Environmental Management

, Volume 51, Issue 1, pp 138–153

Progress, Challenges and Prospects of Eco-Hydrological Studies in the Tarim River Basin of Xinjiang, China

  • Yaning Chen
  • Changchun Xu
  • Yapeng Chen
  • Yongbo Liu
  • Weihong Li

DOI: 10.1007/s00267-012-9823-8

Cite this article as:
Chen, Y., Xu, C., Chen, Y. et al. Environmental Management (2013) 51: 138. doi:10.1007/s00267-012-9823-8


Eco-hydrological research in arid inland river basins has been a focus of geologists and ecologists as it is crucial for maintaining the sustainable development of socio-economy, particularly in ecologically vulnerable areas. Based on the research work carried out in the Tarim River basin of Xinjiang, northwestern China, this paper summarizes synthetically the climate change and associated responses of water resources in the mountainous area, land use and land cover in the oasis, and plants responding to environmental stresses in the desert area of the river basin. Research gaps, challenges, and future perspectives in the eco-hydrological studies of the Tarim River basin are also discussed.


Eco-hydrology Water resources management Tarim River 


Eco-hydrological processes at the basin scale, due to the wide and complicated interactions among geography, environment, hydrology, climate and ecology, have been a focus of many international research programs (IGBP, MAB, IHP, and the UNESCO’S Eco-hydrology). These processes are also key influencing factors for integrated river basin management, especially for water resources management (Alaa and others 2008; Braga 2001; Sokile and others 2003; Gaiser and others 2008). In China, research on eco-hydrological processes have been carried out since early 1990s and mainly in the areas of: (1) forest eco-hydrological processes in headwater regions centering on the problem of water conservation (Jin and others 2003; Chen and Zhao 2008; Zhao and Chen 2001b; Song and others 2003; Wei and others 2011); (2) relative detailed examination of water resources focusing on the mechanism of water use and the responses of typical plants to the environmental stress in arid regions. (3) components and modeling of evapotranspiration (Hou and others 2010; Wu and others 2005; Si and others 2005); (4) surface water quality (Li and others 2010a; Feng and others 2004; Zhu and others 2008); (5) runoff formation, discharge generation and water pathways (Wang and others 2009; Chen and others 2006a; Xi and others 2010); and (6) runoff modeling (Gao and others 2004; Ouyang and others 2007; Li and Williams 2008).

The Tarim River basin is the largest inland river basin in China with unique ecological and hydrological processes. The basin has abundant natural resources and a fragile ecological environment. It is also well known for its environmental problems. The ecological degeneration in the lower reaches of Tarim River fundamentally relates to water shortage. In the river basin, water runs through the processes from the development and transformation in high mountains to the consumption of water resources in the oasis, and to the loss of water resources in the downstream desert. Challenges involved in these processes are typically the development, utilization and protection of water resources in upper and middle reaches and the maintenance of ecology in lower reaches of the river. Water resources play a key role in the sustainable development of local social and economic development and the eco-environmental protection. It is of great importance to conduct research on the eco-hydrological processes in the Tarim River basin. Much fundamental work has been done in the past including research on water resources management and ecological protection (Huang and others 2010; Chen and others 2003a; Li and others 2009), land-use and land-cover change and their environmental effects (Wu and Meng 2004a; Do and others 2008), climate change (Chen and Xu 2005; Xu and others 2006; Xu and others 2009; Zhong and others 2007), water resources variation and human activities (Wu and Cai 2004b; Xu and others 2004; Hao and others 2009a), responses of vegetation, groundwater level and water quality to the ecological water conveyance (Chen and others 2010; Chen and others 2009; Ye and others 2009; Li and others 2010a), and the suitable groundwater level for plant growth and ecological water demand (Hao and others 2009b; Ye and others 2010). However, most of the previous studies were concentrated on specific processes but were lacking in integrated research on all processes involved. Based on the long-term work carried out by our research group in the Tarim River, this paper summarizes synthetically the progress on the issues of climate change, water resources variation, land use and land cover, and the ecological responses of plants to the environmental stresses in the river basin, from which future directions in research on eco-hydrological processes of the river basin are proposed. The aims of this paper are to reveal the responses of climate and hydrological cycle to changing environment, to interpret the relationship between groundwater and vegetation of desert riparian forests, to identify the suitable ecological water level and the minimum ecological water demand for natural restoration of the desert ecosystem, and finally to provide scientific support for restoration of degenerated ecosystems, sustainable management of environment, and efficient use and allocation of water resources of the Tarim River Basin. Findings reported in this paper would be useful in studies of resource development and ecological rehabilitation in other arid inland river basins.

Study Area

Tarim River is located in the northern border of Taklimakan Desert, which is the largest desert in China and the second largest desert in the world. The watershed covers an area of 1.02 × 106 km2. It is formed by 144 rivers belonging to 9 river systems surrounding the Tarim Basin including Aksu River, Kashgar River, Yarkant River, Hotan River, Keriya River, Qargan River, Kaidu-Konqi River, Dina River, and Weigan River (Fig. 1). The average annual runoff at the inlet is 3.98 × 1010 m3 during 1957–2005 with main source of water from snowmelt and precipitation in the surrounding high mountains. Because of the influence of human activities and climate change, surface water of the Qargan River, Keriya River and Dina River had lost their connections to the mainstream successively before 1940s. Since then, the Kashgar River, Kaidu-Konqi River, and Weigan River were also gradually separated from the mainstream. At present, only three headstream rivers, Aksu River, Yarkant River and Hotan River, have surface water supply to the Tarim River mainstream. In addition, water from the Bosten Lake is transported to the irrigated areas in the lower reaches of the Tarim River through Kuta Trunk Canal. Therefore, the current Tarim River basin has a water system pattern of “Four headstreams and one mainstream” (Song and others 2000).
Fig. 1

