Change and Driving Factor Analysis of Eco-Environment of Typical Lakes in Arid Areas

Abstract

In arid regions with scarce water resources, lakes play an extremely vital role in maintaining the ecological environment. Therefore, the Chinese government has launched an ecological water conveyance project in the Tarim River basin in Xinjiang with the aim of restoring the ecological environment of the area. In previous studies, there was no complete evaluation system used to quantify changes in the ecological environment of arid regions after ecological water conveyance. In this paper, Lake Taitema was selected as the study area, which is both a terminal lake in the Tarim River basin and an object of the ecological water conveyance project. This study utilized Landsat TM/OLI satellite remote sensing images and MODIS datasets to build a remote sensing ecological index model and systematically evaluated the changes in the ecological environment and land use types in the Taitema Lake area. A structural equation model was constructed to analyze the correlation between the area of Taitema Lake and its driving factors. The results show that over the selected 20 years, the proportion of pixels with an upward trend (Zc > 0) of the RSEI was 56.5%, while the proportion of pixels with a downward trend (Zc < 0) of the RSEI was 43.5%. The area proportion of regions with poor ecological environment quality decreased by about 40%, and the area proportions of regions with moderate, good, and excellent ecological environment quality increased by 29.7%, 10%, and 0.6%, respectively. By comparing the land use data from 2000 and 2020, the proportion of grassland increased by 6%, the proportion of water area increased by 4.4%, and the proportion of unused land decreased by 9.6%. In summary, after the implementation of the ecological water conveyance project, the ecological environment quality of the Lake Taitema area gradually improved, and ecological water conveyance was the main driving factor of the area change in Lake Taitema.

1. Introduction

Lakes play a vital role in maintaining biodiversity and rich flora and fauna resources in arid regions [1]. Lakes are a link among the interactions of the Earth’s various spheres, as well as important indicators of climate change and human activities [2,3]. In recent years, under the dual effects of global climate change and human social activities, the natural environment has been severely damaged, and the ecological problems caused by the long-term reduction in the inflow water volume of inland lakes have become more prominent [4]. The decline in lakeside vegetation has led to a decrease in biodiversity, shrinking lake and wetland areas, and land desertification [5,6]. Especially in the inland river basins in arid regions, lakes play an irreplaceable role in maintaining the benign cycle of the ecosystem, improving local climate, and maintaining biodiversity [7], thus effectively improving land desertification. As an important indicator of lake changes, the area can intuitively reflect the spatiotemporal distribution characteristics of water resources in the region [8]. Changes in the water area can directly affect the local ecological and social benefits [9]. Changes in lakes can sensitively record changes in local climate and human activities and comprehensively reflect the ecological environment quality of a certain region [10].

Studies on inland lakes in arid regions have shown that ecological water conveyance is directly related to maintaining the area of these lakes and restoring local ecology [11,12,13]. In arid zones, lakes are inextricably linked to the local ecological environment, and both are mutually constrained. The quality of the ecological environment can reflect the overall indicators of a region, such as biodiversity, the degree of land desertification, and the status of water resources such as lakes and rivers. To protect ecological security in arid regions, the Chinese government has implemented ecological water conveyance in various areas, such as the Tarim River, Shiyang River, and Heihe River basins [14,15,16]. Currently, previous scholars have mainly conducted research on the changes in lake ecological environment quality after ecological water conveyance in three aspects: Some scholars have evaluated vegetation recovery after ecological water conveyance based on ecosystem functional indicators, such as net primary productivity (NPP) [17], gross primary productivity (GPP) [18], water use efficiency (WUE) of vegetation [19], and normalized difference vegetation index (NDVI) [20]. Some scholars have evaluated changes in hydrological conditions after ecological water conveyance, such as changes in groundwater depth, groundwater quality, and surface water area [21,22,23]. Others have evaluated soil quality after ecological water conveyance using indicators such as the temperature vegetation dryness index (TVDI) [24], soil nutrients [25], and soil moisture [26]. Researchers have found that after ecological water conveyance, the number of natural vegetation species, growth, and vegetation coverage in the local area have gradually recovered [27,28,29] and biodiversity has continuously increased [30,31]. In most areas where ecological water conveyance has been implemented, hydrological conditions have significantly improved; water quality has greatly improved; and groundwater levels have significantly increased [32]. Soil quality has become more moist, and soil nutrient content is higher closer to rivers [25,33]. In order to achieve ecological water conveyance, Daxihaizi Reservoir was established to regulate water resources and to convey water from the Tarim River to Lake Taitema. So far, the Tarim River ecological water conveyance project has been implemented for more than 20 years, and studies on the middle and lower reaches of the Tarim River and its terminal lake, Lake Taitema, have been carried out from several perspectives, but without correlation analysis among them, mainly focusing on vegetation, hydrology, soil properties, and climate evolution.

