Nothing Special   »   [go: up one dir, main page]
Next Article in Journal
Navigation of a Team of UAVs for Covert Video Sensing of a Target Moving on an Uneven Terrain
Previous Article in Journal
Mapping Polylepis Forest Using Sentinel, PlanetScope Images, and Topographical Features with Machine Learning
Previous Article in Special Issue
Realization of Integrated Regional Ecological Management Based on Ecosystem Service Supply and Demand Flow Networks: An Example from a Dominant Mineral Resources Development Area
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Remote Sensing
Volume 16
Issue 22
10.3390/rs16224272
remotesensing-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Ecosystem Stability in the Ugan–Kuqa River Basin, Xinjiang, China: Investigation of Spatial and Temporal Dynamics and Driving Forces

by
Ting Zhou
1,
Peiyue Zhu
1,
Rongjin Yang
2,†,
Yilin Sun
1,
Meiying Sun
2,
Le Zhang
2 and
Xiuhong Li
1,*,†
1
State Key Laboratory of Remote Sensing Science, Faculty of Geographical Science, Beijing Normal University, Beijing 100875, China
2
State Key Laboratory of Environmental Criteria and Risk Assessment, Chinese Research Academy of Environmental Sciences, No. 8, Da Yang Fang, An Wai, Chao Yang District, Beijing 100012, China
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Remote Sens. 2024, 16(22), 4272; https://doi.org/10.3390/rs16224272
Submission received: 28 September 2024 / Revised: 6 November 2024 / Accepted: 14 November 2024 / Published: 16 November 2024
(This article belongs to the Special Issue Remote Sensing Based Monitoring of Terrestrial Ecosystem Service Bundles, Trade-Offs and Synergies (Second Edition))
Browse Figures
Figure 1
<p>Schematic diagram of the research region. (<b>a</b>) Geographical location of the study area; (<b>b</b>) Remote sensing image map of the study area; (<b>c</b>) Land use types in the study area; (<b>d</b>) Area percentage of different land use types in the study area.</p> "> Figure 2
<p>Conceptual meta-model of the ecological stability drivers. (<b>a</b>) The overall relationship between stability and climate, vegetation, human activities, and fragmentation. (<b>b</b>) The specific modeling relationship between the drivers of ecosystem stability. The pathways illustrate the interconnections among Pre, Tem, ET, LUCC, GDP, NDVI, ED, PD, LSI, and ecosystem resistance and recovery. Unidirectional arrows denote causation.</p> "> Figure 3
<p>Temporal changes in resistance and recovery.</p> "> Figure 4
<p>Spatial trends in ecosystem resistance and recovery. (<b>a</b>) Spatial trend in resistance. (<b>b</b>) Spatial trend in recovery.</p> "> Figure 5
<p>Scatter plot of resistance–recovery trade-off distributions. The data for both resistance and recovery were standardized within the range of 0 to 1, with the intersection where both factors equate to 0 signifying the origin of the coordinates. The delineation was set at 0.5; values below 0.5 for both resistance and recovery indicate a state of low resistance–low recovery, whereas values surpassing 0.5 indicate high resistance–high recovery. Additionally, resistance above 0.5 and recovery below 0.5 indicate high resistance–low recovery, whereas resistance below 0.5 and recovery above 0.5 suggest low resistance–high recovery.</p> "> Figure 6
<p>Drivers of ecosystem stability, illustrating the simulated effects of climate, vegetation, human ac-tivities, and habitat fragmentation on ecosystem stability. The colors of the arrows represent the degree of significance of the relationships, with dashed arrows indicating nonsignificant correlations. Climatic factors are denoted by blue boxes, vegetation variables are denoted by green boxes, human activity variables are denoted by pink boxes, habitat fragmentation variables are denoted by gray boxes, and stability variables are denoted by yellow boxes.</p> "> Figure 7
<p>Impact of drivers on ecosystem stability. (<b>a</b>) Presents a stacked plot showing the proportions of the total effects of these factors on resistance and recovery, with cells representing negative impacts. (<b>b</b>) Showcases a stacked plot illustrating the percentages of direct versus indirect influences of the variables on resistance, with cells denoting negative impacts. Finally, (<b>c</b>) presents a stacked plot displaying the percentages of direct versus indirect influences of the variables on recovery, with cells indicating a negative impact.</p> ">
Versions Notes

Abstract

:
Ecosystem stability plays a pivotal role in safeguarding the enduring well-being of both the natural world and human society. This work explores the uncertainty surrounding changes in ecosystem stability and their response mechanisms at localized scales, focusing on the Ugan–Kuqa River Basin in Xinjiang, China. Based on remote sensing data and spatial lag modeling (SLM), we evaluated the spatial and temporal dynamics of the basin’s stability from 2001 to 2020. Additionally, structural equation modeling (SEM) was employed to assess the impacts of climate conditions, human activities, and habitat fragmentation on ecosystem stability. The results of the study indicated that the basin ecosystem stability tended to increase in the temporal dimension, and that the spatial distribution was greater in the north than in the south. In addition, the trade-off between resistance and recovery in the watershed decreased, with a considerable increase in high-resistance–high-recovery zones. Climate warming and increased humidity have emerged as the predominant factors driving the watershed ecosystem stability.

1. Introduction

The intricate interplay among climate change, escalating water demand, and anthropogenic activities [1] has resulted in a notable increase in the frequency and severity of extreme climatic events such as droughts and heatwaves [2]. These phenomena have instigated a reconfiguration of water reservoirs across diverse geographical scales [3] and are poised to accentuate the discrepant spatial and temporal dispersal of water resources. Drought, characterized primarily by a shortage of water, is a defining climatic phenomenon [4]. Prolonged drought periods heighten the risk of carbon scarcity inside ecosystems [5], produce hydrological breakdowns [6], lower vegetation productivity [7], and hinder the physiological processes within plant life [8]. Together, these repercussions affect the configuration and activities of a dry region’s ecosystem, impacting the capacity to maintain ecosystem equilibrium. Additionally, the vegetation’s responses to drought create feedback loops within the ecosystem, thereby influencing the climate in various ways.
Ecosystems, characterized by complex interconnections and interactions among the social, economic, and natural elements, are essential for life and the development of humans. Stability, a fundamental facet of ecosystem structure and function, plays a crucial role in determining the rise and fall of ecosystems [9]. Ecosystem stability refers to the overall ability of a system to remain within a specific region of attraction while preserving its function and structure in the face of disruptions [10]. There are various definitions and evaluation techniques for assessing the components of ecosystem stability [11]. In this study, we focused on two fundamental indicators of ecological stability: resistance and recovery [12]. Resistance refers to an ecosystem’s ability to withstand external disturbances while maintaining its structure and function [13]; it quantifies the impact of stress on the ecosystem. Recovery, however, denotes the capacity of an ecosystem to recover after a disturbance [14,15], reflecting the swiftness of the ecosystem recovery [16].
Currently, there are numerous studies on ecosystem stability across various scales [17,18,19]. Advancements in remote sensing technologies have paved the way for the investigation of resistance and recovery over extensive temporal spans and vast spatial extents, emerging as a focal point within the realms of geography and ecology. Ruppert et al. [20] assessed the stability of ecosystems in over thirty arid zones worldwide and reported that drought intensity exhibited a linear negative correlation with ecosystem resistance. They pointed out that, particularly under grazing conditions, drought resistance was significantly higher in herbaceous systems dominated by annuals compared to those dominated by perennials. Yao et al. [21] reported that semi-arid zones in China presented the weakest resistance. They reported that the overall resistance gradually increased from arid to semi-humid zones, highlighting a trade-off between recovery and resistance. Specifically, they noted that, compared with forested areas, grasslands had greater recovery but lower resistance. Shao et al. [22] investigated the drought resistance of nine ecosystems via data from 118 flux tower stations around the globe. Their findings revealed that drought resistance was greater in evergreen forests but lower in agricultural lands and shrublands.
The occurrence of drought can lead to changes in the ecosystem structure and function [23]. However, the extent of these alterations and the rate of recovery may depend on the environmental conditions. For example, water, which serves as the primary driver of vegetation productivity within arid-zone ecosystems [24], exerts varying degrees of influence on the stability of these environments through rainfall fluctuations, temperature shifts, and evapotranspiration intensity. In addition to the climate conditions, the intensity, frequency, and spatial extent of human activities can also modify ecosystem components and impact stability [16]. Land use changes fragment natural ecosystems, and habitat fragmentation—one of the most significant factors affecting ecosystems—directly influences the intraecosystem supply, flow, and interactions [25]. Consequently, a combination of factors including climate change, human activity, and habitat fragmentation collectively conspire to engender shifts in ecological stability.
While numerous studies have focused on assessing ecosystem stability at a global scale, these endeavors serve as significant pillars for global evaluations of ecosystem stability and the factors influencing its response to vegetation dynamics. Nonetheless, alongside the repercussions of global climate change, the stability of vegetation is markedly influenced by local conditions [19]. Hence, the meticulous tracking of shifts in ecosystem stability and its responses to local environmental alterations at smaller regional scales has emerged as paramount for sustainable management practices and the effective implementation of vegetation conservation initiatives in arid ecosystems across these localized settings. Moreover, ecological stability is affected by various interacting forces. Research should not only examine the effects of a single driver but also explore the interrelationships among different factors to fully understand the dynamics of ecosystem stability.
Consequently, this research focused on the Ugan–Kuqa River Basin, a representative region of the ‘mountain–oasis–desert’ complex system, and conducted comprehensive studies using remote sensing data. The primary objectives of this research are (1) to quantify the spatial and temporal dynamics of ecosystem stability within the basin from 2001 to 2020; (2) to analyze the trade-off between ecosystem resistance and recovery; and (3) to employ structural equation modeling (SEM) to elucidate the interrelations among climatic factors, human activities, habitat fragmentation, and other stressors. Specifically, this study examines the effects of climatic factors such as precipitation (Pre), temperature (Tmp), and evapotranspiration (ET); anthropogenic influences such as land use changes (LUCC) and gross domestic product (GDP); and metrics of habitat fragmentation such as edge density (ED), patch density (PD), and the landscape shape index (LSI) on the significance of ecosystem resistance and recovery.