Sketch map of the nine river systems in the Tarim River basin

The Tarim River mainstream areas, from Alar to Taitema Lake with 1,321 km in length, do not generate runoff. Instead, all water is supplied from the headstreams. From the hydrological perspective, Tarim River is a typical dissipative river in the arid region. Over the past 50 years, natural runoff in the headstreams has shown an increasing trend. However, water supply to the mainstream has decreased at an average rate of 3.2 × 108 m3 per decade due to the increase of piedmont irrigation areas and water consumption in the headstream regions. Moreover, the construction of reservoirs, flood overflows, and the unreasonable water channeling at upper and middle reaches of the mainstream, have caused a serious water shortage in the lower reaches. The imbalance of water consumption had led to a sharp decline of downstream water volume year after year, an eventual dry-up of the two tail lakes (Lake Lop Nor in 1970 and Lake Tetma in 1972), a complete cutoff of 321 km downstream river reach, a dramatic decrease of groundwater level, a total deterioration of desert vegetation dominated by poplar (Populus euphratica) groves, as well as exacerbating desertification, severely damaging biodiversity, critically endangering regional socio-economic sustainable development and adversely affecting human living conditions. With respect to water resources development and ecosystem protection, the lower reach of the Tarim River has become the most prominent hot spot in arid regions of western China.

Climate Change

Long-Term Trends of Temperature and Precipitation

Based on the temperature and precipitation data collected at 19 observation stations during 1958–2002, Xu and others (2008) detected abrupt changes and long-term trends of climate in the Tarim River basin using the Mann-Whitney and Mann-Kendall non-parametric methods. Their results showed an obvious increase in both temperature and precipitation. The sharp increasing trend started from the mid-1980s and has continued ever since (Fig. 2). The 1990s is the warmest and most humid decade (Yang and He 2003; Wang and others 2006a) over the past half century, and is synchronized with the climate change of the entire Xinjiang region (Shi and others 2002; Hu and others 2002). The overall increase in temperature and precipitation in the basin is obvious but with a distinct spatial and temporal variability. Temporally temperature increased mainly in winter and autumn, and precipitation mainly increased in summer. Spatially temperature increased in the whole basin, while precipitation increased intensively in the headstream areas, e.g., the Aksu River, Yarkant River and Kaidu River basins (Xu and others 2010). This mitigated to some extent the evapotranspiration caused by increasing temperature and prevented the drought aggravation. Over all, the entire basin showed a warming and humidifying trend, although some areas, especially the lower reaches of the Tarim River, are still experiencing drought.
Fig. 2

Anomaly variation of the annual precipitation and average annual temperature based on data collected at 19 observation stations in the Tarim River basin during 1958–2002

Correlation Between Temperature/Precipitation and El-Nino Southern Oscillation (ENSO)

Chen and others (2007) conducted a correlation analysis between annual temperature/precipitation and the Southern Oscillation Index (SOI) during 1957–2003. Their results showed that both parametric and nonparametric tests agreed with the hypothesis H0, that is there is no significant increasing trend, suggesting that there was no significant correlation between SOI and annual temperature/precipitation. Further \( \chi^{2} \)independence test was employed to investigate the correlation between the temperature/precipitation of four seasons and the two phases (El Nino and La Nina) of ENSO (Xu 2007). Results (Table 1) showed that ENSO affected the climate of the study area in certain seasons. During the El Nino years, autumn temperature was generally higher than that in normal years and summer precipitation was also on the high side. This is different from the findings in the east part of northwestern China that higher temperature and less precipitation appeared in the El Nino years reported by Li and Li 2004 and Ye and others 2006. In the La Nina years, temperature decreased in the basin showing the same trend as that in the east part of northwestern China. However, unlike the increased precipitation observed in the east, the precipitation in the basin did not show a significant change.
Table 1

Correlation between the seasonal temperature/precipitation and the El Nino/La Nina events in the Tarim River basin












El Nino years









La Nina years









**Indicates that the χ2 interdependence test is significant at the significance level of 0.05; *indicates that it is significant at the significance level of 0.1

Prediction of Future Climate Scenarios

Liu and others (2007) used the Statistical Downscaling Model (SDSM) to establish the statistical relationship between atmospheric circulation factors from the National Centers for Environmental Prediction (NCEP) and measured climatic variables in the Tarim River basin. The HadCM3 outputs of A2 and B2 scenarios were then placed into the calibrated SDSM to simulate the maximum and minimum temperature and precipitation changes of the basin in 2020s, 2050s and 2080s. The primary conclusions were: (1) On the yearly scale, the future highest and lowest monthly temperature of the basin would both increase, and would be higher under A2 scenario than B2 scenario. Under both scenarios, the increasing rate of the highest temperature would be generally faster than that of the lowest temperature. However, the yearly precipitation of the basin under both scenarios would exhibit only a slight increasing trend, and the increasing rate would become smaller as years go by. (2) On the monthly scale, the highest and lowest temperatures in most months would have an increasing trend except the months of January, February, November and December in which the extreme temperatures sometimes would show a decreasing trend. Moreover, the increasing amplitude would be larger than the decreasing one. The largest variation of the highest temperature would appear in July and December, while that of the lowest temperature would appear in February and June. The variation ranges of seasonal and monthly precipitation would be rather wide and identical under both scenarios. The precipitation from May to August would show an increasing trend. The maximum value of monthly precipitation would appear in December and the minimum would appear in September.