The results of previous studies have shown that vegetation and biodiversity have gradually recovered, water resources conditions in the regions have significantly improved, and soil desertification has decreased as a result of ecological water conveyance. However, there is currently no comprehensive assessment of the changes in the ecological environment of Taitema Lake, and there is no complete system to quantify and compare the changes in the local ecological environment. Further research is needed to understand the driving factors behind changes in the lake area, especially the relationship between ecological water conveyance and Taitema Lake. This study helps to further understand the relationship among ecological water conveyance, lake area, and lake ecology environment and provides a scientific basis for maintaining the water balance of rivers and lakes in arid areas, maintaining their ecological environment and protecting lake water resources. Terminal lakes are an important component of water resources in arid regions, and protecting their surface area and ecological environment is also a crucial measure for maintaining the ecological health of the entire watershed in arid areas. Therefore, this study can provide a scientific basis for ecological management in other arid regions around the world. By constructing a remote sensing ecological index model, this study can objectively and comprehensively assess the changes in the ecological environment of lakes in arid zones. The model excludes the influence of previous, artificially set thresholds and differs from previous methods for evaluating the ecological environment in that it can couple multiple indicators for comprehensive assessment, rather than being based on individual factor analysis.

This article is based on long-time-series MODIS data to clarify the changes in the area, the causes, and the ecological environment of Lake Taitema over the next 20 years. Based on remote sensing data and relevant survey data, this article analyzes the changes in lake area and land use at Taitema Lake from 2000 to 2020 after ecological water conveyance. Using a structural equation model, this article explores the correlation between temporal–spatial changes in the Taitema Lake area and ecological water conveyance, analyzes the driving factors of the temporal–spatial changes in the Taitema Lake area, and identifies the dominant factor affecting the lake’s size changes. By constructing a remote sensing ecological index model, this article systematically evaluates the ecological quality of the Taitema Lake area, assesses the changes in the ecological environment after ecological water conveyance, and helps to further understand the relationship between ecological water conveyance and the ecological environment. The purpose of this paper is to analyze and study the effect of ecological water conveyance projects on the restoration of the ecological environment of degraded arid zone lakes and the correlation between the changes in the area of arid zone lakes and ecological water conveyance. The results show that with the ecological water conveyance project, the area and ecological environment of Lake Taitema have been gradually restored.

2. Materials and Methods

2.1. Overview of the Study Area

The Tarim River basin in Xinjiang is the longest inland river in China [34]. It has experienced phenomena such as lake drying, widespread vegetation degradation, and aggravated land desertification, and the ecological environment is relatively poor. Taitema Lake, in the lower reaches of the Tarim River, is a typical representative of tail water lakes, which can maintain the ecological stability of the source water [35]. Taitema Lake is a huge ecological barrier located between the Quruq Desert and the Taklamakan Desert [36]. It has important ecological significance, and ecological water conveyance is crucial to maintaining the area and protecting ecological security in the arid tail water lake.

Taitema Lake (39°05′–39°45′ N and 87°20′–88°40′ E, with an average elevation of 807 m [37]) is located in the southeastern part of the Tarim basin in Xinjiang Uygur Autonomous Region, in the north of Ruoqiang County, 55 km away from the county seat (Figure 1). Lake Taitema has a warm, temperate, continental desert arid climate, with average annual precipitation of 28.5 mm, maximum annual precipitation of 118.0 mm, average annual evaporation of 2920.2 mm, and maximum annual evaporation of 3368.1 mm. The water sources of Lake Taitema are mainly the Tarim River and the Chechen River [38]. However, the formation of their water resources, water supply frequency, and flood frequency are significantly different. On the one hand, the Cherchen River primarily originates from the Altun Mountains, which are in closer proximity to Taitema Lake. The river outflow mostly enters the lake area. On the other hand, the Tarim River mainly relies on melting ice and snowmelt from the West Tianshan Mountains and the Karakoram Mountains.

Before 1921, for 1500 years, lake water flowed south into Lake Karakorum. In the past, the river water downstream of the Tarim River used to flow eastward into the Kongque River and eventually emptied into Lob Nor in Ruoqiang County. However, the course of the river water has changed, and it now flows southeast into Taitema Lake. According to the topographic map of aerial photographs taken by the National Mapping Bureau in 1959, Highway 218 divides Taitema Lake into two parts: the western part is supplied by the Cherchen River and has a slightly larger water area, while the eastern part is supplied by the Tarim River and has a slightly smaller water area [36]. However, excessive development and utilization of water and soil resources, especially the construction of Daxihaizi Reservoir downstream of the river in 1972, directly led to the drying up of the Tarim River’s and the Cherchen River’s downward streams [37]. As a result, Taitema Lake, as the tail water lake of the Tarim River, significantly shrank and eventually dried up, causing serious ecological degradation in the region [35]. Before 1980, the center of Taitema Lake was about 20 km east of the Kuche–Ruoqiang highway. After it dried up, the lakebed was quickly buried by the desert and became flat land.