2. Materials and Methods

2.1. Research Region

The Ugan–Kuqa River Basin is located in the northern section of the Tarim Basin and the southern foothills of the Tianshan Mountains in the Xinjiang Uygur Autonomous Region [26]. Geographically, it spans from longitudes 82°07′ to 83°41′E and latitudes 40°57′ to 41°55′N. The landscape features significant elevation changes, with higher terrain in the north and south (Figure 1). The region experiences a quintessential temperate continental climate, with an average annual precipitation of 74 mm and an approximate average annual temperature of 11.1 °C, rendering it an arid and exceedingly dry environment.
Land use within the watershed is relatively straightforward, although the ecosystem is highly fragmented and typical of oasis agriculture. Agricultural land constitutes 53.4% of the basin’s area, with crop cultivation predominantly reliant on meltwater from alpine ice and snow, as well as river irrigation. However, factors such as climate warming, accelerated glacial melting, and human activities pose significant threats to the stability of the ecosystem in the Ugan–Kuqa River Basin.

2.2. Data Sources

2.2.1. SPEI Data

The standardized precipitation evapotranspiration index (SPEI) is a widely adopted drought metric on a global scale. This index plays a pivotal role in the surveillance and prediction of meteorological, hydrological, agricultural, and socioeconomic droughts. For this investigation, we sourced the SPEI data from a dataset with a spatial resolution of 1 km and a 12-month temporal scale, as reported by Xia et al. [27]. To meet the specific requirements of our study, we resampled the watershed’s spatial resolution to 500 m via nearest neighbor interpolation.

2.2.2. GPP Data

The functioning of ecosystems in arid zones heavily depends on the ecological ser-vices provided by plants. Gross primary productivity (GPP) serves as a vital indicator of the capacity of ecosystems to sustain themselves and acts as a source of energy for the ecosystem. To assess the productive potential of ecosystems, this study utilized annual GPP data (MOD17A3HGF.006, https://lpdaac.usgs.gov, accessed on 3 May 2024) from 2001 to 2020, with a spatial resolution of 500 m, as provided by MODIS. The GPP dataset is derived from the sum of all 8-day net photosynthesis (PSN) products for each year, where the PSN value represents the difference between total primary production (GPP) and maintenance respiration (MR) [28].

2.2.3. Climate Data

The meteorological data utilized in this research were collected from the Chinese 1-km-resolution monthly precipitation dataset and the Chinese 1-km-resolution monthly mean temperature dataset, which were obtained from the spatiotemporal tripolar environmental big data platform (http://poles.tpdc.ac.cn, accessed on 27 May 2024). Additionally, the ET data were derived from the annual ET dataset with an MODIS resolution of 500 m for the years 2001 to 2020 (MYD16A3GF, https://lpdaac.usgs.gov).

2.2.4. Huaman Activities Data

To assess the impact of human activities on ecosystem stability, this study selected LUCC data and GDP statistics as indicators. The LUCC data were obtained from MODIS land cover types (MCD12Q1, https://lpdaac.usgs.gov) at a resolution of 500 m for the period from 2001 to 2020. The MCD12Q1 dataset includes 17 land cover classes, including 11 natural vegetation classes, 3 developed and mosaic land classes, and 3 nonvegetated land classes. These classes are derived via supervised classification of MODIS Terra and Aqua reflectance data [29].
GDP statistics were employed to characterize human activities in this investigation. The categorization of GDP statistics was obtained from the 1-km-resolution gross domestic product data, which were calibrated on the basis of worldwide nighttime light data for the years 1992 to 2019, as provided by Chen et al. [30]. To align with the requirements of this study, we resampled the spatial resolution of the watershed’s GDP data to 500 m via nearest neighbor interpolation.

2.2.5. Fragmentation Data

Land use changes fragment natural ecosystems [31]. To quantify the degree of eco-system fragmentation in the study area, we utilized land use data from 2001 to 2020 to generate three landscape fragmentation indices: ED, PD, and LSI. ED refers to the extent of the boundaries surrounding patches within a unit area, indicating the complexity of landscape types and assessing the edge effects among various land use patterns within the ecosystem [32]. PD denotes the quantity of patches per unit area, reflecting landscape diversity; a higher value signifies greater fragmentation of landscape patches [33]. LSI measures the regularity of patch shapes, where a more complex shape corresponds to a higher index value.

2.3. Research Methods

2.3.1. Data Preprocessing

In this study, to ensure the integration of multi-resolution data and maintain consistency in the analysis, we preprocessed the multi-source remote sensing data. First, we applied projective transformation to the raster data to ensure that all datasets had consistent coordinate systems, thereby reducing errors and enhancing analysis accuracy. Second, for data with varying spatial resolutions (e.g., 1 km for SPEI data, 500 m for MODIS GPP data, and 30 m for land use data), we utilized the nearest neighbor method to populate the nearest pixel values from the original images into a new image. All datasets were then resampled to a uniform resolution of 500 m to facilitate spatial and temporal analyses at the same scale.

2.3.2. Drought Threshold

The SPEI, a widely employed indicator for detecting drought occurrence, operates on multiple timescales and can effectively reflect moisture conditions across various intervals [34]. In this study, we selected the SPEI with a 12-month timescale, which captures the interannual variability in drought conditions, to analyze drought episodes. The SPEI quantifies the degree of deviation from both dry and wet conditions, where a positive SPEI value signifies wetness and a negative value indicates drought. For the purposes of this research, a drought event is defined according to the national standard meteorological drought level of the People’s Republic of China (Table 1), with drought classified as SPEI ≤ −0.5 [35].