Changes in Water Resources

Changes of Runoff Volume in Headwater Streams

The average annual natural runoff of the Tarim River basin is 3.98 × 1010 m3 and majority of the water supply comes from glacier melt. Since the unduplicated volume of the groundwater is 3.07 × 109 m3, the total volume of the basin is 4.29 × 1010 m3 (Wang and others 2006a). Over the past 50 years, the natural runoff in the three headwater streams of the Tarim River increased gradually with an average annual runoff volume (ARV) of 1.88 × 1010 m3, the maximum ARV of 2.51 × 1010 m3 (in 1994) and the minimum ARV of 1.38 × 1010 m3 (in 1965) (Fig. 3). The ratio of maximum to minimum annual runoff is 1.82 with Cv of 0.14., The annual runoff was close to the average in the years between 1950 and 1980. Higher runoff occurred in 1990s and between 2000 and 2005 with ARV of 1.99 × 1010 and 2.13 × 1010 m3 respectively. These runoff volumes were 1.1 × 109 and 2.5 × 109 m3 higher than the multi-year annual average and increased by 5.6 and 11.75% respectively.
Fig. 3

Variation of annual runoff in the three Tarim headstreams during 1957–2005

Trend of Runoff Volume in the Mainstream

Unfortunately, increased ARV of the headwater streams did not lead to the expected increase of the mainstream runoff volume. The measured annual runoff volume at the four hydrological stations (Alar, Xinqiman, Yingbaza and Kara) on the mainstream all showed a monotonic decreasing trend between 1957 and 2005 (Hao and others 2009a) (Fig. 4). Data recorded at the Alar, Xinqiman and Yingbaza hydrological stations were highly consistent and the three decreasing curves matched each other well. The annual runoff volume measured at the Kara hydrological station, however, did not match the curves analyzed at the other three stations. Runoff values at the Kara station were far lower than those at the Yingbaza hydrological station. This result suggested that the hydrological regime in the middle and lower reaches of the Tarim River was strongly influenced by external factors, such as irrational water diversion, insufficient water resources project control, lack of authoritative governing body, and lack of comprehensive management plan. Water supply from the three headwater streams had decreased from 5.09 × 109 m3 in the 1950s–1960s to 4.36 × 109 m3 between 2000 and 2005, and had lost 7.27 × 108 m3 over the past 50 years.
Fig. 4

Variation of annual runoff at the four hydrological stations in the mainstream of the Tarim River during 1957–2005

A breakpoint test was performed for the time series of runoff at the four mainstream hydrological stations using the Mann-Kendall method (Hao and others 2008b). Results revealed that the breaking points of ARV for each hydrological station along the mainstream from upper to lower reaches (Alar, Xinqiman, Yingbaza and Kara) occurred in 1970, 1972, 1974 and 1974 respectively, reflecting the upper to lower transfer processes of the interfering factors. Human activities, especially the large areas of land development and water diversion for irrigation, were the main influencing factors. The analytical results (Table 2) also showed that the area that consumed the majority of the water in the mainstream is between the Yingbaza and Kara hydrological station. In other words, the middle reaches of the Tarim River is the major water consumption section in the mainstream accounting for 46.58–62.67% of the total consumed water. The lowest water consumption section is between the Alar and Xinqiman hydrological station, which accounted for a maximum amount of 23.49%. Obviously, water usage decreased spatially from upper to lower reaches along the mainstream. However, there was no consistent temporal trend in these different sections. The section between Yingbaza and Kara station has showed a decreasing trend of water consumption in the past.
Table 2

Water consumption in the main stream of the Tarim River





Water consumption/108 m3

Percentage %

Water consumption/108 m3


Water consumption/108 m3

Percentage %




































Percentage: water consumption between the two stations divided by the total water consumption from Alar to Kara

Driving Factors for ARV Change in the Mainstream

As discussed above, ARVs of headwater streams and the mainstream of the Tarim River showed distinctively different trends in the past 50 years with increasing trend in headwater streams and decreasing trend in the mainstream. The decrease of mainstream ARV is mainly caused by intensive human activities (Chen and others 2003b; Hao and others 2008b; Wang and others 2006a; Zhou and others 2002), such as continued population growth, economic and social development, low efficiency of water utilization, heavy irrigation, and disturbance to the natural water cycles by various water conservancy facilities in both the headwater streams and the mainstream of the Tarim River.

Response of Glaciers to Climate Change

In the high mountains of the Tarim basin, there are 11,665 glacier rivers within China territory flowing into the basin water system, with a total cover area of 19,878 km2 and an ice reserve of 2,313 km3 (Liu and others 2006). 613 glacier rivers are located in Kyrgyzstan running into the basin with a total cover area of 2,120 km2 and an ice reserve of 260 km3. The estimated glacier melt accounts for over 40% of river runoff, while the major headstreams of the Tarim River (Yarkant River, Yulongkax River and Kunmalik River) may have 50–80% of the water coming from glacier melt.