In order to improve the increasingly severe ecological problems and water resource crisis in the region, ecological water conveyance has been implemented in the Tarim River basin since 2000 [38]. Since the implementation of ecological water conveyance, the center of Taitema Lake has moved to the west of the Kuche–Ruoqiang highway, about 30 to 40 km westward of the old site. This is because the ecological environment in this area was extremely poor, and the Quruq Desert in the northeast invaded southwest, raising sand and causing the Tarim River’s ecological water to be unable to flow into the original lake and to be only able to move westward.

More than 20 interventions of ecological water conveyance have been carried out in the lower reaches of the Tarim River from 2000 to 2020, and the water head can reach Taitema Lake. The cumulative amount of water delivered to the lake over the 20-year period was approximately 2 × 10⁹ m³, with an average annual amount of 1 × 10⁸ m³. This has resulted in the formation of a lake area spanning several hundred square kilometers. When the ecological water conveyance project was first implemented, the area of Lake Taitema was only 1.79 km²; by 2017, the area expanded to about 402 km², and the average surface area of Lake Taitema over the selected 20 years was 114.78 km². The ecological quality of the Lake Taitema area has also significantly improved.

2.2. Research Program

Based on remote sensing images and measured data, we analyzed the spatiotemporal changes in the area of Lake Taitema and its surrounding ecological environment after ecological water conveyance from 2000 to 2020. With a comparative analysis of land use/cover changes in the Lake Taitema area, we clarified the land use pattern after ecological water conveyance. A structural equation model revealed the relationship between the Lake Taitema area and various driving factors. The study aimed to evaluate the ecological environment of Lake Taitema by constructing a remote sensing ecological index model. The model offers comprehensive insights into the impact of ecological water conveyance on the region’s ecological environment. Additionally, the study quantified the changes in the ecological environment of the area with the use of remote sensing data (Figure 2).

2.3. Data Sources

This study used MODIS product collection as the data source and leveraged the GEE platform to load MODIS data from 2000 to 2020 relative to the study area using the JavaScript API. Atmospheric correction was then performed to calculate the required RSEI data. The remote sensing ecological index was derived from the MOD09A1, MOD13A1, and MOD11A2 datasets released by NASA, with 500 m resolution (16-day synthesis). The NDVI vegetation index was extracted from the MOD13A1 V6 image collection, and time-series LST images for the temperature index were obtained from the MOD11A2 product. The surface reflectance data from the MOD09A1 V6 product were used to extract the humidity and dryness indices.

Land use data were obtained from Institute of Geographic Sciences and Natural Resources Research/Geographic National Conditions Monitoring Cloud Platform of the Chinese Academy of Sciences, with spatial resolution of 30 m. Climate data were sourced from the Chinese monthly average temperature and precipitation dataset, with spatial resolution of 1 km and time range of January 2000–December 2020. The data were CRU’s (The Climatic Research Unit at University of East Anglia, Norwich, UK) 0.5° global climate data and WorldClim’s high-resolution global climate data, downscaled using the Delta spatial downsizing scheme in the Chinese region. Temperature is measured in 0.1 °C, and precipitation is measured in 0.1 mm. Ecological water conveyance and Taitema Lake area data came from the Tarim River Basin Management Bureau. Gross Domestic Product (GDP) is the core indicator of national economic accounting and an important indicator to measure the economic status and development level of a country or region. Since Lake Taitema is a protected area, it belongs to Ruoqiang County, so this paper selected Ruoqiang County’s population and GDP data. Population and GDP data were obtained from Xinjiang Statistical Yearbook from 2000 to 2020 and included end-of-year county GDP (in ten thousand yuan (CNY)) and end-of-year population (in ten thousand people).

2.4. Calculation Method

2.4.1. Mann–Kendall Trend Analysis

The Mann–Kendall test is a nonparametric test method. This method does not require its samples to follow a normal distribution, and a few abnormal values do not interfere with the results. The calculation is relatively simple, which makes it a widely used trend analysis method at present [39].

Construct a time series $X$; successively, compare $X=\left\{x_1,x_2,x_3,\dots ,x_n\right\}$ (where n is the sequence length); and obtain the Mann–Kendall statistical calculation result, $S$, as follows:

$$S={{\displaystyle \sum }}_{i=1}^{n-1}{{\displaystyle \sum }}_{j=i+1}^{n}sgn(x_j-x_i) ,$$

(1)

where $x_i$ and $x_j$ are random variables and $sgn\left(x_j-x_i\right)$ is represented by $sgn\left(\alpha \right)$:

$$sgn\left(\alpha \right)=\left\{\begin{array}{c}+1,\alpha >0\\ 0, \alpha =0 \\ -1,\alpha <0\end{array}\right.,$$

(2)

The standardized test statistic, Z, is as follows:

$$Z=\left\{\begin{array}{c}\frac{S-1}{\sqrt{Var\left(S\right)}},S>0\\ 0 , S=0\\ \frac{S+1}{\sqrt{Var\left(S\right)}}, S<0\end{array}\right.,$$

(3)

In the Mann–Kendall test, it is generally assumed that there is no obvious change trend of time series $X$. The value range of test statistic $Z$ is (−∞, +∞), and confidence interval $\beta$ is specified. If $\left|Z\right|>Z_{1-a/2}$, the original assumption is not tenable, and time series X has an obvious change trend. When Z is positive, it means that time series $X$ shows an upward trend; otherwise, it shows a downward trend.