2.3.3. Components of Stability

The dynamics of ecosystem stability under the impact of drought disruption resemble those of springs [21]. Like springs, which compress when disturbed and gradually recover afterwards, ecosystem stability compresses during drought disturbances and progressively recovers once the disturbance subsides. Therefore, we employed two indicators to evaluate the stability of ecosystems in the face of drought perturbation: the capacity of ecosystems to withstand and endure disturbances (referred to as resistance) and the ability of ecosystems to rebound and restore themselves after the disturbance has ceased (known as recovery). In this study, we measured the resistance of grid cells via the ratio of drought-induced losses to undisturbed losses, as described in Equation (1). A lower magnitude of vegetation loss indicates a greater capacity of the ecosystem to resist drought. In contrast to resistance, which focuses on quantifying vegetation loss, we quantified the recovery of grid cells via Equation (2) by comparing the vegetation productivity after the drought event with that prior to the event. A faster rate of vegetation recovery within the same timeframe implies a greater restorative capacity of the ecosystem.
R e s i s t a n c e = G P P d r o u g h t − ∅ G P P ∅ G P P
In Equation (1), G P P d r o u g h t represents the GPP for an annual SPEI ≤ −0.5, whereas ∅ G P P represents the GPP for nondrought conditions, which is defined as an annual SPEI greater than −0.5.
R e c o v e r y = G P P p o s t − d r o u g h t G P P p r e − d r o u g h t
In Equation (2), G P P p o s t − d r o u g h t refers to the GPP during the year following a specific drought event, whereas G P P p r e − d r o u g h t represents the GPP before that particular drought occurrence.

2.3.4. Trade-Off Between Resistance and Recovery

The trade-off between resistance and recovery in ecosystems reflects a dynamic balance between high resistance and low recovery and vice versa. This concept aims to prevent ecosystems from simultaneously exhibiting low resistance and low recovery, which is crucial for withstanding drought conditions [36]. Simultaneously, the ecological data representing resistance and recovery often manifest distinctive spatial patterns [37]. To elucidate these patterns and understand the spatial dynamics of the resistance–recovery trade-off, we employed a spatial lag model (SLM). With reference to Equation (3), where ρ denotes the spatial autoregressive coefficient (Rho), a positive value signifies a positive relationship between the recovery of a given region and that of its neighboring regions. Moreover, β represents the coefficient associated with the resistance–recovery continuum; specifically, a positive value suggests mitigation of the resistance–recovery trade-off, whereas a negative value implies amplification of this trade-off.
y i = ρ ∑ j w i j y j + X i β + ε i
In Equation (3), y i represents the recovery (dependent variable), whereas ρ represents the spatial autoregressive coefficient (Rho). The term w i j denotes the spatial weight matrix element, indicating the spatial association between regions i and j . X i represents the resistance (explanatory variable), with β being the coefficient of this explanatory variable, and ε i represents the error term.

2.3.5. Analysis of Driving Mechanisms

To investigate the impacts of climate change, human activities, and habitat fragmentation on ecosystem stability, as well as the interactions among these influencing factors, we employed SEM to assess the complex relationships among the multivariate variables. SEM offers a comprehensive approach that combines the strengths of deterministic factor analysis and path analysis [38]. It not only identifies potential variables (factors) and their associations with observed variables [39] but also visually represents causal relationships among variables through path diagrams [40].
SEM quantifies causal relationships and path strengths by constructing and verifying the covariance structure among dependent variables, latent variables, and observed variables. To examine the effects of various influencing factors on the abstract concept of eco-system stability, we selected resistance and recovery as exterior variables representing ecosystem stability. We verified their relationships with other variables, such as Tem, Pre, ET, LUCC, GDP, ED, PD, and LSI. By estimating the coefficients of a linear model that captures the covariance between the exterior variables and the influencing factors, we can ascertain the validity of the relationships between the exterior variables, thereby indicating the causal links between the variables. The specific modeling technique is illustrated in Figure 2.
Simultaneously, we used the SEM fit indicators (e.g., χ2/d.f, RMSEA, RMR, GFI, and AGFI) to iteratively adjust the model, ensuring it achieved optimal explanatory power and goodness-of-fit. The χ2/d.f ratio was utilized to assess the overall fit of the model, with a value less than 3 generally indicating a well-fitted model. RMSEA evaluates the fit be-tween the model structure and the data; lower values indicate better model fit, with values below 0.05 signifying a good fit. RMR measures the residuals between predicted model values and actual observed values, with model residuals considered acceptable when RMR is less than 0.05. GFI characterizes the overall goodness-of-fit of the model, with values above 0.9 typically indicating a better fit. AGFI (Adjusted Goodness-of-Fit Index) introduces a degree-of-freedom adjustment factor based on GFI, serving as a crucial factor for model adjustment. An AGFI greater than 0.9 indicates a good model fit, helping to control model complexity.
To enhance model stability and facilitate the interpretation and comparison of parameter estimates, all variables were standardized using a z-score. This method converts variables to a dimensionless form by subtracting the mean from the original value and di-viding by the standard deviation, effectively removing the effects of scaling due to differences in measurement scales. This ensures that the impacts of climatic, economic, and habitat fragmentation factors can be compared on the same scale. After the standardization process, each variable has a mean of 0 and a standard deviation of 1, thus enabling more interpretable path coefficients in the SEM. This approach also reduces computational errors arising from differences in data scales, leading to more stable and informative model results.

3. Results

3.1. Temporal and Geographical Differences in Resistance and Recovery

Regarding the temporal scale, the ecosystem resistance and recovery demonstrated a general increasing trends (Figure 3). Specifically, resistance exhibited a more significant increase than recovery did, and there was even an increase in the watershed vegetation production following a prior drought episode (resistance > 0). Although the watershed recovery also increased, drought events continued to negatively impact plant production (recovery < 0). The correlation coefficients between resistance, recovery, and SPEI were 0.19 and −0.05, respectively. While these associations were not strong, they did indicate a trend relating resistance and recovery to drought severity and highlighted a trade-off relationship between the two.
The geographical distribution of resistance revealed that the largest areas of resistance were located in the northern half of the basin, whereas the lowest levels were found in the southern section. Over time, the area characterized by heightened resistance gradually shifted from the northern sector of the basin towards the southern sector (Figure 4a). Conversely, the regional dispersion of recovery displayed more variability, featuring localized hotspot effects (Figure 4b). While resistance demonstrated distinct patterns across both the temporal and geographical dimensions, the geographic framework of recovery proved to be more intricate and diverse.
The intricate geographical distribution of recovery was influenced by numerous factors. At the spatial level, the spatial autoregressive effect of recovery diminished from 0.58 (p < 0.01) to −0.16 (p < 0.1), indicating a transition from a positive to a negative correlation (Figure 4b). This suggests that those areas exhibiting greater recovery initially exerted a beneficial influence on the recovery of adjacent regions; however, this effect gradually transformed into a dampening effect. This shift may be linked to environmental changes, human activities, or other external influences.

3.2. The Resistance–Recovery Trade-Off

The connection between resistance and recovery has deteriorated. According to the spatial lag model, the coefficient reflecting the influence of resistance on recovery shifted from positive to negative over the 20-year period, moving from −0.23 (p < 0.01) to 0.18 (p < 0.01). This transition signifies a gradual departure from a trade-off relationship to the emergence of a pattern of high resistance–high recovery or low resistance–low recovery within the resistance–recovery dynamic. The long-term trend in the distribution of resistance–recovery further validated the decline in this trade-off connection, particularly the decrease in the regional presence of high resistance–low recovery (Figure 5). In general, there was a consistent improvement in the stability of the watershed ecosystem. Generally, ecosystems characterized by high resistance and high recovery are often perceived as more stable. Within the watershed, those areas exhibiting poor resistance and low recovery experienced a significant decline, whereas the regions marked with high resistance and high recovery experienced steady expansion (Figure 5). Notably, changes in recovery occurred before changes in resistance. Specifically, recovery underwent initial changes over time, with a noticeable increase in the region characterized by low resistance and high recovery, gradually progressing towards the upper left quadrant of the coordinate system (i.e., in the direction of increasing recovery).