In recent years, glaciers in the upstream of the Aksu River, a main tributary of the Tarim River, have been closely monitored by the Tianshan Glaciological Station of the Chinese Academy of Sciences. Based on the remote observation, the glacier-covering area in the Tomur region has decreased by 5–20% over the past 40 years, and the thickness of the glaciers have decreased even more dramatically in the same time period (Li and others 2010b). Temperature elevation plays a significant role in the melting of compound valley glaciers at low altitudes. It was estimated that tongues of compound valley glaciers account for over half of the total glacier volume in the region (Fig. 5). Because these types of glaciers are located at low altitudes, they are highly sensitive to climate change and are the main source of glacier melt runoff. Field observation and glacier dynamics simulation have suggested that tongues of such glaciers would melt away on the decade scale, which would subsequently cause a remarkable decrease in runoff volume.
Fig. 5

Glacier distribution in the upstream of Aksu River. The dotted areas are tongues of compound valley glaciers which melt more quickly (at several decades scale). The area of these parts is not large but the estimated volume is over the half of the entire glaciers

Glacier water plays a critical role in the upper reaches of the Tarim River. It is likely that in the next 50 years glacier melt will dramatically increase, which may even cause flooding (Li and others 2010b). However, as time goes on, the scale of glacier melt will be smaller and smaller. Once all the glaciers have melted away, the lack of water resources in this region will be disastrous.

Response of Accumulated Snow to Climate Change

Besides the change in climate, the seasonal snow coverage in the basin has also changed to some extent over the past few decades. Xu and others (2008) studied the snow cover distribution and variation in the Tarim River basin using the snow cover area (SCA) derived from NOAA/AVHRR land cover data and the GTOPO30 DEM during 1982–2001. Results showed that since the 1980s, the overall snow coverage area in the basin had slowly increased, and a relatively large fluctuation had been observed since the late 1980s. Compared to the escalating increase of temperature and precipitation, the increase in snow coverage was somewhat retarded. At the altitudes below 2,500 m, snow cover did not decrease in response to temperature increase, but showed a slow increasing trend. In the areas above 2,500 m, the snow coverage gradually decreased. Correlation analysis showed that the accumulated snow at various altitudes was affected by both temperature and precipitation. Precipitation plays a more significant role in the low altitude areas (below 2,500 m), while temperature is the main factor affecting snow cover in high altitude areas (above 5,000 m). Snow accumulation could be attributed to the spatial variation of precipitation and temperature. In the Tarim River basin, precipitation increases firstly and then decreases with the increasing elevation. There is a belt of maximum precipitation in areas ranging from about 3,000 to 4,500 m. Below the belt, precipitation increases and temperature decreases, and precipitation dominates the snow variation. Above the belt, precipitation decreases even to nothing and temperature decreases. The altitudinal belt from 2,500 to 5,000 m is affected by both temperature and precipitation. Annual distribution of the accumulated snow also showed a considerable change in the past years. Compared to 1980s, snowfalls in 1990s were more frequent and snow melting occurred much faster. An obvious positive correlation had been observed between snow cover and precipitation in cold months, while there was no apparent correlation between snow cover and temperature in the cold season.

Changes in Land Cover and Land Use

Current research on land utilization in the Tarim River basin has been mainly focused on the patterns of land use/ land cover, monitoring methods, monitoring scale and dynamic changes. In terms of land types, most attention has been paid to oasis, forest, grassland, farm land and decertified land.

Most oases are distributed in the upstream area of the Tarim River basin. Studies of a typical oasis, Aksu–Awati, by Gong and others (2005) revealed that farm land had increased by about 1.0 × 103 km2 from 1990 to 2000, while the area of wetland had decreased about 50% and sandy land had increased to about 2.5 × 102 km2 in the same time period. Economic development and population growth were the two main factors driving the change of land use and land cover. As a result, in just 22 years (1977–1999), the reclamation area in the lower reaches of the Tarim River had lost 60.6% of water areas and 60.5% of high coverage grasslands. The areas of forests, moderate coverage grasslands, and swamps also decreased significantly by 46.97, 29.49 and 20.30% respectively. On the contrary, saline land and sandy land had increased remarkably by 131.42 and 30.02% respectively. In addition, residential areas, farm lands and low coverage grasslands also showed a significant increase in the study area (Huang and others 2006). Population growth also caused a dramatic increase in food demand. To produce adequate food and fiber, cultivating land was expanded through deforestation. Meanwhile, along with the population growth, the economy and income level were also increased. To meet the demand of economic development and people’s living, urban expansion, road construction and building of other basic infrastructure had started to take place, which had caused a significant increase of construction lands in the river basin (Zhao and others 2009; Yang and others 2006).

Ecological Studies

Physiological and Biochemical Characteristics of Plants Under Drought Stress

Chen and others (2004c and 2004d) collected leaves of Populus euphratica at the nine (A–I) monitoring sections (Fig. 6) in the lower reaches of the Tarim River to investigate the physiological responses of P. euphratica under adverse circumstances. They found that the content of abscisic acid (ABA) in poplars (P. euphratica) increases when the depth to the groundwater table increases. Since the deeper the groundwater table the higher the level of water deprivation, accumulation of endogenous abscisic acid is an active and effective indicator for P. euphratica to cope with drought. Under drought stress, the growth rate of P. euphratica decreases which promotes the accumulation of endogenous assimilates, and thereby improves its water retaining capacity and enhances its adaptability to drought. It was noticed that the ABA content in P. euphratica in the lower reaches of the Tarim River changes with the depth to the groundwater table. This characteristic strongly suggested that accumulation of ABA in P. euphratica is closely related to its drought resistance ability. When the depth to the groundwater table increases, the water availability for plant growth becomes more limited. As a consequence, water content in leaves of P. euphratica tends to decrease, and the lipid peroxidation product, malondialdehyde (MDA), in leaf cell membranes gradually increases resulting in greater damages to P. euphratica.
Fig. 6

Location of the nine ecological monitoring sections (A–I) in the lower reaches of the Tarim River

Through comparison of osmolytes in different plants, it was found that the regulations of soluble sugars and proline (PRO) complemented each other in P. euphratica and Tamarix chinensis. The Pearson Correlation Analysis was conducted between plant physiological indices and the groundwater level. Results showed that indices including soluble sugar, PRO, MDA, and superoxide dismutase (SOD) are positively correlated with the groundwater level, while chlorophyll and peroxidase (POD) are negatively correlated with groundwater level. In addition, under drought stress, plants utilize various coordinated physiological activities to minimize damage (Chen and others 2004a).

