.
,
Vorogushyn
S.
,
Farinotti
D.
,
Gafurov
A.
,
Duethmann
D.
,
Mandychev
A.
&
Merz
B.
2013
What do we know about past changes in the water cycle of Central Asian headwaters? A review
.
Global Planet. Change
110
,
4
25
.
Vaníček
P.
1969
Approximate spectral analysis by least-squares fit
.
Astrophysics and Space Science
4
,
387
391
.
Werth
S.
,
Güntner
A.
,
Schmidt
R.
&
Kusche
J.
2009
Evaluation of GRACE filter tools from a hydrological perspective
.
Geophy. J. Int.
179
(
3
),
1499
1515
.
Wu
D. L.
,
Hays
P. B.
&
Skinner
W. R.
1995
A least squares method for spectral analysis of space-time series
.
Journal of the Atmospheric Sciences
52
(
20
),
3501
3511
.
Xu
M.
,
Ye
B. S.
,
Zhao
Q. D.
&
Wang
J.
2013
Spatiotemporal change of water reserves in the Tianshan mountains, Xinjiang based on GRACE
.
Arid Zone Research
3
,
003
.
Yang
K.
,
Wu
H.
,
Qin
J.
,
Lin
C.
,
Tang
W.
&
Chen
Y.
2014
Recent climate changes over the Tibetan Plateau and their impacts on energy and water cycle: a review
.
Global Planet. Change.
112
,
79
91
.
Yang
T.
,
Wang
C.
,
Chen
Y.
,
Chen
X.
&
Yu
Z.
2015
Climate change and water storage variability over an arid endorheic region
.
J. Hydrol.
529
,
330
339
.
Zhang
Z.-Z.
,
Chao
B. F.
,
Lu
Y.
&
Hsu
H.-T
, .
2009
An effective filtering for GRACE time-variable gravity: Fan filter
.
Geophys. Res. Lett.
36
,
L17311
.
Zhang
G.
,
Yao
T.
,
Xie
H.
,
Kang
S.
&
Lei
Y.
2013
Increased mass over the Tibetan Plateau: from lakes or glaciers?
Geophys. Res. Lett.
40
(
10
),
2125
2130
.
Zhou
Y.
,
Zhang
L.
,
Fensholt
R.
,
Wang
K.
,
Vitkovskaya
I.
&
Tian
F.
2015
Climate contributions to vegetation variations in Central Asian drylands: pre- and post-USSR collapse
.
Remote Sens.
7
,
2449
2470
.