3.3. Drivers of Ecosystem Stability

Figure 6 illustrates the final SEM structure developed. The goodness-of-fit metrics for the final model are as follows: RMSEA = 0.036, GFI = 0.983, and AGFI = 0.946 (Table 2). These metrics serve as indicators affirming that the model aptly captures the intricate interplay among factors and the multidimensional linkages connecting resistance and recovery with their underlying determinants.
Climatic conditions serve as pivotal determinants of ecosystem stability. ET and Pre emerged as the foremost variables influencing variations in ecosystem resistance, with Tmp and ED also exerting considerable effects (Figure 7a). This indicates that increased Pre and ET bolster watershed resistance, whereas elevated Tmp may undermine it. With respect to ecosystem recovery, ET stood out as the most significant factor, closely followed by Pre (Figure 7a), underscoring the critical importance of water availability for ecosystem recovery in arid regions.
The significance of habitat fragmentation as a driver of ecosystem stability varied between resistance and recovery. Compared with PD and the LSI, ED had a notably greater impact on resistance (Figure 7a), with a positive direct influence (Figure 7b), suggesting that, to some degree, an increase in boundary density favors an increase in ecosystem resistance. Conversely, concerning ecosystem recovery, the influence of the LSI carried more weight (Figure 7a), resulting in a negative direct effect (Figure 7c), suggesting that the ecosystem recovery may decrease proportionally as the geometric intricacy of landscape patches intensifies.
The relative contribution of human activities to ecological stability was overshadowed by the impacts of climate factors and habitat fragmentation. LUCC and GDP had positive impacts on ecosystem recovery, indicating that land management and economic growth have, to some extent, enhanced ecosystem resistance to drought. However, GDP also had a negative influence on recovery, revealing that the adverse effects of the environmental constraints resulting from economic expansion can diminish the capacity of ecosystems to recover to their previous state following disturbances.

4. Discussion

Drought poses a significant threat to ecosystem stability in arid regions, profoundly affecting their structure and function. Remote sensing technology, which serves as an instrument for earth observation, offers a distinctive perspective and method for exploring and overseeing ecosystems within arid regions [41,42]. In this work, we assessed the spatial and temporal dynamics of the ecosystem stability in the Ugan–Kuqa River Basin from 2001 to 2020 via a remotely sensed GPP dataset. We also identified the impacts of the climatic factors, human activities, and habitat fragmentation on the ecosystem resistance and recovery through SEM. The determinants of ecosystem stability are intricate and multilayered [43,44]. Even though our model can discriminate the comparative significance of each factor with respect to resistance and recovery, it has limitations in precisely unravelling the operational dynamics and interactive behaviors among these diverse factors. Further studies and analyses are needed to elucidate the mechanisms underlying these interactions.

4.1. Spatial and Temporal Dynamics of Ecosystem Stability in Watershed

To comprehensively scrutinize the geographical and temporal dynamics of water-shed ecosystem stability, we linked the trends of watershed ecosystem stability alterations with the trade-off between resistance and recovery. Our findings revealed that improvements in watershed ecosystem stability were largely associated with a reduction in areas characterized by low resistance and low recovery, as well as an increase in regions with high resistance and high recovery (Figure 5). This trend signifies an increased capacity within the ecosystem to withstand disturbances and recover to its previous state, thereby contributing to enhanced overall stability. In contrast to the findings of Gazol et al. [45], which suggested that the coniferous forests in the arid and semi-arid regions of the southwestern United States and southern Europe exhibited greater drought recovery, our results in the Ugan–Kuqa River Basin demonstrated a significantly higher level of resistance than recovery, with resistance even increasing during drought events (Figure 3).
Upon further evaluation of the land use statistics in the basin, we found that cultivated land accounts for more than 50% of the total area, and this proportion is increasing. Conversely, barren land areas, including gobi and deserts, constitute more than 20% of the watershed but contribute minimally to watershed productivity. Consequently, agriculture, particularly the cultivation of annual crops such as wheat, maize, cotton, and oilseed rape, represents the primary source of production within the watershed. In contrast to the drought-resistant nature of perennial trees, annual vegetation has heightened sensitivity to short-term water availability [46,47], thereby enabling such vegetation to utilize transient resources more efficiently and exhibit greater resistance to drought stress [48,49,50]. Nevertheless, this rapid growth strategy also hints at potential inadequacies in ensuring long-term ecosystem stability.
The internal stability of the watershed revealed a conspicuous gradient, marked by elevated values in the northern region and diminished values in the southern region (Figure 4), closely intertwined with the distinctive geographical position of the Ugan–Kuqa River. Surrounding the Tianshan Mountains to the north, the Taklimakan Desert to the south, and the Gobi Desert to the east and west, the basin forms a quintessential closed oasis [51]. The topography of the basin features elevated terrain in the north and de-pressed terrain in the south, leading to a disparate distribution of water resources, thus fostering substantially greater vegetation coverage and species diversity in the northern expanse region than in the southern region. A growing body of research underscores the positive impact of biodiversity on ecological stability [52,53,54].
Furthermore, our investigations revealed a decreasing trend in the trade-off between resistance and recovery within the watershed, which aligns with the previous research findings. Yao et al. [36] assessed the global spatiotemporal dynamics of resistance and recovery by utilizing remotely sensed vegetation indices and noted a noteworthy reduction in the interplay between these two crucial factors.

4.2. Climate Warming and Wetting Are Major Contributors to the Enhanced Stability of Watershed Ecosystems

Both the increase in overall watershed stability (Figure 3) and the decrease in the resistance–recovery trade-off, accompanied by an increase in the number of high-resistance–high-recovery zones (Figure 5), indicate a paradigm shift in the coping mechanisms of watershed ecosystems in the face of drought. This transformation is driven primarily by the prevailing trend of “warming and wetting” in the arid regions of northwest China [55,56]. Specifically, increased precipitation has led to an expansion of the vegetation cover, mitigating the severity of drought events [57,58]. Despite the negative impact of the rising temperatures on the watershed resistance and recovery (Figure 7a), with higher temperatures directly exacerbating the drought intensity, the warming phenomenon has also accelerated the melting of glaciers and snow in the elevated mountainous areas within the watershed, such as the Tianshan range. In Xinjiang, glacier meltwater runoff contributes approximately 25% of the average multi-year runoff, and the primary source of water in the Ugan–Kuqa River Basin—a typical arid-zone oasis—is the meltwater from ice and snow in the Tianshan Mountains. In recent years, the climate warming and increased humidity in northwestern China have further accelerated the glacier snowmelt rate on the northern slopes of the Tianshan Mountains. For example, the average runoff depth of the glacial meltwater has increased from 508.4 mm in 1985 to 936.6 mm today, representing an 84% rise. This melting of glacial snow has supplied more water to the watershed, contributing to an increase in vegetation greenness [59]. Additionally, an increase in atmospheric carbon dioxide levels has a fertilization effect, contributing to the increase in greenness [60]. These positive impacts may help to balance, to some degree, the negative consequences of rising temperatures.
Moreover, we observed that ET is the most critical component impacting the resistance and recovery of the watershed, functioning in distinct ways for each. ET directly contributed to ecosystem resistance (Figure 7b), whereas its impact on recovery was manifested primarily through indirect pathways (Figure 7c). As a critical indicator of regional hydrothermal conditions, evapotranspiration originates partly from soil evaporation and partly from plant transpiration [61]. Although there are currently no conclusive data concerning whether climatic warming and humidification result in an increase or decrease in evapotranspiration [62,63,64], it is apparent that the expansion of vegetation cover under warming and humidification trends represents the primary factor driving the increase in terrestrial evapotranspiration [65,66]. The increase in watershed evapotranspiration indicates that more water is available for plant growth, which not only enhances vegetation cover but also increases watershed productivity, thereby directly improving the tolerance of the ecosystem to drought. In contrast, recovery, denoting an ecosystem’s ability to revert to its former state following a disturbance, is more sensitive to disruptive variables than resistance. Evapotranspiration influences ecosystem recovery indirectly by altering the soil moisture, plant growth processes, and microbial activity.

