Study on the Spatial–Temporal Variation of Groundwater Depth and Its Impact on Vegetation Coverage in Ejina Oasis

Abstract

Ejina, a representative inland river basin situated in the arid region of northwest China, exhibits a delicate ecological environment and its vegetation coverage is intrinsically linked to regional ecological security. Based on MOD13Q1-NDVI data from 2018 to 2023 and groundwater depth monitoring data during the same period, this study analyzed the spatial–temporal variation characteristics of vegetation coverage and its relationship with groundwater depth in Ejina. It is found that the vegetation coverage in Ejina is generally low and mainly distributed along the riverbanks in the form of strips. During the study period, the overall trend of vegetation coverage showed a fluctuating pattern of first increasing and then decreasing, revealing the fragility of the regional ecology. The groundwater depth shows the characteristic of being higher in the east river than the west, and the trend of groundwater depth along the river flow is first increasing and then decreasing. The spatial groundwater depth indicates that the east river is higher than that of the west river, and the groundwater depth along the river flow first increases and then decreases. In terms of inter-annual changes, the groundwater depth experiences a process of first decreasing and then stabilizing. Further analysis indicates that vegetation growth and coverage in Ejina are significantly affected by water conditions, and areas with high Normalized Difference Vegetation Index (NDVI) values are mainly distributed along the riverbanks. In addition, there is a certain degree of correlation between groundwater depth and NDVI. When the depth of groundwater is too deep or too shallow, the positive correlation between NDVI and groundwater depth increases slightly and the negative correlation decreases slightly. The findings of this study are of great significance for understanding and predicting the response of vegetation coverage to groundwater changes in arid areas, and provide a scientific basis for water resources management and ecological protection in Ejina.

1. Introduction

The ecological environment in northwest China is extremely fragile, and its ecological system’s stability is facing a great threat due to global climate change and human activities [1,2]. Vegetation, as an essential component of the terrestrial ecosystem, directly participates in the material cycle and energy flow process through its interactions with soil, water, and atmosphere, playing an important role in maintaining the balance of the surface ecosystem [3,4]. The dynamic changes in vegetation cover can be used to characterize the ecological and environmental conditions of the region and are considered to be an important indicator of the ecological environment and climate change in arid areas [5,6]. In recent years, with the rapid development of remote sensing technology, it has become the main monitoring method for dynamic changes in vegetation cover and is widely utilized for vegetation change monitoring at different scales [3,7]. The NDVI inverted from remote sensing data is sensitive to the growth rate and vigor of vegetation and is considered to be the best indicator of vegetation cover change. It has been widely applied in related studies at different scales and in various regions at home and abroad [8,9]. Many scholars have conducted extensive research on vegetation coverage changes and related factors in the arid areas of northwest China. Related studies have shown that the change in vegetation NDVI in the arid region of northwest China is highly correlated with the current precipitation and evapotranspiration during the growing season [10]. During prolonged droughts, groundwater contributes significantly to the effective soil moisture required to maintain vegetation growth, and there is a strong correlation between groundwater depth and NDVI [11,12]. In arid areas, water is the foundation for the composition, development, and stability of oasis ecosystems. However, due to the scarcity of precipitation and high evaporation rates in arid regions, groundwater is particularly important for the ecological environment and vegetation ecosystem [13]. When the groundwater level is too shallow, strong evapotranspiration will cause a large amount of submersible evaporation, soil salinity will accumulate in the topsoil, and vegetation growth will be inhibited, which will lead to a decrease in vegetation coverage. When the groundwater level is too deep, the plant roots do not receive enough water to affect their growth. Therefore, in arid desert systems, groundwater has a profound impact on the composition, distribution, and growth rate of vegetation, and ultimately affects the vegetation growth status and coverage.

Ejina is located in the arid interior of northwest China, with little rainfall and scarce water resources, making its ecosystem fragile. The Ejina Oasis plays an important role in maintaining the stability of the desert vegetation ecosystem, but the evolution of desertification in the oasis still needs to be given high attention and importance due to the impact of global climate change and human activities [14,15]. The Ejina Oasis serves as the first line of defense against wind and sand erosion, protecting the ecological environment, and is also an important ecological buffer zone, affecting the ecological security of northwest and north China [16]. Therefore, it is of great scientific significance and research value to study the changes in vegetation cover in Ejina Oasis and its response to groundwater depth. Based on this, this paper used the MOD13Q1-NDVI data from 2018 to 2023 to analyze spatial–temporal change characteristics and dynamic change trends of vegetation cover in Ejina Oasis. Combined with the contemporaneous real-time monitoring data of the National Groundwater Monitoring Network during the same period, this paper explored the impact of groundwater depth on the spatial pattern of vegetation and proposed the range of groundwater that affects vegetation growth, in order to provide a scientific basis for the restoration of the ecosystem and the protection of the fragile environment in this area.

