Preamble

Land use/land cover change responses to ecological water conveyance in the lower reaches of Tarim River, China

WANG Shanshan¹,²,³,⁴*, ZHOU Kefa¹,²,³, ZUO Qiting⁵, WANG Jinlin¹,²,³,⁴, WANG Wei¹,²,³

¹ State Key Laboratory of Desert and Oasis Ecology, Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi 830011, China
⁵ School of Water Conservancy & Environment, Zhengzhou University, Zhengzhou 450001, China

Abstract

The Tarim River is the longest inland river in China and plays a crucial role in protecting the oasis economy and environment of the Tarim Basin. However, excessive exploitation and over-utilization of natural resources, particularly water resources, have triggered a series of ecological and environmental problems, including reduced river flow, deteriorating water quality, dried-up downstream channels, vegetation degradation, and land desertification. This study investigated land use/land cover change (LUCC) responses to ecological water conveyance in the lower reaches of the Tarim River using ENVI (Environment for Visualizing Images) and GIS (Geographic Information System) software for the period 1990–2018. Multi-temporal remote sensing images and ecological water conveyance data from 1990 to 2018 were analyzed. The results indicate that LUCC covered an area of 2644.34 km² during this period, accounting for 15.79% of the total study area. From 1990 to 2018, wetland, farmland, forestland, and artificial surfaces increased by 533.42 km² (216.77%), 446.68 km² (123.66%), 284.55 km² (5.67%), and 57.51 km² (217.96%), respectively. In contrast, grassland and other land use/land cover types such as Gobi, bare soil, and deserts decreased by 103.34 km² (14.31%) and 1218.83 km² (11.75%), respectively. Vegetation area decreased initially then increased, following the order 2010 < 2000 < 1990 < 2018. LUCC in the overflow and stagnant areas of the lower reaches was characterized by fragmentation, irregularity, and complexity. By analyzing LUCC responses to 19 rounds of ecological water conveyance from 2000 to the end of 2018, we propose guidelines for rational development and utilization of water and soil resources and formulate strategies for sustainable development in the lower reaches. This study provides scientific guidance for optimal water resource scheduling in the region.

Keywords: land use/land cover change (LUCC); remote sensing; land use dynamic index; ecological water conveyance; Tarim River

*Corresponding author: WANG Shanshan (E-mail: wangshanshan@ms.xjb.ac.cn)
Received 2020-10-31; revised 2021-08-18; accepted 2021-08-25

1. Introduction

The Tarim River, located in Xinjiang Uygur Autonomous Region, is China's longest inland river and exhibits the dual characteristics of rich natural resources and fragile ecosystems. The region is strategically important for the "Silk Road Economic Belt" initiative and contributes to China's economic and social development (Chen et al., 2020; Zuo et al., 2021). The basin is characterized by an extremely arid environment where most natural processes and human activities are directly or indirectly affected by water shortages (Deng et al., 2017; Chen et al., 2020). The lower reaches of the Tarim River have become a focal point for ecological and environmental research in Northwest China, with issues of ecological restoration and security attracting significant attention from both society and government (Yu et al., 2016; Keram et al., 2019; Zhou et al., 2020).

To restore and protect the downstream ecology, the Chinese government implemented a comprehensive management system for the Tarim River Basin in 2000 (Chen et al., 2004; Ablekim et al., 2016). The development of an intermittent ecological water conveyance project has altered the basin's ecology and environment, resulting in land use/land cover change (LUCC) (Bao et al., 2017; Wang et al., 2021).

Numerous studies have investigated vegetation, biomass, groundwater, LUCC, and ecological benefit assessment responses to ecological water conveyance in the lower reaches (e.g., Shi et al., 2010; Liu et al., 2014; Bao et al., 2017; Chen et al., 2020). Deng et al. (2016) analyzed and evaluated ecological response benefits of vegetation physiology and restoration after water conveyance by monitoring surface water, groundwater, soil water, and vegetation samples through local investigations. Chen et al. (2004) examined the ecological effects of water conveyance by studying vegetation along ecological sections. Zhu et al. (2019) monitored changes in fractional vegetation coverage (FVC) from 2000 to 2017 using time-series NDVI data from MOD13Q1 and explored vegetation growth and recovery under ecological water conveyance. Liu et al. (2014) studied vegetation restoration from 1999–2010 using Landsat TM/ETM and CBERS/CCD remote sensing images based on soil-adjusted vegetation indices.