4.3. The Dual Effects of Anthropogenic Activities and Fragmentation on Ecological Stability

LUCC and GDP positively influence the strengthening of ecosystem resistance; how-ever, GDP expansion has a detrimental impact on recovery (Figure 7b,c). These findings demonstrate that human activities can have both beneficial and adverse effects on ecological responses to drought. Owing to the ecological management initiative in the Ugan–Kuqa River Basin, the tree coverage within the basin has increased [67]. Trees, with greater drought resistance than annual or biennial herbaceous plants, benefit from deeper root systems [68], mitigating the decline in productivity caused by drought. However, upon further quantifying the impact of human activities on the ecological stress within the watershed, the ecological stress coefficient displayed an upward trajectory, rising from 0.328 in 2000 to 0.330 in 2010 and further to 0.394 in 2020. The rate of increase over the last decade exceeded that of the previous ten years by 32-fold, which was attributed primarily to the expansion of cultivated and urban areas. Consequently, accurate assessment of the positive and negative implications of human activities on the stability of watershed ecosystems, as well as understanding this dualistic aspect of human endeavors, is crucial for sustaining the well-being of oasis–desert ecosystems and promoting enduring socioeconomic development.
The response of ecological stability to habitat fragmentation is complex. PD, ED, and the LSI all play beneficial roles in enhancing ecosystem resistance (Figure 7b). Nevertheless, an increase in the LSI notably diminished ecosystem recovery (Figure 7b), illustrating the dual effects of habitat fragmentation on stability. Habitat fragmentation has the potential to increase ecosystem heterogeneity, which is particularly evident in the transitional zones between farmland and grassland, where these peripheral areas often present the highest levels of species diversity and abundance [69,70,71]. This contributes significantly to increased ecosystem resistance. Conversely, habitat fragmentation, instigated by human activities such as urban expansion and agricultural advancements, unavoidably disrupts the natural patches of the landscape within a watershed, leading to negative repercussions for the continuity and overall stability of ecosystems [72,73].

4.4. Limitations

This article examined the ecosystem stability of the Ugan–Kuqa River Basin based on remote sensing data. Although remote sensing technology is a powerful tool for evaluating ecosystem stability at large scales and over extended time periods, its spatial and temporal resolution may affect the accuracy and detailed depiction of model results. Furthermore, this study evaluated ecosystem stability on an annual basis; however, ecosystems respond to various pressures in different ways, which can result in gradual, sudden, or lagged effects. Therefore, future research should focus on quantifying the response mechanisms and response scales of ecosystems to diverse stressors, as well as developing techniques to analyze changes in ecological stability across different scales. This approach will contribute to a deeper understanding of the dynamic aspects of ecosystems and their sustainable management.

5. Conclusions

On the basis of GPP data, this study performed a spatiotemporal dynamic evaluation of the resistance and recovery of the Ugan–Kuqa River Basin ecosystem, examining the trade-off between the two, and analyzed the relative significance of climatic factors, anthropogenic activities, and habitat fragmentation on ecosystem stability. This study revealed that the ecosystem stability of the Ugan–Kuqa River Basin improved significantly from 2001 to 2020, with greater stability observed in the northern region compared to the southern areas. The trade-off between resistance and recovery in the basin decreased, and the area exhibiting high resistance and high recovery significantly increased.
The analysis of the determining factors demonstrated that climate warming and humidification emerged as pivotal elements contributing to the enhancement of ecosystem stability. In contrast, the influence of human activities on these ecosystems was minimal compared to that of habitat fragmentation, although both exhibited dual characteristics. Therefore, it was essential to protect and intervene in critical ecological zones within the Ugan–Kuqa River Basin. Such measures would facilitate the prediction of potential im-pacts of future climate change on dry ecosystems and enable the integration of ecological conservation factors into land use planning, thereby ensuring the long-term stability and sustainability of these ecosystems.

Author Contributions

Methodology, M.S., L.Z. and X.L.; formal analysis, P.Z.; data curation, Y.S.; writing—original draft preparation, T.Z.; writing—review and editing, R.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the National Key Research and Development Project of China (No. 2021YFC3201500) and the Special project of National Natural Science Foundation of China (No. 42442035).