2. Materials and Methods

2.1. Overview of the Study Area

The study area (40°51′~42°32′ N, 100°05′~101°24′ E), located in the westernmost part of Inner Mongolia Autonomous Region, China, is adjacent to Alxa Right Banner to the southeast, Gansu Province to the west and southwest, and Mongolia to the north. The elevation ranges from 892 m to 1567 m, with a total area of approximately 7357 km² (Figure 1). Ejina Banner is a typical continental arid area characterized by little rainfall, high evaporation, abundant sunshine, and frequent sandstorms. The average annual temperature is 8.3 °C, the average annual precipitation is 37.0 mm, the average annual evaporation is 3841.5 mm, and the weather is often windy, with strong winds often accompanied by sandstorms. The landscape is mainly composed of deserts, low mountains, rivers, lakes, and oases, with most of the area being covered by deserts and a zone where oases and deserts intermingle. The vegetation community is relatively poor, with the main vegetation types being desert riverbank forests, desert shrub forests, and desert meadows [15,17].

2.2. Data Sources and Processing

In this study, the NDVI data used were obtained from the NASA website (https://ladsweb.modaps.eosdis.nasa.gov, accessed on 10 October 2024) of the US Aerospace Agency. The MODIS 13Q1 NDVI data around the lower reaches of the Heihe River, also named the Ejina River, from 2018 to 2023 were downloaded from the website, with a spatial resolution of 250 m and a temporal resolution of 16 days. To ensure the accuracy of NDVI data, this study combined quality control documents with Savitzky–Golay (S-G) filtering to eliminate noise caused by cloud layers and atmospheric aerosols, thereby obtaining high-quality monthly vegetation index products.

Groundwater data mainly come from the China Groundwater Monitoring Project Information Service Network (http://jcgc.cigem.cn/, accessed on 10 October 2024). There are a total of 24 observation wells within the study area, with a time span of nearly 6 years (2018~2023). The data types include the latitude and longitude coordinates, ground elevation, and monthly groundwater depth information, with a data accuracy of 0.01 m.

2.3. Research Methods

There is a significant linear correlation between vegetation coverage and NDVI, but due to the influence of clouds and rain, remote sensing data cannot effectively observe the ground, and this effect is manifested as low-value noise in NDVI. The maximum synthesis method can effectively eliminate the influence of low-value noise [18]. Therefore, this study performs maximum processing on the monthly NDVI data collected twice a month to obtain a 6-year monthly sequence dataset and analyzes the changes in NDVI of the Ejina Oasis.

Groundwater depth data were mainly derived from 24 shallow groundwater level monitoring wells within the study area. The spatial analysis module in Arcgis 10.7 software, with the same resolution as the DEM, was applied to generate the contour map of groundwater depth within the study area. By dividing the area into different depth intervals and calculating the average NDVI value for each interval, a curve graph of groundwater depth versus NDVI average value can be drawn to analyze the relationship between groundwater table depth and NDVI spatial distribution.

The Pearson correlation is often employed as a statistical measure to quantify the degree of linear correlation between two variables [13,19]. To investigate the impact of groundwater depth changes on vegetation coverage, this study used 29,661 pixels within the study area to calculate the correlation coefficient between changes in vegetation coverage and groundwater depth changes from 2018 to 2023. The calculation formula is as follows:

$$R={\displaystyle \frac{{{\displaystyle \sum }}_{i=1}^n(X_i-\stackrel{-}{X})(Y_i-\stackrel{-}{Y})}{\sqrt{{{\displaystyle \sum }}_{i=1}^n(X_i-\stackrel{-}{X}{)}^2}+\sqrt{{{\displaystyle \sum }}_{i=1}^n(Y_i-\stackrel{-}{Y}{)}^2}}}$$

(1)

In this formula, R is the Pearson correlation coefficient, X is the annual groundwater depth (m), Y is the annual value, and represents the years from 2018 to 2023.