LUCC is an important indicator for measuring regional ecological sustainability, and studying LUCC is essential for understanding environmental changes and promoting sustainable development (Hao and Ren, 2009; Schirpke et al., 2012). However, few studies have utilized multi-source remote sensing and long-term time series data to monitor and analyze regional environmental changes and ecological responses.

The ecological water conveyance project in the lower reaches has lasted nearly 20 years. This study monitored and analyzed LUCC and ecological responses based on multi-temporal remote sensing images obtained between 1990 and 2018 (a 29-year period). Specifically, we used ENVI, GIS, and statistical analysis software to examine the spatial distribution and temporal dynamics of LUCC, as well as land use change trends and spatial characteristics under intermittent water conveyance conditions.

2. Study Area

The lower reaches of the Tarim River, located in southern Xinjiang, refer to the section from Daxihaizi Reservoir to Taitema Lake, spanning approximately 428 km (Sun et al., 2011). Riparian forests form a "green corridor" in this arid region, blocking the connection between the Taklamakan and Kuruktag Deserts (Zhou et al., 2020; Zhang et al., 2021). Influenced by the climate of these two deserts, the lower reaches experience drought with minimal rainfall and snowfall, strong evaporation, and high wind frequency in spring and autumn. The region has a typical northern temperate continental arid desert climate, with annual average temperatures of 10.7–11.5°C, precipitation of 17.4–42.0 mm, and evaporation reaching 2500–3000 mm (Xu et al., 2019; Li et al., 2021). The ecological environment is extremely sensitive and fragile.

Since 2001, the local government has launched the ecological water conveyance project for the Tarim River Basin, implementing unified water scheduling and emergency ecological water transport for downstream rescue, restoration, and protection (Ablekim et al., 2016; Chen et al., 2019). From 2000 to the end of 2018, 19 ecological water transfers (27 phases) were conducted, delivering a total volume of 707.0×10⁸ m³ (Fig. 1), with water reaching Taitema Lake more than 16 times.

The study area is geographically located at 39°00ʹ–41°30ʹN and 86°00ʹ–89°30ʹE (Fig. 2). Taitema Lake is the lowest point in the region at 801.50 m. To investigate all areas affected by ecological water conveyance, we established buffer zones of 30–50 km on both sides of the main conveyance channel, focusing on the region between these buffer zones.

3. Methods

3.1 Data Acquisition

This study used Landsat TM, ETM, and OLI images from 1990, 2000, 2010, and the summer and autumn of 2018. These included multi-scene remote sensing data with track numbers 143/31, 142/31, 142/32, 142/33, 141/31, 141/32, 141/33, 141/34, and 140/33. We geometrically corrected the 2000, 2010, and 2018 images using an image-to-image registration method based on the 1990 images. After correction, the image data were mosaicked from nine sets of remote sensing images. We selected the scene with the largest area and best data quality as the reference image and adjusted other images accordingly. The FLAASH module in ENVI 5.3 software was used for atmospheric correction to eliminate atmospheric effects.

3.2 Random Forest Algorithm

The random forest (RF) algorithm is an ensemble learning method for data classification and regression (Breiman, 2001). Compared with other methods, RF performs well with smaller training sample sets, requires shorter computation time, and provides more accurate results through out-of-bag error estimation (Demarchi et al., 2014). RF has proven to be a highly robust discrimination method when remotely sensed spectra represent combinations of various materials. The algorithm was implemented using the image RF toolbox of EnMAP-Box v1.3. Figure 3 shows the methodological flowchart from imaging to results, including pre-processing.