2.4.2. Mann–Kendall Mutation Analysis

On the basis of time series $X$, construct order sequence $S_k$. $S_k$ indicates that the value at time $i$ is greater than the cumulative number of values at time $j$, and the results are as follows:

$$S_k={{\displaystyle \sum }}_{i=1}^k{r}_i k=1,2,\dots ,n,$$

(4)

$$r_i=\left\{\begin{array}{c} 1, x_i>x_j\\ 0, else\end{array}\right. \left(j=1,2,\dots ,i\right) ,$$

(5)

The defined test statistic, $U{F}_k$, is as follows:

$$U{F}_k=\left|S_k-E\left(S_k\right)\right|/\sqrt{Var\left(S_k\right)} \left(k=1,2,\dots ,n\right),$$

(6)

where the formulas of the mean and variance of $S_k$ are as follows:

$$E\left(S_k\right)=k\left(k+1\right)/4,$$

(7)

$$Var\left(S_k\right)=k\left(k-1\right)\left(2k+5\right)/72,$$

(8)

In the Mann–Whitney mutation analysis, $U{F}_k$ follows the standard normal distribution. Specify significance level $\beta$, and query the normal distribution table; if $\left|U{F}_k\right|>U_{a/2}$, it means that order sequence $S_k$ has an obvious trend of change. When $U{F}_k$ is positive, it means that order column $S_k$ is in an upward trend; otherwise, it is a downward trend. UF and UB are the two curves used in the Mann–Kendall mutation test to determine the specific mutation time. If the value of UF or UB is greater than 0, it indicates an upward trend of the series, and less than 0 indicates a downward trend. When they exceed the critical straight line, it indicates a significant upward or downward trend. The range beyond the critical line is determined as the time region where the mutation occurs. The intersection of the UF and UB curves is within the confidence interval [−1.96 1.96]; the specific year of the intersection is determined; and the moment corresponding to the intersection is the time when the mutation starts. The intersection point is the time of the start of the mutation, which means that the parameter shows a mutational increase in that year.

2.4.3. Remote Sensing Ecological Index Model

Using remote sensing image data, the normalized difference vegetation index (NDVI), the dryness index (NDBSI), the humidity index (WET), and Land Surface Temperature (LST) were extracted to represent four indicators of greenness, dryness, humidity, and heat, respectively, and the remote sensing ecological index (RSEI) was constructed to analyze and evaluate the changes in the ecological environment quality of Taitema Lake [21].

(1)

Greenness index

The normalized difference vegetation index (NDVI) was chosen to express the greenness index [40]. This index is the best indicator of plant growth status and vegetation density distribution. It is constructed by using the special red-light absorption valley of vegetation and the characteristics of high reflectance in the near-infrared band. The formula is as follows:

$$NDVI=\left({\rho }_{NIR}-{\rho }_{Red}\right)/\left({\rho }_{NIR}+{\rho }_{Red}\right),$$

(9)

(2)

Dryness index

The dryness index (NDBSI) was chosen to represent the dryness index, that is, the degree of “drying” of the surface. This indicator is not the opposite of the humidity indicator. It also includes the hardening degree of the ground. Therefore, in addition to bare soil, it is also necessary to consider building land and analyze the status of land desertification and land degradation using soil brightness. The formula is as follows:

$$NDBSI=\left(SI+IBI\right)/2,$$

(10)

where SI is the bare soil index and IBI is the construction index, which is calculated using the following formula:

$$SI=\left[\left({\rho }_{SWIR1}+{\rho }_{Red}\right)-\left({\rho }_{NIR}+{\rho }_{Blue}\right)\right]/\left[\left({\rho }_{SWIR1}+{\rho }_{Red}\right)+\left({\rho }_{NIR}+{\rho }_{Blue}\right)\right],$$

(11)

$$IBI=\frac{\left\{2{\rho }_{SWIRI}+\left({\rho }_{SWIR1}+{\rho }_{NIR}\right)-\left[\frac{{\rho }_{NIR}}{{\rho }_{NIR}+{\rho }_{Red}}+\frac{{\rho }_{Green}}{\left({\rho }_{Green}+{\rho }_{SWIR1}\right)}\right]\right\}}{\left\{2{\rho }_{SWIRI}+\left({\rho }_{SWIR1}+{\rho }_{NIR}\right)+\left[\frac{{\rho }_{NIR}}{{\rho }_{NIR}+{\rho }_{Red}}+\frac{{\rho }_{Green}}{\left({\rho }_{Green}+{\rho }_{SWIR1}\right)}\right]\right\}},$$

(12)

(3)