3. Results

3.1. Spatial–Temporal Variation in Vegetation Coverage

This study used the NDVI pixel binomial model to calculate the vegetation coverage degree in Ejina over the past 6 years (2018~2023) and referenced the literature on vegetation coverage classification in arid regions [19]. Combined with field actual investigation, the vegetation coverage was divided into four grades: bare land (0–0.05), low vegetation coverage (0.05–0.25), medium vegetation coverage (0.25–0.5), and high vegetation coverage (0.5–1.0), thus obtaining the spatial distribution of vegetation coverage in 2018, 2020, and 2023 (Figure 2). The calculation results show that in 2018, 2020, and 2023, the areas with moderate to high vegetation coverage in the Ejina region accounted for only 22.5%, 24.1%, and 21.3%, respectively. The vegetation coverage in the study area is generally low, with bare land and low vegetation coverage being the main features. The areas with relatively higher vegetation coverage are mainly located in the upper and lower reaches of the east river, the west river, and the East and West Juyan Lake basin. Over the past 6 years, the overall trend of vegetation coverage has shown a fluctuating pattern of first increasing and then decreasing (Figure 2c), but the inter-annual trends of vegetation coverage are different. The high coverage area shows a trend of decreasing year by year, with a reduction rate of 25.71 km²/a. The low coverage area shows a trend of increasing annually, with an increase rate of 96 km²/a. The medium coverage area shows a trend of increasing coverage first and then decreasing, while the bare land shows a trend of decreasing first and then increasing. The medium coverage area shows an up-and-down trend, while the bare land shows a down-and-up trend. Due to the effects of natural factors and human activities, the rate of vegetation coverage is undergoing migration and transformation every year. Through a matrix analysis of vegetation coverage, it was found that the areas of low and medium vegetation coverage rose significantly from 2018 to 2020, increasing by 470.00 km² and 166.70 km², respectively. The increase in low vegetation coverage was mainly provided by bare land transfer, while the increase in medium vegetation coverage was mainly transferred from low vegetation coverage (Figure 2b). From 2020 to 2023, the vegetation coverage degree showed a slight decrease, mainly due to the transfer of medium-high vegetation coverage to low coverage and bare land, but the magnitude of the transfer was small. It is evident that the trend of vegetation degradation in Ejina has improved significantly, remaining relatively stable on the whole, but it is susceptible to fluctuations due to natural environment and human activities, and the regional ecological security still faces considerable challenges [20].

3.2. Spatial–Temporal Variations of Groundwater

3.2.1. Annual Variation

The Heihe River is divided into the east river and west river after entering the lower reaches, and the water depth has a distinct spatial distribution pattern. In this study, the east and west rivers are divided into four river basins upstream and downstream, respectively, and the dynamic change characteristics of the groundwater depth are explored using data from 2018 to 2023. Overall, the groundwater depth near the rivers is smaller than in areas far from the rivers, and the west river is generally lower than the east river. Along the flow direction of the East River, the groundwater depth shows a trend of first increasing and then decreasing, and is relatively large in the East Juyan River. Along the flow direction of the west river, the groundwater depth also shows the same trend, and it is smallest in Aohan Taolai and largest in the North Gobi to the West Juyan River. In general, along the river flow direction, the groundwater depth of the east and west rivers shows a trend of first increasing and then decreasing (Figure 3a).

From the annual variation in groundwater depth, the water depth of the east river and west River is generally the smallest in March and the largest in August, showing a clear seasonal change pattern (Figure 3b). In spring and summer (March to August), the water level tends to decrease, while in autumn and winter (September to the following February), it shows an increasing tendency. However, the groundwater depth in the upper and lower reaches of the river shows obvious seasonal variations, with the change in water level depth in the upper reaches being relatively small, and that in the lower reaches being relatively large. According to the latency and investigation data of water supply from the river to the groundwater, it can be seen that from the upper reaches of the east and west rivers to their lower reaches, the minimum average groundwater depth over many years transits from around March to April or May, and the maximum average depth also has similar changes in trend. This is mainly because groundwater is replenished by surface runoff, which starts from the upper section of the east and west rivers and then affects the middle and lower sections of the river.

3.2.2. Inter-Annual Variations

In terms of inter-annual changes, the groundwater depth in the lower reaches of the Heihe River showed a trend of first declining and then stabilizing overall from 2018 to 2023 (Figure 4b). The groundwater depth in the upper reaches of the river was shallow and fluctuated little, with a range of fluctuation generally less than 0.20 m. The groundwater depth of the river end showed larger inter-annual fluctuations, with the lower section of the east river showing a downward trend, with a decline of about 0.30 m; the lower section of the west river showed a trend of first declining and then rising, with the water level by 6.21 m in 2018 to 7.58 m in 2018–2021, and then rising year by year to 7.36 m in 2021–2023.

In this study, the groundwater depth change in the lower reaches of the Heihe River over the past 6 years (2018~2023) was interpolated and calculated to obtain a map of groundwater depth variation (Figure 4a). From 2018 to 2020, the trend of water level changes in the lower reaches of the Heihe River was quite distinct, with the water level in the upper section of the river changing basically stably, the middle section rising, and the lower section dropping significantly. Among them, the water level in the northern part of the lower reaches of the Heihe River fell the most, and the downward trend gradually decreased from north to south, until the middle section, where the water level rose slightly, with an increase of 0.1–0.5 m/a. Other areas showed no significant variation. The water level in the northern part of the lower reaches of the Heihe River continued to show a downward trend, but the decline slowed down, showing a slightly downward trend, with a decline of 0.1–0.3 m/a. In the middle and southern sections, it showed no significant trend, with a variation of less than 0.2 m/a.

Based on the comprehensive investigation and analysis, the southern part of the lower reaches of the Heihe River receives surface water recharge from the Heihe River, with water levels and depths being shallow and relatively stable, with small fluctuations [21]. In the middle and northern parts of the area, water levels and depths increase gradually due to water hydrological geological conditions and uneven distribution of surface water. The surface water recharge becomes less stable and the fluctuation range also increases [21,22]. In the northern part of the area, surface water does not recharge, and the water level and depth are the largest. Due to insufficient infiltration of atmospheric precipitation, the water level and depth decrease most significantly.

3.3. Influence of Groundwater Depth Variation on Vegetation Coverage

3.3.1. Groundwater Depth and Distribution of Vegetation Coverage

Groundwater is an important component of the ecosystem in arid areas and has a significant impact on vegetation growth and distribution [11,23]. In order to reveal the relationship between groundwater depth and vegetation coverage distribution in Ejina, spatial distribution maps of groundwater depth and vegetation index were overlaid at different times (Figure 5a). The calculation shows that the high values of NDVI in Ejina are distributed in stripes along the east and west riverbanks. When the groundwater depth is between 4–6 m, vegetation grows well and the coverage is high. When the groundwater depth exceeds 8 m, vegetation coverage generally deteriorates.

To investigate the relationship between groundwater depth and vegetation growth, we divided the groundwater depth into intervals of 0.1 m and statistically calculated the average NDVI value for each interval, then forming a relationship curve between the NDVI values and groundwater depth distribution for 2018, 2020, and 2023 (Figure 5b). In 2018, 2020, and 2023, when the groundwater depth exceeded 5.4 m, 5.7 m, and 4.9 m, respectively, the NDVI value decreased significantly in the range of 0.06–0.08. This indicates that there is a range of optimal groundwater depth for vegetation growth in the lower reaches of the Heihe River, which can be considered as the groundwater depth threshold for vegetation growth [24,25]. When the groundwater depth is less than the threshold, the vegetation index shows a marked high value, and the vegetation grows well. When the groundwater depth exceeds the threshold, the vegetation index value drops sharply, and the decline is obvious. By comparing the relationship between groundwater depth and NDVI at different times, the suitable groundwater depth for vegetation growth in the lower reaches of the Heihe River is approximately 4–5 m, with slight fluctuations in the upper and lower ranges. The statistical results are consistent with the conclusions of related studies [23], but the statistically suitable water level interval is slightly different. This may be due to the source of the data and the difference in accuracy.

3.3.2. Analysis of the Correlation Between Groundwater Depth and NDVI

To investigate the impact of groundwater depth changes on vegetation coverage in Ejina, this study calculated the Pearson correlation coefficient between groundwater depth and vegetation coverage (Figure 6a). The calculation shows that the area where the groundwater depth is positively correlated with NDVI accounts for 39.61%, while the area where the correlation is negative accounts for 37.63%. The distribution of the related areas is relatively scattered and exhibits discontinuous spot-like distribution. The distribution of the correlation is scattered and discontinuous, with a patchy distribution. Meanwhile, this study divided the groundwater depth range into 4.8 m and 8.5 m intervals and analyzed the correlation with each interval (Figure 6b). When groundwater depth is less than 4.8 m, the area with a positive correlation accounts for 40.49%, while the area with a negative correlation accounts for 37.31%. When groundwater depth is between 4.8 and 8.5 m, the area with a positive correlation accounts for 37.49%, while the area with a negative correlation accounts for 38.28%. When groundwater depth is greater than 8.5 m, the area with a positive correlation accounts for 42.13%, while the area with a negative correlation accounts for 37.23%. This shows that when the groundwater depth in Ejina is too shallow or too deep, the area where NDVI is related to groundwater depth increases, indicating that groundwater has a stronger impact on vegetation coverage.

Through the above analysis, there is a certain degree of correlation between groundwater depth and vegetation index in Ejina, which may be positive, negative or no obvious correlation, depending on various factors, especially the influence of soil water and salt conditions [26,27]. Ejina belongs to the most arid area in China, with little rainfall and strong evaporation. When the groundwater depth is shallow, the salt in the groundwater is more likely to migrate to the soil surface with the evaporation of water, causing a large accumulation of salt on the soil surface, deepening the degree of salt stress faced by plants, and being unfavorable to the growth of shallow-rooted and salt-intolerant plants. With the increase in groundwater depth, the degree of salinization on the soil surface will decrease due to the weakening of evaporation, but the soil water content will significantly decrease, which makes it difficult for plant roots to obtain enough water, thus affecting the growth and survival of vegetation [28]. In summary, there is both positive and negative correlation between groundwater depth and vegetation index in arid areas, and this complex relationship is affected by various factors, including the dependence of plants on water, the adaptability of the soil environment, and the ecological strategies of vegetation itself.

4. Discussion

4.1. Driving Factors of Groundwater Depth

The groundwater depth in Ejina varies greatly in spatial scale, and after field investigation, the shallow groundwater dynamics in the area can be divided into four types: hydrological type, underground runoff–evaporation type, underground runoff–evapotranspiration type, and artificial interference type [29,30]. The hydrological type is mainly distributed in the riverbank zone, with riverbank seepage recharge and vegetation transpiration as the main sources. Groundwater depth experiences rapid fluctuations during the initial river flow period, but then declines steadily under the influence of riverbank vegetation transpiration during the summer. The underground runoff–evaporation type mainly exists in the desert-gobi belt, where the groundwater depth shows a weak seasonal fluctuation under the dual influence of lateral recharge and submarine evaporation. The underground runoff–evapotranspiration type is mainly distributed in oasis areas, where groundwater levels show marked seasonal fluctuations under the influence of evapotranspiration in groundwater-dependent ecosystems and lateral groundwater recharge. The artificial interference type is present within the artificial oasis, where groundwater levels, affected by groundwater extraction and artificial recharge, fluctuate more dramatically during irrigation periods. Therefore, the dynamic changes in Ejina’s groundwater depth are mainly influenced by surface runoff infiltration recharge, regional groundwater flow, agricultural irrigation, and shallow groundwater evaporation.

Due to the influence of the special soil texture of Ejina, Infiltration recharge by surface runoff is the main way to renew shallow groundwater in the bank area of the Ejina River. Different soil textures have different capillary water properties, which determine soil water holding capacity and infiltration [25]. According to field investigations, the riverbed of the Ejina River is mainly composed of sandy sediments, with a good infiltration property. The downstream areas such as Gurinai and Juyan Lake are mostly composed of lacustrine deposits, and the soil texture is mainly clay and sand. The infiltration of groundwater into the riverbed is significantly linearly related to the river discharge. Meanwhile, the Ejina River is a typical intermittent and seasonal river, and the time when water is diverted from the middle reaches of the Heihe River to the lower reaches is mainly concentrated in spring and autumn. The river’s water flow is easily affected by human intervention upstream and downstream, and it has strong temporal and spatial variability [31]. Thus, in the mode of artificial water regulation, the upper reaches of the river receive long-term surface water recharge, with sufficient water volume and shallow water depth. However, as the river flows from the upstream to the downstream, the amount and duration of surface water recharge gradually decrease, leading to a deeper and more fluctuating groundwater depth, which makes the groundwater level easily affected by other variables.

Shallow groundwater evaporation is the main way for groundwater discharge [32,33]. Most of the area in Ejina is a desert-gobi zone, and the evaporation intensity is mainly affected by the lithology and structure of the unsaturated zone and the depth of the groundwater depth. The smaller the depth of the groundwater, the greater the evaporation intensity (Figure 7). Groundwater evaporation has a greater impact on the groundwater depth of the ecosystem dependent on groundwater. The evaporation of groundwater-dependent vegetation not only causes seasonal decline in shallow groundwater level, but also causes daily fluctuations in groundwater level, which is manifested as the groundwater level begins to decline during the day under the evaporation of vegetation, and the groundwater level slowly rises at night when evaporation is basically stopped or very weak, under the influence of lateral groundwater recharge.

4.2. Relationship Between Vegetation Coverage and Groundwater Depth

In arid and semi-arid areas, water is an important factor in maintaining the growth of vegetation and the stability of ecosystems [34]. Ejina is a typical arid area in western China, and there is a complex interplay between vegetation coverage and groundwater depth, which plays a crucial role in the stability and restoration of vegetation ecosystems [34]. Ejina has low rainfall and high evaporation, and its surface rivers are easily affected by human intervention in the middle reaches of the Heihe River. Groundwater becomes an important and stable source of water for plant growth in the area, and groundwater level changes directly affect soil moisture content, which in turn influences the species composition, quantity, growth status, and vegetation coverage of plants.

Through field investigation, it was found that in areas with shallow groundwater depth, the vegetation types are more diverse, including various shrubs and herbaceous plants, while in areas with deep groundwater depth, the vegetation types are relatively simple, mainly composed of drought-resistant shrubs and herbaceous plants. The different vegetation types and distribution have a direct impact on vegetation coverage [35]. On the other hand, the vegetation coverage is higher in areas with shallow groundwater depth because the vegetation root system can more easily absorb water, which is conducive to vegetation growth [36]. In contrast, the vegetation coverage is lower in areas with deep groundwater depth because the vegetation root system cannot absorb enough water, limiting vegetation growth. Through calculation, it was found that the NDVI value of Ejina is related to the depth of groundwater to a certain extent, that is, the vegetation development degree is obviously higher in the area with groundwater depth of 2–6 m, and the NDVI value decreases obviously when the groundwater depth exceeds 4.8 m. Meanwhile, it is shown that there is no obvious correlation between vegetation index and groundwater depth in some areas, which may be due to the fact that vegetation growth is affected by a variety of factors such as water, soil, salt, temperature, and human activities [6,35,37,38,39]. For a certain period of time, When the change in groundwater level exceeds a certain range, other influencing factors may become the main controlling factors, and the formed vegetation growth pattern will not have a significant correlation with the groundwater depth.

5. Conclusions

This study takes the Ejina region, an arid area in northwest China, as the research object, using MOD13Q1-NDVI data from 2018 to 2023, combined with groundwater depth data from the same period, to explore the relationship between the spatial–temporal changes of groundwater depth and vegetation coverage. The main conclusions are as follows:

(1) The vegetation coverage in the Ejina region is generally low, with bare land and low vegetation coverage as the main features. The vegetation mainly extends along the river direction and is distributed in a strip-like pattern. Between 6a (2018~2023), the overall trend of vegetation coverage is fluctuating, with an increase followed by a decrease. However, the inter-annual trends of each vegetation coverage class are different, and the vegetation coverage is easily affected by the natural environment and human activities, reflecting the region’s strong ecological vulnerability.

(2) The groundwater depth in Ejina has obvious spatial and temporal heterogeneity characteristics. In terms of spatial distribution, the water level in the west river area is generally lower than that in the east river area. Along the river flow direction, the groundwater depth in both the east river and west river shows a trend of first increasing and then decreasing. From the annual variation, the groundwater depth in the study area shows a distinct seasonal change characteristic, with the lowest level usually observed in March and the highest in August. From the inter-annual scale, the groundwater depth shows a trend of first decreasing and then stabilizing. The fluctuation of water level in the upper reaches of the river is smaller, while the fluctuation in the lower reaches is larger.

(3) Water conditions are an important factor affecting vegetation growth and coverage in Ejina. The high values of NDVI are mainly distributed in strip-like patterns along the east-west riverbanks. When the depth of groundwater is between 4–6 m, vegetation grows well and its coverage is high in these areas. However, vegetation coverage generally deteriorates when the depth of groundwater exceeds 8 m. There was an optimal groundwater depth interval for vegetation growth in the Ejina area, and when the groundwater depth was too deep or too shallow, the vegetation cover showed a decreasing trend.

(4) There is a certain correlation between groundwater depth in Ejina and vegetation index. When the groundwater depth is too deep or too shallow, the correlation between NDVI and groundwater depth will be slightly enhanced, and the negative correlation between NDVI and groundwater depth is the largest when the groundwater depth is between 4.8–8.5 m.