Based on land use/land cover characteristics and planning in the lower reaches (He et al., 2010), we classified land use/land cover units according to Level-1 of the China land resource classification system, including meadowland, forestland, wetland, farmland, artificial surfaces, and other types such as Gobi, bare soil, and deserts (Fig. 4). The overall classification accuracies were 82.27%, 85.92%, 83.21%, and 87.21% for 1990, 2000, 2010, and 2018, respectively, meeting the requirements for land use/land cover monitoring.

3.3.1 Single Land Use Dynamic Index (P₁)

The single land use dynamic index (P₁; %) indicates quantitative changes in a particular land use/land cover pattern for a given region over a specific time period (Wang et al., 2017).

$$
P_1 = \frac{U_b - U_a}{U_a} \times \frac{1}{T} \times 100\%
$$

where $U_a$ and $U_b$ represent the quantities of a particular land use/land cover pattern in the early and late stages of the research period (km²), respectively, and $T$ denotes the timespan (years).

3.3.2 Spatial Variability of a Single Land Use Pattern (P₂)

The spatial variability index (P₂) was introduced because P₁ only reflects quantitative changes but not the degree of spatial change. The P₂ (%) value reveals the dynamic degree of spatial change in specific land use/land cover patterns (Wang et al., 2017).

$$
P_2 = \frac{\Delta U_{in} + \Delta U_{out}}{2U_a} \times \frac{1}{T} \times 100\%
$$

where $\Delta U_{in}$ and $\Delta U_{out}$ represent the areas transferred into and out of a land use/land cover type (km²), respectively, during the study period.

4. Results

4.1 Spatial Distribution of LUCC

The distribution of land use/land cover patterns in 1990, 2000, 2010, and 2018 (Fig. 4) reveals LUCC characteristics over the 29-year study period. To explore these changes more accurately, we calculated a transition matrix for 1990 and 2018 (Table 1). LUCC varied significantly, covering 2644.34 km² or 15.79% of the total area (16,761.03 km²).

Forestland covered the largest area in 2018 (5303.06 km²), representing an increase of 284.55 km² (5.67%) compared to 1990. Meadowland area was largest in 1990 (722.14 km²) and smallest in 2010 (493.09 km²), decreasing by 103.34 km² (14.31%) between 1990 and 2018. Farmland area increased continuously, covering 361.20, 517.11, 690.67, and 807.88 km² in 1990, 2000, 2010, and 2018, respectively, for a total increase of 446.68 km² (123.66%). Wetland area showed the greatest increase of 533.42 km² (216.77%), more than doubling in size. Artificial surfaces increased from 26.39 km² in 1990 to 83.90 km² in 2018, a growth rate of 217.96% (57.51 km²). In contrast, other land use/land cover types (Gobi, bare soil, deserts) decreased by 1218.83 km² or approximately 11.75% in 2018 compared to 1990.

To better analyze the area, location, and trend changes of LUCC, we conducted a superposition analysis of the 1990 and 2018 classification maps (Fig. 5). Artificial surfaces such as mining fields and transportation lands expanded rapidly by 217.96% from 1990 to 2018, primarily converted from meadowland (13.77%), forestland (8.54%), farmland (12.02%), and wetland (0.10%). Wetland area increased by 533.42 km² (216.77%), mainly converted from other land types (42.50%), forestland (31.27%), and meadowland (8.20%). Farmland increased by 446.68 km² (123.66%), primarily replacing forestland (35.15%) and meadowland (19.13%). Meadowland was mainly replaced by farmland (19.13%) and artificial surfaces (13.77%), while Gobi, bare soil, and deserts were converted primarily to artificial surfaces (43.80%) and farmland (42.50%). Although artificial surfaces did not occupy a large proportion of the study area by 2018, their growth rate was substantial, increasing from 26.39 km² to 83.90 km² (217.92%).

4.2 Land Use Dynamic Indices

Table 2 presents P₁ and P₂ values for the periods 1990–2000, 2000–2010, and 2010–2018.