Gas Exchange and Chlorophyll Fluorescence in P. euphratica Under Drought Stress

Chen and others (2011) compared the daily gas exchange, light response, PN–Ci curve and chlorophyll fluorescence of P. euphratica leaves in areas with different depths to the groundwater table (4.91, 6.93 and 8.44 m). Results showed that when the depth to the groundwater table increased from 4.91 to 6.93 m or 8.44 m, the photosynthesis rate of P. euphratica (PN) (10:00 am), initial chlorophyll fluorescence (F0), maximum chlorophyll fluorescence (Fm), actual PSII photochemical efficiency (ΦPSII), electron transport rate (ETR), non-photochemical quenching coefficient (NPQ) and mid-day leaf water potential (Ψmidday) all showed significant changes, indicating that the drought stress level was increased (Table 3). When the groundwater table was between 6.93 and 8.44 m, the above parameters did not vary significantly, suggesting that P. euphratica was likely experiencing the same level of drought stress. The maximum photochemical efficiency (Fv/Fm), apparent quantum efficiency (φ) and Rubisco carboxylation rate (Vcmax) did not show significant changes when the groundwater table was between 4.91 and 8.44 m, implying that Populus euphratica has a very high drought stress tolerance. In addition, the photosynthetic capacity of the plant was not irreversibly damaged in these areas.
Table 3

Chlorophyll fluorescence parameters of P. euphratica grown at different groundwater depths

Measurement time


Groundwater depth (m)




Predawn (6:00)


304.3 ± 3.2a

261.0 ± 8.4b

249.7 ± 3.1b


2078 ± 26a

1929 ± 33b

1885 ± 23.0b


0.85 ± 0.00

0.86 ± 0.00

0.87 ± 0.00


5.83 ± 0.08a

6.46 ± 0.24b

6.56 ± 0.08b

Morning (10:00–11:00)


0.444 ± 0.030a

0.3144 ± 0.021b

0.338 ± 0.015b


160.1 ± 7.87a

124.9 ± 6.67b

132.8 ± 6.00b


0.465 ± 0.036a

0.622 ± 0.025b

0.593 ± 0.018b


0.59 ± 0.035

0.499 ± 0.022

0.514 ± 0.022


1.173 ± 0.183a

2.661 ± 0.268b

2.455 ± 0.153b

Noon (12:00)


341.70 ± 4.73a

288.4 ± 7.1b

300.0 ± 6.5b


1880 ± 69a

1589 ± 48.0b

1635 ± 35b


0.81 ± 0.01

0.82 ± 0.00

0.81 ± 0.01


4.52 ± 0.30

4.51 ± 0.12

4.44 ± 0.21

Mean values ± SE are shown. Different letters in the same line indicate significant difference (P < 0.05)

Water Potential of the Plant and Its Impacting Factors

Fu and others (2006) analyzed the stem water potential of Tamarix Chinensis, leaf water potential in different shaped leaves of P. euphratica, and the relationship between environmental factors and water potential of these two plants. Results indicated that the daily change of Tamarix Chinensisstem water potential has a roughly linear decreasing trend, and it is strongly correlated with the temperature and relative air humidity. When temperature increased and relative air humidity became low, the transpiration water loss would increase, which subsequently caused the decrease of the plant water potential. Meanwhile, the stem water potential has a significant correlation with the groundwater level. As the depth to the groundwater table increased, the soil water content would decrease, thereby leading to the decrease of stem water potential. If the change of groundwater level was large, groundwater availability would become the key impact factor for Tamarix chinensisstem water potential. If the change of groundwater level was not significant, climate factors would act as the main water potential regulators.

The water potential of the two major heterophylls of P. euphratica (ovate and lanceolate leaves) has roughly the same trend in daily and monthly changes, which somewhat matches the temperature trend. When the groundwater level was low and the water content in soil was relatively high, the leaf water potential of P. euphratica did not decrease as temperature raised, but instead gradually increased, suggesting that temperature has a low impact on P. euphratica leaf water potential; When the groundwater level was relatively high and the soil water content was limited, the leaf water potential of P. euphratica gradually decreased as the temperature increased, indicating an increased impact of temperature on P. euphratica leaf water potential.

Change of Sap Flow in P. euphratica and Its Environmental Explanations

Sap flow volume of the P. euphratica trunk has a distinctive diurnal rhythm, being strong in the day time and weak at night. The sap discharge process follows a parabolic shape with an apparent rise, peak and fall (Zhou and others 2008; Ma and others 2010). The typical daily changes of sap flow in P. euphratica trunks are identical from month to month. Between late April and early May, the trunk sap flow starts to rise, but is still weak with no apparent diurnal rhythm. In late June, sap flow of P. euphratica reaches to a higher level. Between July and September, the sap flow becomes strong and the peak value appears in August. Starting from late October, sap flow of P. euphratica becomes weak, and a winter characteristic sap flow is observed in mid-November with weak flows both in the day time and at night. However, the sap flow does not stop in P. euphratica in winter (from November to the next April). The seasonal change of trunk sap flow in P. euphratica is closely correlated with its phenology (Zhou and others 2008).