Data Availability Statement

Data are contained within the article.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Sadiqi, S.S.J.; Hong, E.-M.; Nam, W.-H.; Kim, T. Review: An integrated framework for understanding ecological drought and drought resistance. Sci. Total Environ. 2022, 846, 157477. [Google Scholar] [CrossRef] [PubMed]
  2. IPCC. Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change; Core Writing Team, Lee, H., Romero, J., Eds.; IPCC: Geneva, Switzerland, 2023; pp. 35–115. [Google Scholar]
  3. Huang, Z.; Yuan, X.; Liu, X. The key drivers for the changes in global water scarcity: Water withdrawal versus water availability. J. Hydrol. 2021, 601, 126658. [Google Scholar] [CrossRef]
  4. Haile, G.G.; Tang, Q.; Li, W.; Liu, X.; Zhang, X. Drought: Progress in broadening its understanding. WIREs Water 2020, 7, e1407. [Google Scholar] [CrossRef]
  5. Wolf, S.; Paul-Limoges, E. Drought and heat reduce forest carbon uptake. Nat. Commun. 2023, 14, 6217. [Google Scholar] [CrossRef]
  6. Li, W.; Pacheco-Labrador, J.; Migliavacca, M.; Miralles, D.; Hoek van Dijke, A.; Reichstein, M.; Forkel, M.; Zhang, W.; Frankenberg, C.; Panwar, A.; et al. Widespread and complex drought effects on vegetation physiology inferred from space. Nat. Commun. 2023, 14, 4640. [Google Scholar] [CrossRef] [PubMed]
  7. Gampe, D.; Zscheischler, J.; Reichstein, M.; O’Sullivan, M.; Smith, W.K.; Sitch, S.; Buermann, W. Increasing impact of warm droughts on northern ecosystem productivity over recent decades. Nat. Clim. Change 2021, 11, 772–779. [Google Scholar] [CrossRef]
  8. Stefanidis, S.; Rossiou, D.; Proutsos, N. Drought Severity and Trends in a Mediterranean Oak Forest. Hydrology 2023, 10, 167. [Google Scholar] [CrossRef]
  9. Gao, X.; Huang, P.; Wang, K. Assessment of the ecosystem stability of Shapotou Arid Desert Nature Reserve in Ningxia, China. Acta Ecol. Sin. 2019, 39, 6381–6392. [Google Scholar] [CrossRef]
  10. Noy Meir, I. Stability in arid ecosystems and the effects of man on it. In Proceedings of the 1st International Congress of Ecology, The Hague, The Netherlands, 8–14 September 1974; pp. 220–225. [Google Scholar]
  11. Van Meerbeek, K.; Jucker, T.; Svenning, J. Unifying the concepts of stability and resilience in ecology. J. Ecol. 2021, 109, 3114–3132. [Google Scholar] [CrossRef]
  12. Kéfi, S.; Domínguez-García, V.; Donohue, I.; Fontaine, C.; Thébault, E.; Dakos, V. Advancing our understanding of ecological stability. Ecol. Lett. 2019, 22, 1349–1356. [Google Scholar] [CrossRef]
  13. Justus, J. Complexity, Diversity, and Stability. In A Companion to the Philosophy of Biology; John Wiley & Sons, Ltd.: Hoboken, NJ, USA, 2007; pp. 321–350. [Google Scholar] [CrossRef]
  14. Hillebrand, H.; Langenheder, S.; Lebret, K.; Lindström, E.; Östman, Ö.; Striebel, M. Decomposing multiple dimensions of stability in global change experiments. Ecol. Lett. 2018, 21, 21–30. [Google Scholar] [CrossRef] [PubMed]
  15. Hillebrand, H.; Kunze, C. Meta-analysis on pulse disturbances reveals differences in functional and compositional recovery across ecosystems. Ecol. Lett. 2020, 23, 575–585. [Google Scholar] [CrossRef]
  16. Donohue, I.; Hillebrand, H.; Montoya, J.M.; Petchey, O.L.; Pimm, S.L.; Fowler, M.S.; Healy, K.; Jackson, A.L.; Lurgi, M.; McClean, D.; et al. Navigating the complexity of ecological stability. Ecol. Lett. 2016, 19, 1172–1185. [Google Scholar] [CrossRef] [PubMed]
  17. Hoover, D.L.; Pfennigwerth, A.A.; Duniway, M.C. Drought resistance and resilience: The role of soil moisture–plant interactions and legacies in a dryland ecosystem. J. Ecol. 2021, 109, 3280–3294. [Google Scholar] [CrossRef]
  18. Li, Y.; Zhang, W.; Schwalm, C.R.; Gentine, P.; Smith, W.K.; Ciais, P.; Kimball, J.S.; Gazol, A.; Kannenberg, S.A.; Chen, A.; et al. Widespread spring phenology effects on drought recovery of Northern Hemisphere ecosystems. Nat. Clim. Change 2023, 13, 182–188. [Google Scholar] [CrossRef]
  19. Abel, C.; Aestre, F.T.; Berdugo, M.; Agesson, T.; Abdi, A.M.; Orion, S.; Ensholt, R. Vegetation resistance to increasing aridity when crossing thresholds depends on local environmental conditions in global drylands. Commun. Earth Environ. 2024, 5, 379. [Google Scholar] [CrossRef]
  20. Ruppert, J.C.; Harmoney, K.; Henkin, Z.; Snyman, H.A.; Sternberg, M.; Willms, W.; Linstädter, A. Quantifying drylands’ drought resistance and recovery: The importance of drought intensity, dominant life history and grazing regime. Glob. Change Biol. 2015, 21, 1258–1270. [Google Scholar] [CrossRef]
  21. Yao, Y.; Fu, B.; Liu, Y.; Li, Y.; Wang, S.; Zhan, T.; Wang, Y.; Gao, D. Evaluation of ecosystem resilience to drought based on drought intensity and recovery time. Agric. For. Meteorol. 2022, 314, 108809. [Google Scholar] [CrossRef]
  22. Shao, X.; Zhang, Y.; Ma, N.; Zhang, X.; Tian, J.; Xu, Z.; Liu, C. Drought-induced ecosystem resistance and recovery observed at 118 flux tower stations across the globe. Agric. For. Meteorol. 2024, 356, 110170. [Google Scholar] [CrossRef]
  23. Berdugo, M.; Delgado-Baquerizo, M.; Soliveres, S.; Hernández-Clemente, R.; Zhao, Y.; Gaitán, J.J.; Gross, N.; Saiz, H.; Maire, V.; Lehmann, A.; et al. Global ecosystem thresholds driven by aridity. Science 2020, 367, 787–790. [Google Scholar] [CrossRef]
  24. Pan, X. A preliminary study on the stability of oasis ecosystem in arid area. Quat. Sci. 2001, 21, 345–351. [Google Scholar]
  25. Li, D.; Yang, Y.; Xia, F.; Sun, W.; Li, X.; Xie, Y. Exploring the influences of different processes of habitat fragmentation on ecosystem services. Landsc. Urban Plan. 2022, 227, 104544. [Google Scholar] [CrossRef]
  26. Huang, S.; Ding, J.; Liu, B.; Ge, X.; Wang, J.; Zou, J.; Zhang, J. The Capability of Integrating Optical and Microwave Data for Detecting Soil Moisture in an Oasis Region. Remote Sens. 2020, 12, 1358. [Google Scholar] [CrossRef]
  27. Xia, H.; Zhao, X.; Sha, Y. HSPEI: A 1-km spatial resolution SPEI Dataset across Chinese Mainland from 2001 to 2022. Sci. Data Bank 2024, 11, 479–494. [Google Scholar] [CrossRef]
  28. Running, S.W.; Glassy, J.M.; Thornton, P.E. MODIS Daily Photosynthesis (PSN) and Annual Net Primary Production (NPP) Product (MOD17) Algorithm Theoretical Basis Document; SCF At-Launch Algorithm ATBD Documents 1999; University of Montana: Missoula, MT, USA, 1999. [Google Scholar]
  29. Liu, Y.; Liu, R. A Simple Approach for Mapping Forest Cover from Time Series of Satellite Data. Remote Sens. 2020, 12, 2918. [Google Scholar] [CrossRef]
  30. Chen, J.; Gao, M.; Cheng, S.; Hou, W.; Song, M.; Liu, X.; Liu, Y. Global 1 km × 1 km gridded revised real gross domestic product and electricity consumption during 1992–2019 based on calibrated nighttime light data. Sci. Data 2022, 9, 202. [Google Scholar] [CrossRef]
  31. Liu, J.; Coomes, D.A.; Gibson, L.; Hu, G.; Liu, J.; Luo, Y.; Wu, C.; Yu, M. Forest fragmentation in China and its effect on biodiversity. Biol. Rev. 2019, 94, 1636–1657. [Google Scholar] [CrossRef]
  32. Xiong, Y.; Pan, R.; Xu, G.; Jiao, L.; Li, K. A comparison of spatial and temporal characteristics of urban expansion in India during 1990–2014. Prog. Geogr. 2019, 38, 271–282. [Google Scholar]
  33. Wu, J.; Luo, K.; Zhao, Y. The evolution of urban landscape pattern and its driving forces of Shenzhen from 1996 to 2015. Geographic. Res. 2020, 39, 1725–1738. [Google Scholar]
  34. Vicente-Serrano, S.M.; Beguería, S.; López-Moreno, J.I. A Multiscalar Drought Index Sensitive to Global Warming: The Standardized Precipitation Evapotranspiration Index. J. Clim. 2010, 23, 1696–1718. [Google Scholar] [CrossRef]
  35. GB/T 20481-2017; Grades of Meteorological Drought. China Standard Press: Beijing, China, 2017.
  36. Yao, Y.; Liu, Y.; Song, J.; Tao, S.; Li, Y.; Wu, T.; Wang, Y.; Wang, S.; Fu, B. Declining Tradeoff Between Resistance and Resilience of Ecosystems to Drought. Earths Future 2024, 12, e2024EF004665. [Google Scholar] [CrossRef]