The P₁ values for meadowland and forestland were negative during the first two decades (1990–2000 and 2000–2010), indicating area decreases. Meadowland declined sharply, reaching a minimum of –2.63% during 2000–2010. During 2010–2018, P₁ values for meadowland and forestland were 2.66% and 0.98%, respectively, indicating recovery. The total increase in P₁ from 2000 to 2018 was 5.29% for meadowland and 1.11% for forestland, demonstrating rapid recent growth. Farmland area increased across all three decades, as shown by positive P₁ values. Wetland area decreased from 1990 to 2000 (P₁ = –1.06%) but increased substantially during the latter two decades with P₁ values of 6.71% and 11.36%, coinciding with increased water supply facilities. Artificial surfaces showed positive P₁ values during 2000–2010 (12.54%) and 2010–2018 (0.84%), with more significant changes during 2000–2010 that decreased after 2010. Other land types such as Gobi, bare soil, and deserts exhibited successive decreases in P₁ values (–0.12% and –0.98% for 2000–2010 and 2010–2018), indicating steady decline.

The P₂ values reveal that meadowland and forestland areas generally increased from 1990 to 2018, though trends varied significantly. P₂ values for meadowland and forestland during 2010–2018 were 9.30% and 4.35%, respectively, indicating significant spatial dynamics and extensive land conversion in late 2009. Farmland P₂ values showed minimal changes across the three periods (4.31%, 3.79%, and 4.73%), implying steady conversion. Wetland P₂ values demonstrated an increasing trend (3.11%, 7.42%, and 17.31%), with the largest value in 2010–2018. Artificial surfaces had higher P₂ values during 2000–2010 and lower values during 2010–2018, suggesting reduced spatial disturbance impacts.

5. Discussion

5.1 Trends of LUCC

Effective utilization of remote sensing data to analyze LUCC requires consideration of three aspects: location, direction, and process (Hasselmann et al., 2010; Jiang et al., 2015). Understanding change locations is vital for ecosystem monitoring and helps analyze and explain causes (Huang et al., 2006; Chen et al., 2008). Analyzing LUCC in response to ecological changes and human intervention is useful for understanding trends (Li et al., 2017; El-Tantawi et al., 2019). This study analyzed LUCC using satellite images from 1990, 2000, 2010, and 2018, with superposition of classification maps facilitating understanding of changes, locations, and trends.

The results indicate substantial farmland development near Qiala Reservoir during the study period. Most forestland extending outward from both river sides toward the desert and Gobi has become wetland, suggesting expanded water-receiving areas for vegetation (Ye et al., 2010; Keyimu et al., 2017). By 2018, most meadowland around Daxihaizi Reservoir had converted to forestland and partially to wetland. The small river course downstream of Daxihaizi Reservoir has widened significantly, forming a larger channel with numerous puddles and pits. Farmland area around Tikanlik Town (40°37ʹ47.53ʺN, 87°41ʹ23.91″E) in Ruoqiang County increased substantially, forming contiguous tracts. Notably, Taitema Lake had no water in 1990 and 2000, but water volume increased from 2010 to 2018, making it the second largest lake in southern Xinjiang. The surrounding environment improved simultaneously, with forestland area on both river sides increasing by 284.55 km² (5.67%), primarily in river overflow and stagnant pool areas. This suggests that river overflow promoted local vegetation recovery and the formation of fragmented, irregular forestland.

5.2 LUCC Responses to Ecological Water Conveyance

To understand LUCC responses, we analyzed wetland and vegetation areas (including forestland and meadowland) from 1990 to 2018. Wetland area changed in the order: 2000 < 1990 < 2010 < 2018 (Fig. 6), with similar areas in 1990 and 2000, increasing from 368.08 km² in 2010 to a maximum of 786.10 km² in 2018. The 11th water conveyance project in 2010 (June–November) lasted 139 days with a total volume of 3.89×10⁸ m³ (Fig. 1). In 2018, two conveyance periods occurred in February and August, totaling 88 days and 4.29×10⁸ m³—an increase of 0.39×10⁸ m³ compared to 2010. This demonstrates substantial wetland improvement.

Vegetation area followed the trend 2010 < 2000 < 1990 < 2018 (Fig. 6), decreasing by 138.20 km² between 1990 and 2000 and by 241.81 km² between 2000 and 2010, but recovering with a 609.21 km² increase between 2010 and 2018. This indicates a degradation-then-restoration process.

Many studies suggest that water conveyance benefits require several years to become visible and cannot be observed immediately (Wan, 2012; Guo et al., 2017; Wang and Guo, 2018). Similar to gradual pre-conveyance vegetation degradation, ecological responses of downstream vegetation and groundwater are slow on spatio-temporal scales (Chen et al., 2015a, b; Keyimu et al., 2017). We analyzed relationships between vegetation area in 2010 and 2018 and previous years' water conveyance (2009 and 2017). In December 2009, cumulative conveyance lasted 26 days with 0.11×10⁸ m³, reaching Kaerdayi. In 2017, conveyance occurred in April, May, and December, totaling 215 days and 8.33×10⁸ m³, affecting 2018 vegetation growth. Additionally, early 2018 conveyance in February addressed spring ecological water supply. Annual vegetation area changes were positively correlated with cumulative ecological water conveyance from previous years, with one-, two-, three-, four-, and five-year cumulative volumes responding to LUCC. Therefore, short-term and long-term effects of cumulative ecological water conveyance effectively promote natural vegetation restoration along both river sides (Ye et al., 2009; Zhang et al., 2013).

Variations in vegetation and wetland areas are influenced by multiple factors including climate, water volume, conveyance days, timing, average conveyance rate, and human factors (Lu and Jiang, 2009; Bai et al., 2015; Xu et al., 2015; Ling et al., 2019; Wang et al., 2020). This study indicates that river runoff in both vertical and horizontal directions is the main groundwater recharge source in the lower reaches, with groundwater level fluctuation amplitude gradually decreasing with distance both vertically and horizontally (Wu and Cai, 2004; Yu et al., 2012; Li et al., 2017). Remote sensing interpretation and field monitoring reveal that shallow groundwater levels increased due to ecological water conveyance. Monitoring data show groundwater depth increased from 9.8–10.1 m to 2.1–5.3 m at 1 km from the main channel, while mineralization increased from 1.1–3.0 g/L to 5.3–7.8 g/L during 2000–2017 (Ablekim et al., 2016; Chen et al., 2019).

Previous studies showed vegetation in the lower reaches has been restored, with area increasing to 2285.00 km² (a 362.00 km² increase). Sandy area decreased by 854.00 km², and plant species increased from 17 to 46, including hogweed, white thorn, camel thorn, reed, sand jujube, and licorice as main types (Fan et al., 2013, 2014; Ye et al., 2014a, b; Chen et al., 2015a, b, 2020). Drought-tolerant trees and shrubs gradually recovered, and floral biodiversity was restored. After years of continuous conveyance, flooded poplar forestland reached 712.00 km², restoring extensive poplar forests that had died from water scarcity. Artificial surfaces showed an increasing trend while Gobi, bare soil, and desert areas decreased significantly, indicating increasing human activity impacts since 2000 (Hartmann et al., 2016; Bao et al., 2017; Huang et al., 2018). However, increased vegetation area and recovery were still concentrated in certain areas, requiring further efforts to ensure long-term, stable, and scientifically sound ecological water conveyance and rational water resource utilization.

This study explored ecological water conveyance effects from the LUCC perspective, though the effects are not limited to LUCC. LUCC is a complex process requiring consideration of human activities and climate change (Chen et al., 2020; Li et al., 2021). This study focused on ecological water conveyance impacts, but climate factors such as temperature and precipitation, plus natural factors like topography and soil organic matter content, may also affect LUCC and warrant future investigation.

6. Conclusions

Based on multi-scene and multi-temporal remote sensing dynamic monitoring data, we investigated LUCC responses to ecological water conveyance in the lower Tarim River reaches during 1990–2018. The LUCC area was 2644.34 km², accounting for 15.79% of the total area. During the study period, wetland, farmland, forestland, and artificial surfaces increased, while grassland and other types including Gobi, bare soil, and deserts decreased. Improved vegetation was concentrated along both river sides in overflow and stagnant areas, indicating that river overflow promoted local vegetation restoration, particularly meadow-to-wetland and meadow-to-farmland conversions. Long-term, intermittent, and stable artificial ecological water delivery is a critical driver of ecological and environmental changes in the lower reaches.

Land use/land cover affects the regional natural environment, including groundwater, microclimates, natural vegetation, and wildlife, significantly impacting human lifestyles and living standards. Using multi-scene and multi-temporal remote sensing monitoring, this study provides guidance for rational water and soil resource development and utilization, and proposes Tarim River development strategies based on ecological security, serving as a scientific guideline for optimal water resource regulation and ecological water allocation.

Acknowledgements: This study was supported by the Key Project of National Natural Science Foundation of China-Xinjiang Joint Fund (U1803241), the Key Project of Xinjiang Uygur Autonomous Region Talent Special Plan-Tianshan Outstanding Youth (2019Q033), the Chinese Academy of Sciences (2017-XBQNXZ-B-019), and the Science and Technology Plan Major Projects of the Xinjiang Uygur Autonomous Region, China (2021A03001-3).

References

Ablekim A, Kasimu A, Kurban A, et al. 2016. Evolution of small lakes in lower reaches of Tarim River based on multi-source spatial data. Geographical Research, 35(11): 2071–2090. (in Chinese)
Bai Y, Xu H L, Zhang Q Q, et al. 2015. Evaluation on ecological water requirement in the lower reaches of Tarim River based on groundwater restoration. Acta Ecologica Sinica, 35(3): 630–640. (in Chinese)
Bao A, Huang Y, Ma Y, et al. 2017. Assessing the effect of EWDP on vegetation restoration by remote sensing in the lower reaches of Tarim River. Ecological Indicators, 74: 261–275.
Breiman L. 2001. Random forests. Machine Learning, 45: 5–32.
Chen Y N, Zhang X L, Zhu X M, et al. 2004. Ecological effect analysis of water conveyance in the downstream of Tarim River in Xinjiang. Science in China Series D: Earth Sciences, 34(5): 475–482. (in Chinese)
Chen Y N, Pang Z H, Chen Y P, et al. 2008. Response of riparian vegetation to water-table changes in the lower reaches of Tarim River, Xinjiang Uygur Autonomous Region, China. Hydrogeology Journal, 16: 1371–1379.
Chen Y N, Li Z, Fan Y T, et al. 2015a. Progress and prospects of climate change impacts on hydrology in the arid region of northwest China. Environmental Research, 139: 11–19.
Chen Y N, Li W H, Xu C C, et al. 2015b. Desert riparian vegetation and groundwater in the lower reaches of the Tarim River basin. Environmental Earth Sciences, 73: 547–558.
Chen Y N, Hao X M, Chen Y P, et al. 2019. Research on water system connectivity and ecological protection countermeasures in Tarim River Basin, Xinjiang. Proceedings of the Chinese Academy of Sciences, 34(10): 1156–1164. (in Chinese)
Chen Y N, Zhang X L, Fang G H, et al. 2020. Potential risks and challenges of climate change in the arid region of northwestern China. Regional Sustainability, 1(1): 20–30.
Demarchi L, Canters F, Claude C, et al. 2014. Assessing the performance of two unsupervised dimensionality reduction techniques on hyperspectral APEX data for high resolution urban land-cover mapping. ISPRS Journal of Photogrammetry and Remote Sensing, 87: 166–179.
Deng M J, Zhou H Y, Xu H Q, et al. 2016. Research on the ecological operation in the lower reaches of Tarim River based on water conveyance. Scientia Sinica Technologica, 46(8): 864–876. (in Chinese)
Deng M J, Yang P N, Zhou H Y, et al. 2017. Water conversion and strategy of ecological water conveyance in the lower reaches of the Tarim River. Arid Zone Research, 34(4): 717–726. (in Chinese)
El-Tantawi A M, Bao A M, Chang C, et al. 2019. Monitoring and predicting land use/cover changes in the Aksu-Tarim River Basin, Xinjiang, China (1990–2030). Environmental Monitoring and Assessment, 191: 480–498.