Humidity index

The humidity component (WET) was selected to represent the humidity index. The humidity component is obtained using remote sensing tassel cap transformation, which is closely related to vegetation and soil moisture. The formula is as follows:

$$WE{T}_{TM}=0.0315{\rho }_{Blue}+0.2021{\rho }_{Green}+0.3102{\rho }_{Red}+0.1594{\rho }_{NIR}-0.6806{\rho }_{SWIR1}-0.6109{\rho }_{SWIR2},$$

(13)

$$WE{T}_{OLI}=0.1511{\rho }_{Blue}+0.1973{\rho }_{Green}+0.3283{\rho }_{Red}+0.3407{\rho }_{NIR}-0.7117{\rho }_{SWIR1}-0.4559{\rho }_{SWIR2},$$

(14)

(4)

Heat index

In this paper, surface temperature (LST) was selected to represent the heat index, and surface temperature is derived from thermal infrared radiation.

To sum up, in all formulas, ${\rho }_{NIR}$ is the reflectance of the near-infrared band; ${\rho }_{SWIR1 }$ and ${\rho }_{SWIR2}$ are the reflectance of shortwave infrared band 1 and the reflectance of shortwave infrared band 2, respectively. ${\rho }_{Red}$, ${\rho }_{Blue}$, and ${\rho }_{Green }$ are the reflectance of the red band, blue band, and green band, respectively.

After calculating all the indicators, we needed a standard to measure the quality of the ecological environment, so we needed to use the principal component analysis method. Principal component analysis (PCA) is an analytical and simplified data processing technique in multivariate statistical methods. Due to the different measurement standards of each indicator, it was necessary to normalize the indicator and unify the value of the indicator between 0 and 1. The normalization formula is as follows:

$$N{I}_i=\left(I_i-I_{min}\right)/(I_{max}-I_{min}),$$

(15)

where $N{I}_{i }$ is the value of an index after normalization, $I_i$ is the value of this index at pixel $i$, $I_{min}$ is the minimum value, and $I_{max}$ is the maximum value of this indicator.

The four normalized indices were analyzed using ENVI software. In order to make a large PC1 represent good ecological condition, the initial ecological index ($RSE{I}_0$) was obtained by subtracting the calculated PC1 from 1, and the final remote sensing ecological index ($RSEI$) was obtained by normalizing it. The formula is as follows:

$$RSE{I}_0=1-PC1,$$

(16)

$$RSEI=\left(RSE{I}_0-RSE{I}_{0min}\right)/\left(RSE{I}_{0max}-RSE{I}_{0min}\right),$$

(17)

where RSEI is the remote sensing ecological index, and the larger the value is, the higher the ecological environment quality is; otherwise, the lower the ecological quality is.

2.4.4. Structural Equation Model

The structural equation model can simultaneously analyze a group of equations with interrelation, especially those with causality between potential exogenous variables and potential endogenous variables. The structural equation model is divided into measurement model and structural model. The structural equation is as follows:

$$\eta =B\eta +\Gamma \xi +\zeta ,$$

(18)

where $\eta$ is the potential dependent variable matrix, $\xi$ is the potential independent variable matrix, $\Gamma$ and $B$ are the structure coefficient matrices, and $\zeta$ is the residual matrix of the structural equation. The measurement equation is as follows:

$$\left\{\begin{array}{c}X={\Lambda }_X\xi +\delta \\ Y={\Lambda }_Y\eta +\epsilon \end{array}\right.,$$

(19)

where $X$ and $Y$ are the measurement variable matrices of $\xi$ and $\eta$ respectively; ${\Lambda }_X$ and ${\Lambda }_Y$ are both measurement coefficient matrices; and $\delta$ and $\epsilon$ are the residual matrices of the measurement equation.

3. Results

3.1. Temporal and Spatial Variations in Lake Area of Taitema Lake

According to the data of the Taitema Lake area from 2000 to 2020, the Mann–Kendall test was used to analyze the trend and mutation of the lake area. The test results show that the Taitema Lake area increased in the selected 20 years, passing the 0.05 significance test, indicating a significant increasing trend of the Taitema Lake area.

From an annual perspective, the inter-annual variation in the area of Taitema Lake was as follows: The area of Taitema Lake was only 1.8 km² in 2000 and reached 402.12 km² in 2017, which was the largest area in nearly 20 years (Figure 3a). The average lake area over 20 years was 114.8 km², with an annual growth rate of 5.58 km² a⁻¹, and the area change was significant (p < 0.05). Based on the position of the intersection of the UF and UB curves, it was determined that the increase in the area of Taitema Lake was a sudden change, specifically starting from around 2002 and 2011 (Figure 3b). The sudden change test of Taitema Lake indicates that there were an extremely significant expansion in 2011 (p < 0.01) and a significant expansion in 2002 (p < 0.05).

Due to factors such as snow cover and lake ice, this article selected the monthly average area changes in Taitema Lake from May to October for study (Figure 4a). The overall trend of the lake area from May to October was one of increase, which was basically consistent with the annual scale of lake area changes. The annual changes in the Taitema Lake area were as follows: From May to June, the area showed a shrinking trend, decreasing from 211.6 km² to 173.3 km², and then an overall expanding trend month after month, reaching the maximum value in October. From a monthly scale perspective (Figure 4b), the area of Taitema Lake was 26.79 km² in October 2002, and it expanded to 139.71 km² by May 2003. During this period, the ecological water conveyance volume increased from 3.31 × 10⁸ m³ to 6.25 × 10⁸ m³. By August 2009, the area of Taitema Lake shrank to 1.79 km², and the corresponding ecological water conveyance volume was 0.11 × 10⁸ m³. With the increase in ecological water conveyance volume, the area of Taitema Lake reached 149.95 km² in June 2011. The maximum area of Taitema Lake occurred in October 2017, which was also the peak of the selected 20 years in terms of ecological water conveyance volume.

3.2. Temporal and Spatial Change Characteristics of Land Use at Taitema Lake

From 2000 to 2020, the overall grassland area showed an increasing trend, especially in the eastern region of Lake Taitema, where the growth was the most obvious. The water area also expanded, gradually forming a lake from the original small pond, and the unused land area continued to decrease (Figure 5).

The proportions of land use types in Taitema Lake in 2020 are shown in

Table 1. Proportion of the land use area of Taitema Lake between 2000 and 2020. (Unit: %.).

Date	Wood Land	Grassland	Water	Construction Land	Unused Land
2000	1.6	22.6	0.4	0.0	75.4
2010	0.6	22.6	3.3	0.1	73.3
2015	0.6	21.1	4.5	0.1	73.6
2020	0.6	28.6	4.8	0.1	65.8
	−0.9	6.0	4.4	0.1	−9.6

. By comparing the land use ratio of Lake Taitema from 2000 to 2020, we can see that the proportion of grassland increased by 6%, the largest increase; the proportion of water area increased by 4.4% compared with 2000; the proportion of construction land increased by 0.1%; the proportion of unused land decreased by 9.2%. Among them, the proportion of grassland was 22.6% in 2000, and the proportion of water was only 0.4%. With ecological water conveyance, the proportion of water area increased year by year and reached 4.8% in 2020, while the proportion of grassland area as a whole showed an increasing trend and increased to 28.6% in 2020.

3.3. Temporal and Spatial Variation Characteristics of Eco-Environmental Quality of Taitema Lake

Based on the analysis of remote sensing ecological indices of Taitema Lake from 2000 to 2020, it could be observed that the proportion of poor ecological condition showed an overall downward trend, while the proportions of moderate, good, and excellent ecological conditions showed an overall upward trend, indicating that the ecological condition of the area gradually improved (Figure 6).

Among them, in 2000, the proportion of poor ecological condition was 94.7%, which decreased to 54.4% by 2020, a decrease of about 40.4%; the proportion of moderate ecological quality increased from 5.3% to 35%, an increase of about 29.7%, which was the largest increase (Figure 7); the increase in the proportion of good ecological quality was second, with 0% in 2000 and 10% in 2020; the proportion of excellent ecological quality increased by 0.6% over the 20 years, indicating that poor ecological quality is gradually being replaced by moderate, good, and excellent quality. Due to ecological water conveyance, the area of Taitema Lake had significant increases in 2012 and 2017. In 2012, the proportion of poor ecological condition was 38.7%, which was the smallest in the selected 20 years. The proportions of moderate and good ecological conditions were 46.1% and 14.7%, respectively. In 2017, the proportions of moderate and good ecological conditions were 38.2% and 13.2%, respectively.

According to the RSEI mean analysis of 20 a, the proportion of areas with poor habitat quality was 47.7%, while the proportions of areas with medium, good, and excellent habitat quality were 32.6%, 18.4%, and 1.4%, respectively. The sum of the three was greater than the proportion of areas with poor habitat quality, indicating an improvement in the ecological condition of Taitema Lake (Figure 8). Using the Mann–Kendall trend test method to examine the ecological quality change on an element-by-element basis and referring to the trend test value ($Z_c$), the spatial distribution of the RSEI trend test after ecological water conveyance was obtained. From 2000 to 2020, 56.5% of the area of elements in Lake Taitema showed an increasing trend of the RSEI ($Z_c$ > 0), while 43.5% of the area showed a decreasing trend of the RSEI ($Z_c<0$).

In particular, the proportion of pixels with a significant upward trend of the RSEI ($Z_c$ > 1.96) was 15.6%, while the proportion of pixels with a significant downward trend of the RSEI ($Z_c$ < −1.96) was smaller, only 2.1%. The proportions of pixels with a non-significant upward trend (0 < $Z_c$ < 1.96) and a non-significant downward trend (−1.96 < $Z_c$ < 0) of the RSEI were 40.9% and 41.4%, respectively. Overall, this indicates a significant improvement in the ecological condition of the Taitema Lake area after ecological water conveyance.

3.4. Patterns of Variation in the Driving Factors of Lake Taitema

The 20-year average annual precipitation in the Lake Taitema area showed an overall upward trend with a significant increase, and the annual average temperature varied steadily. The overall increase in ecological water conveyance was basically consistent with the inter-annual variation trend of the area of Lake Taitema.

Over the selected 20 years, the annual average precipitation in the Taitema Lake area generally increased, with a significant increase in amplitude, while the average annual temperature remained stable (Figure 9). Over the selected 20 years, the annual precipitation in the Taitema Lake area showed a significant increasing trend (p < 0.05), with multi-year average precipitation of 36.1 mm, reaching the highest annual precipitation in 2019. The average temperature over the years was 11.7 °C, with no significant trend (p > 0.05), fluctuating between 11 and 12 °C.

The overall ecological water conveyance volume increased, and its trend was consistent with the inter-annual variation trend of the Taitema Lake area (Figure 10). The annual average ecological water conveyance volume over the selected 20 years was 4.02 × 10⁸ m³, and in 2011, the conveyance volume reached 8.52 × 10⁸ m³. Compared with 2010, the area of Taitema Lake expanded from 28.79 km² to 151.06 km², which coincided with the time of sudden change at Taitema Lake. In 2017, the conveyance volume reached the maximum in the 20 years considered, equal to 1.209 × 10⁹ m³, and as a consequence, the area of Taitema Lake reached 402 km², also the largest in 20 years.

4. Discussion

4.1. Analysis on the Factors of Lake Area Change Trend of Taitema Lake

The analysis of the curves of ecological water conveyance and area changes showed that the area change of Lake Taitema was highly consistent with the trend of ecological water conveyance, so it was speculated that the area change of Lake Taitema might have a close relationship with ecological water conveyance. Temperature changes in Lake Taitema were flat over the selected 20 years, as higher temperatures lead to the evaporation of water from the lake, which in turn affects the lake area. The overall change in rainfall fluctuated little, but the change in rainfall was similar to the overall trend of the change in the area of Lake Taitema. Therefore, the specific effects of temperature and rainfall on the lake need to be analyzed.

As Taitema Lake is situated in Ruoqiang County, which is also a natural reserve, this article selected the population and GDP data of Ruoqiang County as the driving factors to analyze Taitema Lake (Figure 11). With the analysis of the results of the structural equation model, it is possible to see that ecological water conveyance, precipitation, and GDP had a positive effect on the area of Taitema Lake, while population and temperature had a negative effect on the area of Taitema Lake.

Among them, the path coefficient of ecological water conveyance to the area of Taitema Lake was 0.87, which passed the test with p < 0.05 (Figure 12a). The linear relationship between ecological water conveyance and lake area was significant, indicating that ecological water conveyance had a significant impact on the area of Taitema Lake. Precipitation had a secondary impact on the area (Figure 12b), and precipitation also had a positive effect on ecological water conveyance. Temperature had a negative effect on the area of Taitema Lake, and temperature also limited ecological water conveyance. Based on the above analysis, it is possible to see that ecological water conveyance had the largest positive impact on the area of Taitema Lake. As ecological water conveyance and temperature change, the area of Taitema Lake is expected to change accordingly [41], but the main cause of the changes in the area of Taitema Lake is the ecological water conveyance. Therefore, during the rainy season, ecological water conveyance can be reduced or temporarily stopped to avoid the waste of resources. During the high-temperature season, ecological water conveyance is necessary to maintain the lake area and protect the local ecology. Due to the scarce rainfall and high evaporation in the Taitema Lake area [42], water resources mainly come from the Daxihaizi Reservoir supply, which performs ecological water conveyance downstream of the Tarim River. In summary, the change in the area of Taitema Lake is largely influenced by ecological water conveyance, and ecological water conveyance has a significant impact on inland lakes in arid areas [16]. Therefore, if we want to restore the lakes in arid areas, ecological water conveyance projects should be considered.

4.2. Analysis of the Suitable Area of Taitema Lake

Taitema Lake is difficult to maintain at a fixed and large scale [43], and it is not necessary to do so. Currently, it is the only lake downstream of the Tarim River, serving as a terminal lake where the Tarim River and the Cherchen River converge. As an inland lake at the end of the desert, Taitema Lake has no other source of water supply except for the water inflow of the Cherchen River and the Tarim River. Therefore, the size of the lake is mainly affected by the water inflow of the two rivers. The total water inflow into the lake is closely related to the water inflow of the two rivers. However, due to the obvious intra-annual concentration (about 70% of the water inflow from July to September), there are significant inter-annual differences in the natural runoff of the rivers in the arid area [44]. In addition, the strong evapotranspiration of the lake area, the flat lake-bottom shape, and other factors determine that there must be strong intra-annual and inter-annual changes in the lake area of Taitema Lake.

From 2000 to 2020, the average ecological water conveyance to Taitema Lake was 4.02 × 10⁸ m³ per year, while the average lake area during this period was 114.79 km². The analysis showed a significant linear relationship between ecological water conveyance and the lake area (p < 0.05), which can be used to estimate the lake area at different water levels. Based on the linear fitting equation, the lake area reached its maximum in 2017, which was also the year with the highest water conveyance. The increase in lake area in 2003 was also due to an increase in water conveyance. In the drought years from 2007 to 2009, the lake area decreased to about 35 km² in 2007, and the vegetation area decreased by 4.97% in 2008 and 2009. The lake and its surrounding areas formed fish-scale sand dunes again in 2008 and 2009 [45], leading to mild desertification. If the lake area of Taitema Lake is continuously less than 30 km², the ecological environment in the Taitema Lake area will rapidly deteriorate. It was not until ecological water conveyance exceeded 2 × 10⁹ m³ in 2010 that the trend of ecological degradation was reversed. Therefore, in order to avoid serious ecological degradation in the lake area, the surface area of Taitema Lake should not be less than 30 km². Using data analysis of the vegetation area, the average RSEI value, and the lake area around Lake Taitema, it was estimated that when the lake reaches its maximum area of 110 km², the corresponding vegetation area is 232 km². Without maintaining a large lake area, the natural vegetation area and quality of the lake area can be maintained in good condition. At the same time, from the results of the RSEI, it is possible to see that the vegetation of the lake area can adapt to the strong annual and inter-annual variations in the lake area. Therefore, it is not necessary to maintain a larger and fixed lake area. A too-small lake area would cause poor water conditions for vegetation distribution, especially in the vegetation distribution areas far away from the lake area [46], affecting the normal growth of vegetation. Prolonged “small lake area” conditions would lead to ecological environmental problems such as decreased vegetation quality and decreased area.

The restoration of Lake Taitema marks the end of the dry season in the main stream of the Tarim River and the lower reaches of the Cherchen River, and the restoration of ecological water supply to the downstream green corridor, achieving the goal of comprehensive management of the Tarim River basin in the near future. From the current water surface formation of Lake Taitema, it is observed that the water volume and peak flow rate of the Tarim River into the lake are both high [47,48], indicating the limited capacity of the Tarim River source, and upper and middle reaches regulation projects. Due to various factors, such as source water inflow and losses along the way, the lake area of Lake Taitema can only be rapidly expanded in the short term using ecological water conveyance during the flood season, while ensuring the supply of water for domestic and industrial use. Despite improvements in water supply efficiency in the lower reaches of the Tarim River, there is still room to optimize water resource use and achieve the unified regulation of water resources throughout the entire basin. Therefore, it is necessary to further strengthen the capacity of source and mainstream regulation projects in the Tarim River basin [49]. Water resources are the primary limiting factor in ecological environment safety, and sustainable economic and social development in arid areas [50,51]. For Taitema Lake, it is important to use the currently available ecological reservoirs for artificial regulation and operation. To prevent ecological water from entering Taitema Lake all at once and to avoid large-scale attenuation of the lake area during hot and dry periods, leading to environmental deterioration, a more scientific and reasonable water resource regulation plan should be developed based on the relationship between water discharge from the Tarim River and the lake area [52]. According to analysis and research on the average water level change process and frequency of Taitema Lake over the past 20 years, taking into account ecological water conveyance and climate factors, maintaining a water area of 30–110 km² for Taitema Lake is relatively suitable.

5. Conclusions

(1)

Ecological water conveyance has significantly increased the area of Lake Taitema. Over the selected 20 years, ecological water conveyance expanded the area of Lake Taitema from 1.8 km² to approximately 402 km². The overall trend of expansion of the area of Lake Taitema is evident at both annual and monthly scales, and the lake began to expand after the head of ecological water conveyance reached it. This indicates that ecological water conveyance can alter the area of lakes.

(2)

From 2000 to 2020, significant changes occurred in the land use types of the Taitema Lake area, with an increase in grassland and water areas and a gradual decrease in unused land. The largest change in the proportion of unused land occurred over 20 years, with a decrease of about 9.6%, while the proportion of grassland and water increased by 6.0% and 4.4%, respectively. At the beginning of ecological water conveyance, the water area was very small, and the grassland was sparse. However, as ecological water conveyance progressed, the previously small lake area became a patchy lake, and the grassland became denser. This indicates that the spatial conversion of land use in the Taitema Lake area showed a trend of improvement due to ecological water conveyance.

(3)

Ecological water conveyance has gradually improved the habitat quality of Taitema Lake, and areas with poor ecological quality have been gradually replaced by areas with moderate, good, and excellent ecological quality. The Mann–Kendall test results show that over the selected 20 years, the proportion of pixels with an increasing RSEI ($Z_c$ > 0) in the Taitema Lake area was 56.5%, while the proportion of pixels with a declining RSEI ($Z_c$ < 0) was 43.5%, indicating a significant improvement in ecological quality in the Taitema Lake area over the selected 20 years. The change in ecological water conveyance was generally consistent with the change in the ecological environment. It is possible to see that ecological water conveyance plays an extremely important role in restoring ecology. In arid areas with scarce rainfall and insufficient water resources, the most direct and effective method to restore ecology is ecological water conveyance.