Sap flow of the P. euphratica trunk is also closely associated with climate factors. The flow volume is influenced by the following factors in an order of: photosynthetically active radiation (PAR) > relative humidity (H) > temperature (Ta) > wind speed (S). Relative humidity has a higher impact than temperature on the sap flow volume indicating that transpiration of P. euphratica is very sensitive to humidity of the air. P. euphratica can adjust its water usage according to the air humidity, and therefore it possesses physiological and ecological adaptability to the extreme drought condition (Zhou and others 2008).

Water Distribution in Plant Roots and Its Ecological Effect

Hao and Chen (2010) measured the root sap flow rate of P. euphratica and the synchronous soil water content surrounding the root system during the growth season using a HRM (Heat Ratio Method) stem flow meter and a soil water neutron probe. Results showed that the sap flow rate of both taproots and lateral roots changed dramatically during a day (Fig. 7). The taproot sap flow rate has a remarkable single-peak pattern with the rate rising dramatically between 8:00 and 10:00 and dropping significantly between 20:00 and 23:00. The taproot sap flow rate was maintained at a low speed between 23:00 and 8:00 with the lowest rate at 0.9 cm/h. A much higher rate was recorded at daytime between 10:00 and 20:00, with a maximum value of 13.87 cm/h. The sap flow rate of lateral roots has a similar daily trend with that of the taproots. However, there is also a notable difference that the lateral root sap flow rate has negative values between 21:00 and 8:00. The lack of leaf surface water vapor pressure is the major factor causing the negative sap flow of lateral roots. When “water hoisting” occurs in P. euphratica, the soil water content significantly increases, particularly in the soil (60–120 cm depth) within a 4 m radius of the main trunk. It is estimated that the amount of water hoisted by the P. euphratica root system accounts for at least 10% of the water used for transpiration.
Fig. 7

Sap flow velocity (Vh) in the tap root and lateral root of P. euphratica Oliv. During September 10–22, 2008. a, b, and c show the diurnal and nocturnal variations of tap root of three trees, respectively, over 4 days; d, e, and f show the diurnal and nocturnal variations of lateral root of three trees, respectively, over 4 days. Points in d, e and f present the sap velocities of the lateral roots with different diameters

Species Diversity Associated with the Change of Groundwater Levels

The plant species in the middle and lower reaches of the Tarim River consist of Salicaceae, Tamaricaceae, Leguminosae, Apocynaceae and Gramineae. The representative plants are P. euphratica, Tamarix spp., Phragmites communis, Poacynum hendersonii and Alhagi sp. The species diversity shows a notable gradient change in regions with different groundwater levels, reflecting drought stress levels of the major desert vegetations. Research (Hao and others 2008a) showed that as the depth to the groundwater table increased, the species diversity indices, (Simpson, Shannon-Wiener, Margalef and Alatalo), increased first and then decreased, and the maximum index values appeared at a depth to the groundwater table between 2 and 4 m. When the depth to the groundwater table exceeded 6 m, species diversity in the lower reaches of the Tarim River started to decrease dramatically. Therefore 4 m is a critical groundwater level to determine whether the abundance of desert plants is affected, and 6 m is another critical groundwater level to prevent damage of the desert plants.

Determination of Suitable Ecological Groundwater Level

Plant physiology techniques have been used to examine the ecological water levels of representative species in the desert riverbank forest community in the lower reaches of the Tarim River. Through the analysis of physiological and ecological responses of individual plants with different groundwater levels, researchers aimed to identify the critical groundwater level that is suitable for plant growth. It was found that when the groundwater level is relatively low, changes in each physiological index of P. euphratica and T. chinensis are very small. When the groundwater level increases, the content of soluble sugars and the level of the endogenous plant hormone abscisic acid increased linearly. Meanwhile, it was also found that under a certain level of drought stress, T. chinensis has a much stronger physiological response than P. euphratica, suggesting that T. chinensis is more sensitive to groundwater level change while P. euphratica is more resistant to drought (Chen and others 2006b). All physiological and biochemical indices clearly showed that the suitable ecological water level for P. euphratica, in terms of depth to the groundwater table, is 2.0–4.0 m, the level that endangers normal plant growth is 4.0-9.0 m, and when the groundwater level is ≥9.0 m, the condition becomes critical for P. euphratica. For T. chinensis, when the groundwater level exceeds 3.12 m, growth inhibition starts to appear, and a groundwater level greater than 5 m will cause severe drought stress to T. chinensis. When the depth to the groundwater table is ≥8.83 m, the survival of T. chinensis is greatly endangered (Chen and others 2004b; Zhuang and others 2007).

Ecological Water Requirement

Ecological water requirement (EWR) is the amount of water resources required to maintain certain environmental functions or goals (current status, recovery or development) (Wang and others 2002; Wang and Chen 2000). EWR is a hot topic of current research in ecology, environment, and water resources management. It is also a critical question to be answered in the protection and reconstruction of ecologically fragile environments.