  37. Ver Hoef, J.M.; Peterson, E.E.; Hooten, M.B.; Hanks, E.M.; Fortin, M. Spatial autoregressive models for statistical inference from ecological data. Ecol. Monogr. 2018, 88, 36–59. [Google Scholar] [CrossRef]
  38. Fan, Y.; Chen, J.; Shirkey, G.; John, R.; Wu, S.R.; Park, H.; Shao, C. Applications of structural equation modeling (SEM) in ecological studies: An updated review. Ecol. Process. 2016, 5, 19. [Google Scholar] [CrossRef]
  39. Thompson, B. Exploratory and Confirmatory Factor Analysis: Understanding Concepts and Applications; APA: Washington, DC, USA, 2004. [Google Scholar] [CrossRef]
  40. Klem, L. Path analysis. In Reading and Understanding Multivariate Statistics; American Psychological Association: Washington, DC, USA, 1995; pp. 65–97. [Google Scholar]
  41. Sun, Q.; Zhang, X.; Jiang, M. Eco-environmental Variables Estimation from Remotely Sensed Data and Eco-environmental Assessment: Models and System. Acta Sci. Nat. Univ. Pekin. 2011, 47, 1073–1080. [Google Scholar] [CrossRef]
  42. Zhou, L.; Wu, J.; Zhang, J. Remote Sensing-based Drought Monitoring Approach and Research Progress. Sci. Geogr. Sin. 2015, 35, 630–636. [Google Scholar] [CrossRef]
  43. Wang, Y.; Liu, B.; Zhao, J.; Ye, C.; Wei, L.; Sun, J.; Chu, C.; Lee, T.M. Global patterns and abiotic drivers of ecosystem multifunctionality in dominant natural ecosystems. Environ. Int. 2022, 168, 107480. [Google Scholar] [CrossRef]
  44. Zhao, Y.; Liu, S.; Liu, H.; Wang, F.; Dong, Y.; Wu, G.; Li, Y. Multilevel driving mechanism of ecosystem multidimensional stability in the Yangtze River Economic Belt: A hierarchical linear model approach. J. Clean. Prod. 2024, 449, 141513. [Google Scholar] [CrossRef]
  45. Gazol, A.; Camarero, J.J.; Anderegg, W.R.L.; Vicente-Serrano, S.M. Impacts of droughts on the growth resilience of Northern Hemisphere forests. Glob. Ecol. Biogeogr. 2017, 26, 166–176. [Google Scholar] [CrossRef]
  46. Smith, T.; Boers, N. Global vegetation resilience linked to water availability and variability. Nat. Commun. 2023, 14, 498. [Google Scholar] [CrossRef]
  47. Gong, H.; Wang, G.; Fan, C.; Zhuo, X.; Sha, L.; Kuang, Z.; Bi, J.; Cheng, T. Temporal accumulation and lag effects of precipitation on carbon fluxes in terrestrial ecosystems across semi-arid regions in China. Agric. For. Meteorol. 2024, 356, 110189. [Google Scholar] [CrossRef]
  48. Madon, O.; Médail, F. The ecological significance of annuals on a Mediterranean grassland (Mt Ventoux, France). Plant Ecol. 1997, 129, 189–199. [Google Scholar] [CrossRef]
  49. Krieger, A. Temporal dynamics of an ephemeral plant community: Species turnover in seasonal rock pools on Ivorian inselbergs. Plant Ecol. 2003, 167, 283–292. [Google Scholar] [CrossRef]
  50. Xi, L.; Gou, Q.; Wang, G.; Song, B. The responses of typical annual herbaceous plants to drought stress in a desert-oasis ecotone. Acta Ecol. Sin. 2021, 41, 5425–5434. [Google Scholar] [CrossRef]
  51. Yasen, M.; Shawuti, M.; Aishan, T.; Mijiti, R.; Abudumiti, Y.; Reheman, M. Spatial-Temporal Characteristics of Cropland in the Ugan-Kuqa River Delta Oasis. Sci. Agric. Sin. 2017, 50, 3506–3518. [Google Scholar] [CrossRef]
  52. Tilman, D.; Downing, J.A. Biodiversity and stability in grasslands. Nature 1994, 367, 363–365. [Google Scholar] [CrossRef]
  53. Loreau, M.; De Mazancourt, C. Biodiversity and ecosystem stability: A synthesis of underlying mechanisms. Ecol. Lett. 2013, 16, 106–115. [Google Scholar] [CrossRef]
  54. Isbell, F.; Craven, D.; Connolly, J.; Loreau, M.; Schmid, B.; Beierkuhnlein, C.; Bezemer, T.M.; Bonin, C.; Bruelheide, H.; De Luca, E.; et al. Biodiversity increases the resistance of ecosystem productivity to climate extremes. Nature 2015, 526, 574–577. [Google Scholar] [CrossRef]
  55. Chen, F.; Xie, T.; Yang, Y.; Chen, S.; Chen, F.; Huang, W.; Chen, J. Discussion of the “warming and wetting” trend and its future variation in the drylands of Northwest China under global warming. Sci. China Earth Sci. 2023, 66, 1241–1257. [Google Scholar] [CrossRef]
  56. Ding, Y.; Liu, Y.; Xu, Y.; Wu, P.; Xue, T.; Wang, J.; Shi, Y.; Zhang, Y.; Song, Y.; Wang, P. Regional Responses to Global Climate Change: Progress and Prospects for Trend, Causes, and Projection of Climatic Warming-Wetting in Northwest China. Adv. Earth Sci. 2023, 38, 551–562. [Google Scholar] [CrossRef]
  57. Li, M.; Du, J.; Li, W.; Li, R.; Wu, S.; Wang, S. Global Vegetation Change and Its Relationship with Precipitation and Temperature Based on GLASS-LAI in 1982-2015. Sci. Geogr. Sin. 2020, 40, 823–832. [Google Scholar] [CrossRef]
  58. Zhang, B.; Tian, L.; Zhao, X.; Wu, P. Feedbacks between vegetation restoration and local precipitation over the Loess Plateau in China. Sci. China Earth Sci. 2021, 64, 920–931. [Google Scholar] [CrossRef]
  59. Zhang, R.; Ouyang, Z.; Xie, X.; Guo, H.; Tan, D.; Xiao, X.; Qi, J.; Zhao, B. Impact of Climate Change on Vegetation Growth in Arid Northwest of China from 1982 to 2011. Remote Sens. 2016, 8, 364. [Google Scholar] [CrossRef]
  60. Piao, S.; Wang, X.; Park, T.; Chen, C.; Lian, X.; He, Y.; Bjerke, J.W.; Chen, A.; Ciais, P.; Tømmervik, H.; et al. Characteristics, drivers and feedbacks of global greening. Nat. Rev. Earth Environ. 2019, 1, 14–27. [Google Scholar] [CrossRef]
  61. Zhao, Y.; Chen, Y.; Wu, C.; Li, G.; Ma, M.; Fan, L.; Zheng, H.; Song, L.; Tang, X. Exploring the contribution of environmental factors to evapotranspiration dynamics in the Three-River-Source region, China. J. Hydrol. 2023, 626, 130222. [Google Scholar] [CrossRef]
  62. Huo, Z.; Dai, X.; Feng, S.; Kang, S.; Huang, G. Effect of climate change on reference evapotranspiration and aridity index in arid region of China. J. Hydrol. 2013, 492, 24–34. [Google Scholar] [CrossRef]
  63. Li, Z.; Chen, Y.; Shen, Y.; Liu, Y.; Zhang, S. Analysis of changing pan evaporation in the arid region of Northwest China. Water Resour. Res. 2013, 49, 2205–2212. [Google Scholar] [CrossRef]
  64. Xing, W.; Wang, W.; Shao, Q.; Yu, Z.; Yang, T.; Fu, J. Periodic fluctuation of reference evapotranspiration during the past five decades: Does Evaporation Paradox really exist in China? Sci. Rep. 2016, 6, 39503. [Google Scholar] [CrossRef]
  65. Deng, X.; Liu, Y.; Liu, Z.; Yao, J. Temporal-spatial dynamic change characteristics of evapotranspiration in arid region of Northwest China. Acta Ecol. Sin. 2017, 37, 2994–3008. [Google Scholar] [CrossRef]
  66. Yang, Y.; Roderick, M.L.; Guo, H.; Miralles, D.G.; Zhang, L.; Fatichi, S.; Luo, X.; Zhang, Y.; McVicar, T.R.; Tu, Z.; et al. Evapotranspiration on a greening Earth. Nat. Rev. Earth Environ. 2023, 4, 626–641. [Google Scholar] [CrossRef]
  67. Li, D. Aksu, Xinjiang: Continuously building ecological projects to expand three major benefits. Farmers’ Daily, 26 April 2022; No. 003. [Google Scholar] [CrossRef]
  68. Stuart-Haëntjens, E.; De Boeck, H.J.; Lemoine, N.P.; Mänd, P.; Kröel-Dulay, G.; Schmidt, I.K.; Jentsch, A.; Stampfli, A.; Anderegg, W.R.L.; Bahn, M.; et al. Mean annual precipitation predicts primary production resistance and resilience to extreme drought. Sci. Total Environ. 2018, 636, 360–366. [Google Scholar] [CrossRef]
  69. Fahrig, L. Effects of Habitat Fragmentation on Biodiversity. Annu. Rev. Ecol. Evol. Syst. 2003, 34, 487–515. [Google Scholar] [CrossRef]
  70. Fahrig, L. Ecological Responses to Habitat Fragmentation Per Se. Annu. Rev. Ecol. Evol. Syst. 2017, 48, 1–23. [Google Scholar] [CrossRef]
  71. Kleijn, D.; Rundlöf, M.; Scheper, J.; Smith, H.G.; Tscharntke, T. Does conservation on farmland contribute to halting the biodiversity decline? Trends Ecol. Evol. 2011, 26, 474–481. [Google Scholar] [CrossRef] [PubMed]
  72. Zhang, X.; Zhou, J.; Li, M. Analysis on spatial and temporal changes of regional habitat quality based on the spatial pattern reconstruction of land use. Acta Geogr. Sin. 2020, 75, 01000160. [Google Scholar] [CrossRef]