Fan Y, Chen Y, Liu Y, et al. 2013. Variation of base flows in the headstreams of the Tarim River Basin during 1960–2007. Journal of Hydrology, 487: 98–108.
Fan Y, Chen Y, Li W. 2014. Increasing precipitation and base flow in Aksu River since the 1950s. Quaternary International, 336: 186–194.
Guo H, Jiapaer G, Bao A, et al. 2017. Effects of the Tarim River's middle stream water transport dike on the fractional cover of desert riparian vegetation. Ecological Engineering, 99: 333–342.
Hao H M, Ren Z Y. 2009. Land use/land cover change (LUCC) and eco-environment response to LUCC in Farming-Pastoral Zone, China. Agricultural Sciences in China, 8(1): 91–97.
Hartmann H, Snow J A, Su B, et al. 2016. Seasonal predictions of precipitation in the Aksu-Tarim River basin for improved water resources management. Global and Planetary Change, 147: 86–96.
Hasselmann F, Csaplovics E, Falconer I, et al. 2010. Technological driving forces of LUCC: Conceptualization, quantification, and the example of urban power distribution networks. Land Use Policy, 27(2): 628–637.
He L H, Bao Q S, Sui M M. 2010. Application of land use status classification in land use status survey. Anhui Agricultural Science, (7): 440–442. (in Chinese)
Huang Q, Wang R H, Wu S X. 2006. Dynamic change of landscape pattern and hydrological & geochemical characteristic analysis before and after the emergency water transportation in the lower reaches of the Tarim River, China. Chinese Journal of Geochemistry, 25(1): 274–275.
Huang S, Wortmann M, Duethmann D, et al. 2018. Adaptation strategies of agriculture and water management to climate change in the Upper Tarim River basin, NW China. Agricultural Water Management, 203: 207–224.
Jiang P, Cheng L, Li M, et al. 2015. Impacts of LUCC on soil properties in the riparian zones of desert oasis with remote sensing data: A case study of the middle Heihe River basin, China. Science of the Total Environment, 506–507: 259–271.
Keram A, Halik Ü, Keyimu M, et al. 2019. Gap dynamics of natural Populus euphratica floodplain forests affected by hydrological alteration along the Tarim River: Implications for restoration of the riparian forests. Forest Ecology and Management, 438: 103–113.
Keyimu M, Halik Ü, Kurban A. 2017. Estimation of water consumption of riparian forest in the lower reaches of Tarim River, northwest China. Environmental Earth Sciences, 76: 547–558.
Li Q Y, Sun Y W, Yuan W L, et al. 2017. Streamflow responses to climate change and LUCC in a semi-arid watershed of Chinese Loess Plateau. Journal of Arid Land, 9(4): 609–621.
Li W W, Huang F, Shi F Z, et al. 2021. Human and climatic drivers of land and water use from 1997 to 2019 in Tarim River basin, China. International Soil and Water Conservation Research, 9(4): 532–543.
Ling H, Guo B, Zhang G, et al. 2019. Evaluation of the ecological protective effect of the "large basin" comprehensive management system in the Tarim River basin, China. Science of the Total Environment, 650: 1696–1706.
Liu G, Kurban A, Duan H, et al. 2014. Desert riparian forest colonization in the lower reaches of Tarim River based on remote sensing analysis. Environmental Earth Sciences, 71: 4579–4589.
Lu X, Jiang T. 2009. Larger Asian rivers: Climate change, river flow and sediment flux. Quaternary International, 208(1–2): 1–2.
Schirpke U, Leitinger G, Tappeiner U, et al. 2012. SPA-LUCC: Developing land-use/cover scenarios in mountain landscapes. Ecological Informatics, 12: 68–76.
Shi Y, Wang R, Fan L, et al. 2010. Analysis on land-use change and its demographic factors in the original-stream watershed of Tarim River based on GIS and statistic. Procedia Environmental Sciences, 2: 175–184.
Sun Z, Chang N B, Opp C, et al. 2011. Evaluation of ecological restoration through vegetation patterns in the lower Tarim River, China with MODIS NDVI data. Ecological Informatics, 6(2): 156–163.
Wan H M. 2012. Remote sensing measure of vegetation restoration in the lower reaches of Tarim River. MSc Thesis. Urumqi: Xinjiang Agricultural University. (in Chinese)
Wang F, Yang S, Wei Y, et al. 2021. Characterizing soil salinity at multiple depth using electromagnetic induction and remote sensing data with random forests: A case study in Tarim River Basin of southern Xinjiang, China. Science of the Total Environment, 754: 142–153.
Wang X X, Chen Y N, Li Z, et al. 2020. The impact of climate change and human activities on the Aral Sea Basin over the past 50 years. Atmospheric Research, 245: 105–125.
Wang Y, Song G E, Wang P. 2017. Study on the change characteristics and spatial-temporal pattern of land use types in Songnen Gao Pingyuan. Resources and environment in arid areas, 31: 61–66. (in Chinese)
Wang Y G, Guo Y H. 2018. Characteristics of oasis in the lower reaches of Tarim River before and after ecological water conveyance. Desert and Oasis Weather, 12(2): 37–42. (in Chinese)
Wu X Q, Cai Y L. 2004. Land cover changes and landscape dynamics assessment in lower reaches of Tarim River in China. Chinese Geographical Science, 14(1): 28–33.
Xu H L, Fan Z L, Yang P N, et al. 2015. Tarim River recent management evaluation and comprehensive watershed planning recommendations. Arid Land Geography, 38(4): 645–651. (in Chinese)
Xu M, Wang X, Sun T, et al. 2019. Water balance change and its implications to vegetation in the Tarim River Basin, Central Asia. Quaternary International, 523: 25–36.
Ye Z X, Chen Y N, Li W H, et al. 2009. Effect of the ecological water conveyance project on environment in the lower Tarim River, Xinjiang, China. Environmental Monitoring and Assessment, 149: 9–17.
Ye Z X, Chen Y N, Li W H. 2010. Ecological water demand of natural vegetation in the lower Tarim River. Journal of Geographical Sciences, 20(2): 261–272.
Ye Z X, Chen Y N, Li W H. 2014a. Ecological water rights and water-resource exploitation in the three headwaters of the Tarim River. Quaternary International, 336: 20–25.
Ye Z X, Chen Y N, Zhang X L. 2014b. Dynamics of runoff, river sediments and climate change in the upper reaches of the Tarim River, China. Quaternary International, 336: 13–19.
Yu G A, Disse M, Huang H Q, et al. 2016. River network evolution and fluvial process responses to human activity in a hyper-arid environment – Case of the Tarim River in Northwest China. CATENA, 147: 96–109.
Yu P J, Xu H L, Ye M, et al. 2012. Effects of ecological water conveyance on the ring increments of Populus euphratica in the lower reaches of Tarim River. Journal of Forest Research, 17(5): 413–420.
Zhang X Q, Chen Y N, Li W H, et al. 2013. Restoration of the lower reaches of the Tarim River in China. Regional Environmental Change, 13: 1021–1029.
Zhang X Q, Xia Q, Chen Y N, et al. 2021. Effects of ecological water conveyance on gross primary productivity of vegetation in Tarim River in recent 20 years. Arid Land Geography, 44(3): 718–728. (in Chinese)
Zhou H H, Chen Y N, Zhu C G, et al. 2020. Climate change may accelerate the decline of desert riparian forest in the lower Tarim River, Northwestern China: Evidence from tree-rings of Populus euphratica. Ecological Indicators, 111: 105–197.
Zhu C M, Li J L, Shen Z F, et al. 2019. Spatiotemporal dynamics of vegetation activities in the lower reach of the Tarim River based on MODIS intensive time series data. Resources Science, 41(3): 591–600. (in Chinese)
Zuo J P, Xu J H, Chen Y N, et al. 2021. Downscaling simulation of groundwater storage in the Tarim River basin in northwest China based on GRACE data. Physics and Chemistry of the Earth, 123: 103–142.