Studies on the ecological water requirements in the lower reaches of the Tarim River revealed that the natural vegetations in the area required 3.2 × 108 m3 of water in the 1980s (Ye and others 2007). However, the water requirement dropped sharply to 2.4 × 108 m3 in 2005 (Ye and others 2008). This is because both the area of vegetation and the level of groundwater dropped dramatically since the 1980s. In 2005, the ecological water requirement volume was 1.95 × 108 m3 in the critical growth period (April–September), accounting for 81% of the annual water requirement by the vegetations, and is 3.3 times the water requirement in non-growth season. The peak growth period of the vegetation is in May, June and July, during which transpiration water loss was tremendous and more water was needed accounting for 47% of the annual requirement. Vegetation transpiration rates were much lower in other months and the ecological water requirement was consequently less. Therefore, around April is the best time for ecological irrigation. Considering the physiological characteristics of the plants and the hydrological conditions, August and September are also optional times for irrigation in matching with the ripening time of plant seeds and offering a suitable condition for the next growth cycle (Chen and others 2003b).

Challenges and Problems

Scale and Complexity

Tarim River basin is a compound ecosystem in the arid region covering high mountain glaciers, grasslands, oases in the plains and Gobi deserts. The ecosystem elements are interdependent and mutually constrained. Water resources are a crucial link maintaining this compound ecosystem. Moreover, desertification and oasis development occur concurrently due to human activities. Therefore, the eco-hydrological process in the Tarim basin has its special complexity (Zhao and Chen 2001a).

In arid regions, strong spatial heterogeneities of weather, terrain conditions, geological structure, and plant water consumption make it a great challenge to upscale a local process to regional projections. Since research objectives would be different at different scales, sometimes the difficulty is not about the scale transformation technology but rather the quest for an essential meaning of the study. Observations can only reflect the pattern and process of that scale. Structural and functional heterogeneity or nonlinearity of different scales makes the scale transformation very complex and difficult (Xia and others 2003). In the lower reaches of the Tarim River, the purpose of ecological water requirement research based on plant photosynthesis and sap flow is to reconstruct the entire damaged ecosystem. However, most experimental observations were conducted on individual plants resulting in a failure of data integrity. Therefore, the scale issue must be taken into consideration for both experimental design and model construction in the eco-hydrological research of the river basin. The mismatch of scales in ecology and hydrology should be particularly avoided.

Data Problem

Many difficulties exist in the study of eco-hydrological processes including interdisciplinary, conceptual, and technical issues. Taking the ecological water requirement for example, the current studies are mainly concentrated on the conceptualization and definition, while the calculating methods focus more on water balance, heat balance, water and sediment balance, and water and salt balance. These research lacks consideration of the mutual feedback functions between ecological system and hydrological processes, the relationship between available water moisture and productivity of different species, and the systematic quantitative methods based on rigorous physiology, ecology and physics. In addition, data accumulation and consolidation were generally carried out independently on ecology, hydrology or soil content, and data fusion was almost nonexistent. Meanwhile, due to the lack of observational and experimental data, the ecological process model based upon observations at a limited number of locations compromised our understanding on the complex feedback process of interactions between ecology and hydrology.

Development of New Methodologies

The research of hydrological effects on the change of spatial ecological pattern and process in the basin has just started, and now is still at the stage of seeking statistical patterns and construction of mathematical models. The establishment of an eco-hydrological simulation model by coupling ecological and hydrological processes will be an incarnation of eco-hydrological application in water environment management. The “nested” technology on eco-hydrological models and the modern information technology should be put into full utilization, so as to enhance the practicality and flexibility of the model (Yan and others 2005). Field tests should not only cover the impact of ecological type and vegetation coverage on hydrological processes, but also consider the corresponding hydrological process mechanisms during the concurrent ecological pattern change. In addition, effects of hydrological events and processes on ecological pattern and vegetation types should be investigated, and the leading factors at different temporal and spatial scales in eco-hydrological processes should be identified, so as to create a more realistic model for the eco-hydrological mechanism (Xia and others 2003).

Integration of Different Processes

Zhao and Chen (2001a) studied the interactions between vegetation and hydrology in arid regions and described the ecological patterns, the hydrology mechanism during the ecological process, and the spatial/temporal coupling of hydrological and ecological processes (i.e., the eco-hydrology). However, currently there is still a lack of integration of multidisciplinary research to identify the role of water in ecosystems and environments of arid regions. Therefore, as an interdisciplinary branch of science, eco-hydrology will play an important role in filling in this gap. From the eco-hydrology perspective, two issues need to be investigated: (1) how does the hydrological process affect the distribution, structure, function and dynamics of ecosystems? and (2) how does the feedback from biological processes affect the hydrological cycle?

Planning and Prospects

The major tasks of eco-hydrological research in arid inland river basins, with the Tarim River basin as a typical representative, are as follows:

(1) To obtain key scientific and technologic supports needed for the study of ecological processes and recovery-reconstruction of damaged ecosystems at the middle and lower reaches of the Tarim River. Studies should be focused on the groundwater and riverbank forest vegetation systems in the arid desert regions, in order to examine the changing vegetation-soil-groundwater system and salinification process and the interaction between the desert riverbank forest vegetation community and hydrological processes. Important scientific issues, such as suitable groundwater level and ecological water requirement and maintaining the ecological security of the lower reach Tarim River, should be addressed. Integrated research and demonstration on key technologies to recover/ reconstruct deteriorated ecosystems should be carried out.

(2) To solve the stability problems of the oasis ecosystem and the oasis-desertification process in the Tarim River basin. Research should be focused on key technologies and important scientific questions at different scales, such as the ecological process in an oasis, oasis stability, and technologies related to the efficient use of oasis soil-water resources (internal material recycling, fertility cultivation, agricultural water conservation, joint use and inter-conversion of surface water and groundwater, and productivity improvement). Experiments and demonstration of high level water-saving irrigation models and joint use technology of surface water and groundwater and productivity improvement technologies should be carried out.

(3) To solve issues facing water resources allocation and management at the basin scale, Research should focus on the trends of water recycling and availability of water resources, as well as their relationships with climate change and anthropogenic formation-transformation of river runoffs. Scientific configuration of the water resources in the Tarim River basin between the socio-economic system and the ecological environment system should be discussed. An optimal space-time configuration and management strategy for use of the multi-source and multi-destination water resources in the basin should be proposed.


Eco-hydrological processes at a basin scale are very complex with spatial and temporal changes in water resources and water utilization which are related to climatology, hydrology, land use/land cover, plant physiology and ecology. These different research fields are interrelated when the basin is treated as a whole. This paper uses the concept of “eco-hydrological processes” to integrate the interdisciplinary study, which is different from the traditional and generic eco-hydrological research. When classified on the spatial scale, the traditional research mainly focuses on the micro-study of the relationship between plants and water moisture based on SPAC, the meso-scale study of plants and hydrology based on SVAT, and the meso-or-large scale study of surface coverage change and hydrological system (Mu and others 2011). Today, large-scale basins or regions often consist of agriculture-forest-grassland compound ecosystems instead of single vegetation forms. Furthermore, great differences exist among vegetation types, soils, climate, runoff yield processes, water balance, water quality, and water cycle models in different basins. It remains unclear how to apply the hydrological behaviors of vegetation in specific locations to the whole basin. Investigating the changes of large-scale hydrological system under different vegetation types has been a research focus in recent years.

The Tarim River basin is a typical representative of arid inland river basins, and has been facing tremendous eco-environmental challenges. Although the Chinese government has invested heavily to improve the eco-environment in the basin since 2000, little progress has been made. The fundamental reason is that there is no or weak integrated river basin management. The improvement of the basin environment will depend on the coordinated and stable development of all sub-systems. Any disruption in any sub-system will affect the whole basin environment. Today the importance of environmental management has been well accepted (Keene and Pullin 2011; Argent and others 1999; Margerum 1999; Staudenrausch and Flugel 2001; Wang and others 2006b; Burger 2008; Raymond and others 2010; Hipel and Walker 2011). Many large trans-boundary rivers in Europe, such as the Thames, the Rhine and the Danube, have been undergoing successful basin environmental management (Libor and others 2004; Peter and others 2006). Particularly the management of the Rhine and the early warning system of the Danube are successful examples. The Tarim River basin is located at the northwestern part of China. Water resources availability is the bottleneck of the socioeconomic development of the basin. If the relationship between water resources and all the sub-systems (economy, society, nature and ecology) are well defined, the eco-environmental problems can be controlled effectively. However, the strategies of basin water management are not implemented effectively because of the outdated technology and ecological concepts and the involvement of multiple stakeholders. This indicates the need for an integrated environmental management of the river basin, especially the water resources management.

The first priority is to establish and improve the water resources management and regulatory mechanism, such as establishing a reasonable water allocation program and transfer mechanism, and introducing the water market adjustment mechanism and strengthening the centralization of basin management. Meanwhile, strict water use permission and water quality supervision systems should be set up to ensure the quality and supply of water resources. The market adjustment may include ecological compensation, water expenditure compensation, water price adjustment, and incentive system. Specifically, the centralized water resources management by the Tarim River Basin Management Bureau (TRBMB) should be strengthened. At present, only three river systems, Kaxgar River, Weigan River and the mainstream of Tarim River are controlled by the TRBMB. Six other river systems are administered by the local governments. As a result, water resources cannot be allocated and used efficiently. It is advised that the boundaries of administrative areas should be broken, the regulatory agencies of headwaters should be merged to the TRBMB, and water allocation in irrigated areas should be administered by the regions, prefectures and divisions.

Surface water and groundwater exchange reciprocally and frequently at the mountain pediment alluvial-proluvial fan of the Tarim River Basin. About 90% of the groundwater at the plain is from the surface water. However, water utilization in the headwaters has mainly centered on the surface water, and the utilization of groundwater is relatively inadequate. Some irrigated areas are even undergoing soil salinization due to the rising groundwater level. Therefore, groundwater exploitation should be conducted at a reasonable level to increase the water resources. Meanwhile the water resources expenditure should be decreased by adjusting the industrial structures, adopting the modern irrigation technology, and reducing the cultivated land area. The appropriate scale of oasis should be based on the availability of water resources.

It is also important to strengthen the ecological and hydrological research to provide scientific and technological support for the protection and sustainable utilization of water resources, the protection and restoration of ecosystems, and the stable and sustainable development of the society and economy in the river basin.


This study was jointly supported by the National Basic Research Program of China (973 Program: 2010CB951003), the National Natural Science Foundation of China (Grant No. 91025025) and the West Light Foundation of the Chinese Academy of Sciences (XBBS200907). We thank Professor Zongxue Xu and Zhongqin Li for their assistance in the work. We also thank the editor and the anonymous reviewers for their valuable comments.

Copyright information

© Springer Science+Business Media, LLC 2012

Authors and Affiliations

  • Yaning Chen
    • 1
  • Changchun Xu
    • 2
  • Yapeng Chen
    • 1
  • Yongbo Liu
    • 3
  • Weihong Li
    • 1
  1. 1.State Key Laboratory of Desert and Oasis Ecology, Xinjiang Institute of Ecology and Geography Chinese Academy of SciencesUrumqiChina
  2. 2.Key Laboratory of Oasis Ecology, School of Resources and Environmental ScienceXinjiang UniversityUrumqiChina
  3. 3.Department of GeographyUniversity of GuelphGuelphCanada

Personalised recommendations