  73. Zheng, L.; Wang, Y.; Li, J. Quantifying the spatial impact of landscape fragmentation on habitat quality: A multi-temporal dimensional comparison between the Yangtze River Economic Belt and Yellow River Basin of China. Land Use Policy 2023, 125, 106463. [Google Scholar] [CrossRef]
Remotesensing 16 04272 g001
Figure 1. Schematic diagram of the research region. (a) Geographical location of the study area; (b) Remote sensing image map of the study area; (c) Land use types in the study area; (d) Area percentage of different land use types in the study area.
Figure 1. Schematic diagram of the research region. (a) Geographical location of the study area; (b) Remote sensing image map of the study area; (c) Land use types in the study area; (d) Area percentage of different land use types in the study area.
Remotesensing 16 04272 g001
Remotesensing 16 04272 g002
Figure 2. Conceptual meta-model of the ecological stability drivers. (a) The overall relationship between stability and climate, vegetation, human activities, and fragmentation. (b) The specific modeling relationship between the drivers of ecosystem stability. The pathways illustrate the interconnections among Pre, Tem, ET, LUCC, GDP, NDVI, ED, PD, LSI, and ecosystem resistance and recovery. Unidirectional arrows denote causation.
Figure 2. Conceptual meta-model of the ecological stability drivers. (a) The overall relationship between stability and climate, vegetation, human activities, and fragmentation. (b) The specific modeling relationship between the drivers of ecosystem stability. The pathways illustrate the interconnections among Pre, Tem, ET, LUCC, GDP, NDVI, ED, PD, LSI, and ecosystem resistance and recovery. Unidirectional arrows denote causation.
Remotesensing 16 04272 g002
Remotesensing 16 04272 g003
Figure 3. Temporal changes in resistance and recovery.
Figure 3. Temporal changes in resistance and recovery.
Remotesensing 16 04272 g003
Remotesensing 16 04272 g004
Figure 4. Spatial trends in ecosystem resistance and recovery. (a) Spatial trend in resistance. (b) Spatial trend in recovery.
Figure 4. Spatial trends in ecosystem resistance and recovery. (a) Spatial trend in resistance. (b) Spatial trend in recovery.
Remotesensing 16 04272 g004
Remotesensing 16 04272 g005
Figure 5. Scatter plot of resistance–recovery trade-off distributions. The data for both resistance and recovery were standardized within the range of 0 to 1, with the intersection where both factors equate to 0 signifying the origin of the coordinates. The delineation was set at 0.5; values below 0.5 for both resistance and recovery indicate a state of low resistance–low recovery, whereas values surpassing 0.5 indicate high resistance–high recovery. Additionally, resistance above 0.5 and recovery below 0.5 indicate high resistance–low recovery, whereas resistance below 0.5 and recovery above 0.5 suggest low resistance–high recovery.
Figure 5. Scatter plot of resistance–recovery trade-off distributions. The data for both resistance and recovery were standardized within the range of 0 to 1, with the intersection where both factors equate to 0 signifying the origin of the coordinates. The delineation was set at 0.5; values below 0.5 for both resistance and recovery indicate a state of low resistance–low recovery, whereas values surpassing 0.5 indicate high resistance–high recovery. Additionally, resistance above 0.5 and recovery below 0.5 indicate high resistance–low recovery, whereas resistance below 0.5 and recovery above 0.5 suggest low resistance–high recovery.
Remotesensing 16 04272 g005
Remotesensing 16 04272 g006
Figure 6. Drivers of ecosystem stability, illustrating the simulated effects of climate, vegetation, human ac-tivities, and habitat fragmentation on ecosystem stability. The colors of the arrows represent the degree of significance of the relationships, with dashed arrows indicating nonsignificant correlations. Climatic factors are denoted by blue boxes, vegetation variables are denoted by green boxes, human activity variables are denoted by pink boxes, habitat fragmentation variables are denoted by gray boxes, and stability variables are denoted by yellow boxes.
Figure 6. Drivers of ecosystem stability, illustrating the simulated effects of climate, vegetation, human ac-tivities, and habitat fragmentation on ecosystem stability. The colors of the arrows represent the degree of significance of the relationships, with dashed arrows indicating nonsignificant correlations. Climatic factors are denoted by blue boxes, vegetation variables are denoted by green boxes, human activity variables are denoted by pink boxes, habitat fragmentation variables are denoted by gray boxes, and stability variables are denoted by yellow boxes.
Remotesensing 16 04272 g006
Remotesensing 16 04272 g007
Figure 7. Impact of drivers on ecosystem stability. (a) Presents a stacked plot showing the proportions of the total effects of these factors on resistance and recovery, with cells representing negative impacts. (b) Showcases a stacked plot illustrating the percentages of direct versus indirect influences of the variables on resistance, with cells denoting negative impacts. Finally, (c) presents a stacked plot displaying the percentages of direct versus indirect influences of the variables on recovery, with cells indicating a negative impact.
Figure 7. Impact of drivers on ecosystem stability. (a) Presents a stacked plot showing the proportions of the total effects of these factors on resistance and recovery, with cells representing negative impacts. (b) Showcases a stacked plot illustrating the percentages of direct versus indirect influences of the variables on resistance, with cells denoting negative impacts. Finally, (c) presents a stacked plot displaying the percentages of direct versus indirect influences of the variables on recovery, with cells indicating a negative impact.
Remotesensing 16 04272 g007
Table 1. SPEI drought classification.
Table 1. SPEI drought classification.
ClassTypeSPEI
1NormalSPEI > −0.5
2Abnormal drought−1.0 < SPEI ≤ −0.5
3Moderate drought−1.5 < SPEI ≤ −1.0
4Severe drought−2.0 < SPEI ≤ −1.5
5Extreme droughtSPEI ≤ −2.0
Class: drought level category. Type: corresponding drought type. SPEI: standardized precipitation evapotranspiration index (SPEI), used to assess the degree of drought, with lower values indicating more severe drought.
Table 2. Performance of ecosystem stability driver modeling.
Table 2. Performance of ecosystem stability driver modeling.
χ2/d.f.pRMSEARMRGFIAGFI
SEM17.958 < 0.010.0360.0420.9830.946
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zhou, T.; Zhu, P.; Yang, R.; Sun, Y.; Sun, M.; Zhang, L.; Li, X. Ecosystem Stability in the Ugan–Kuqa River Basin, Xinjiang, China: Investigation of Spatial and Temporal Dynamics and Driving Forces. Remote Sens. 2024, 16, 4272. https://doi.org/10.3390/rs16224272

AMA Style

Zhou T, Zhu P, Yang R, Sun Y, Sun M, Zhang L, Li X. Ecosystem Stability in the Ugan–Kuqa River Basin, Xinjiang, China: Investigation of Spatial and Temporal Dynamics and Driving Forces. Remote Sensing. 2024; 16(22):4272. https://doi.org/10.3390/rs16224272

Chicago/Turabian Style

Zhou, Ting, Peiyue Zhu, Rongjin Yang, Yilin Sun, Meiying Sun, Le Zhang, and Xiuhong Li. 2024. "Ecosystem Stability in the Ugan–Kuqa River Basin, Xinjiang, China: Investigation of Spatial and Temporal Dynamics and Driving Forces" Remote Sensing 16, no. 22: 4272. https://doi.org/10.3390/rs16224272

APA Style

Zhou, T., Zhu, P., Yang, R., Sun, Y., Sun, M., Zhang, L., & Li, X. (2024). Ecosystem Stability in the Ugan–Kuqa River Basin, Xinjiang, China: Investigation of Spatial and Temporal Dynamics and Driving Forces. Remote Sensing, 16(22), 4272. https://doi.org/10.3390/rs16224272

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop