An Evaluation of Ecosystem Quality and Its Response to Aridity on the Qinghai–Tibet Plateau

Abstract

Exploring the response of spatial and temporal characteristics of ecological quality change to aridity on the Qinghai–Tibet Plateau (QTP) can provide valuable information for regional ecological protection, water resource management, and climate change adaptation. In this study, we constructed the Remote Sensing Ecological Index (RSEI) and Standardized Precipitation Evapotranspiration Index (SPEI) based on the Google Earth Engine (GEE) platform with regional characteristics and completely analyzed the spatial and temporal variations of aridity and ecological quality on the QTP in the years 2000, 2005, 2010, 2015, and 2020. Additionally, we explored the responses of ecological quality to aridity indices at six different time scales. The Mann–Kendall test, correlation analysis, and significance test were used to study the spatial and temporal distribution characteristics of meteorological aridity at different time scales on the QTP and their impacts on the quality of the ecological environment. The results show that the ecological environmental quality of the QTP has a clear spatial distribution pattern. The ecological environment quality is significantly better in the south-east, while the Qaidam Basin and the west have lower ecological environment quality indices, but the overall trend of environmental quality is getting better. The Aridity Index of the QTP shows a differentiated spatial and temporal distribution pattern, with higher Aridity Indexes in the north-eastern and south-western parts of the plateau and lower Aridity Indexes in the central part of the plateau at shorter time scales. Monthly, seasonal, and annual-scale SPEI values showed an increasing trend. There is a correlation between aridity conditions and ecological quality on the QTP. The areas with significant positive correlation between the RSEI and SPEI in the study area were mainly concentrated in the south-eastern, south-western, and northern parts of the QTP, where the ecological quality of the environment is more seriously affected by meteorological aridity.

1. Introduction

According to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC), as global warming continues, more regions are expected to experience an increased frequency and severity of aridity. Aridity, as one of the most devastating and widespread extreme climatic conditions, has severe impacts on water resources, ecosystems, agricultural production, and economic development [1]. The main types of aridity include meteorological, agricultural, hydrological, and socio-economic aridity [2,3]. The degree of aridity significantly influences vegetation cover types and structures, thereby affecting the quality of local ecosystems [4]. With ongoing global warming, both the intensity and frequency of aridity are expected to increase, further impacting vegetation productivity. Prolonged and intense aridity can even lead to substantial changes in ecosystem quality [5]. Therefore, understanding the response of ecological environment quality to meteorological aridity is crucial for regional ecological protection and sustainable development.

The ecological environment forms the foundation of human survival, economic development, and social progress. However, with advances in science and technology and the intensification of human activities, ecosystems globally have suffered considerable damage. This has led to an increasing frequency of ecological problems and poses serious threats to regional ecological security. Conducting ecological quality analyses and evaluations can provide targeted recommendations for environmental protection and is vital for promoting harmony between humanity and nature, as well as sustainable development. The Qinghai–Tibet Plateau (QTP) is characterized by a complex landscape, harsh natural environment, and extremely fragile ecosystem. In recent years, significant climate change and intensified human activities have resulted in varying degrees of degradation of the plateau’s ecosystem. Consequently, the ecological quality of the QTP has garnered considerable attention from both domestic and international scholars, leading to numerous studies on vegetation changes, grassland degradation, and related issues. For example, recent research has analyzed the spatio-temporal patterns of Net Primary Productivity (NPP) changes in the QTP’s vegetation over the past two decades, clarifying its relationship with various influencing factors [6]. Other studies have examined the spatio-temporal variations in water use efficiency and the effects of climatic factors on the QTP [7], as well as the ecological vulnerability of the QTP, identifying vegetation as a major driver [8]. Additionally, research has explored the combined effects of climate change and human activities on the spatio-temporal dynamics of vegetation in the QTP by conducting correlation analyses using data from meteorological stations and regional economic statistics [9]. Studies have demonstrated the effectiveness of the Remote Sensing Ecological Index (RSEI) in monitoring and assessing the ecological conditions of the QTP. For instance, study has highlighted significant spatio-temporal variations in ecological quality across the region over the past two decades, with periods of ecological degradation primarily driven by rising temperatures and overgrazing. Notably, between 2015 and 2020, overgrazing in the south-western regions of the QTP was identified as a major factor contributing to poor ecological conditions, as indicated by lower RSEI values [10]. Additionally, studies focused on ecological assessments within the QTP, such as the research conducted in the Changtang Nature Reserve, have used the RSEI to evaluate changes in ecological quality over extended periods. These studies have revealed both periods of ecological deterioration and subsequent recovery, underscoring the value of the RSEI in tracking long-term ecological trends [11]. Furthermore, study on ecological vulnerability across the QTP has utilized the RSEI to map and analyze patterns of vulnerability, demonstrating its utility in guiding environmental protection and policy-making [12]. These studies underscore the RSEI’s effectiveness as a reliable tool for assessing environmental quality and informing ecological management strategies in the Qinghai–Tibet Plateau. However, focusing on changes in specific ecosystem features alone makes it difficult to synthesize a comprehensive assessment of the plateau’s overall ecological state.

Aridity is a multi-scale phenomenon. The variability in water availability, including soil moisture, groundwater, reservoir storage, river discharge, and snowpack, exemplifies the complexity of aridity [13]. The time lag between water input and the availability of usable resources varies significantly. Therefore, understanding the accumulation time scales of water scarcity, which functionally distinguishes between hydrological, environmental, agricultural, and other forms of aridity, is critical. For instance, the hydrological system’s response to precipitation can differ widely over time [14], determined by the varying frequencies of hydrological and climatic variables [15]. Consequently, the Aridity Index must be associated with a specific time scale to facilitate the monitoring and management of different water resources. The Aridity Index is an effective method for assessing aridity, simplifying the complex phenomenon and measuring the severity of aridity events [16]. Numerous aridity indices have been proposed internationally, with commonly used indices for meteorological aridity monitoring including the Relative Humidity Index (M), Standardized Precipitation Index (SPI), and Standardized Precipitation Evapotranspiration Index (SPEI) [17]. Among these, the SPEI is widely used due to its integration of multiple climate factors and its applicability across different timescales. The Standardized Precipitation Evapotranspiration Index (SPEI) evolves from the Normalized Precipitation Index (NPI) [18], incorporating factors such as precipitation, temperature, and potential evapotranspiration to provide a comprehensive measure of water balance and aridity conditions [19].

In the study, the SPEI at several time scales of 1, 3, 6, 9, 12, and 24 are selected to reflect different characteristics of meteorological drought. By analyzing different time scales, the severity and impact range of drought can be assessed more comprehensively. The one-month SPEI (SPEI-1) serves as a critical indicator for monitoring short-term meteorological anomalies, such as seasonal fluctuations or abrupt arid events. As the time scale expands, the three-month SPEI (SPEI-3) delineates the drought conditions over a complete season, providing a more integrated view of seasonal aridity patterns. The six-month SPEI (SPEI-6) extends this assessment to a semi-annual basis, offering insights into medium-term drought dynamics. The annual SPEI (SPEI-12) synthesizes the year-long precipitation and evaporation trends, reflecting the annual drought status and its implications for agricultural planning and water resource management. Furthermore, the two-year SPEI (SPEI-24) encapsulates long-term drought conditions, revealing persistent drought patterns and the potential long-term impacts of climate change on hydrological cycles and ecosystem health.

In recent years, numerous approaches have been developed for monitoring ecosystem status and detecting changes. Remote sensing techniques, celebrated for their effectiveness and precision, have become a staple in ecological assessments. These include indices such as the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), Normalized Difference Water Index (NDWI), Normalized Difference Moisture Index (NDMI), and Land Surface Temperature (LST). The Remote Sensing-based Ecological Index (RSEI) [20] was introduced as a comprehensive ecological assessment tool, integrating these remote sensing indices with socio-economic dimensions. The RSEI combines four key remote sensing indicators: NDVI for vegetation greenness, Wetness Index (WET) for moisture, NDBSI for aridity, and LST for thermal conditions. The specific contributions of each factor to ecological quality are determined by covariance-based Principal Component Analysis (PCA). The synthesis of these indicators into the RSEI offers an objective and meaningful approach to assessing ecological quality across different regions and ecosystems [21,22,23]. Therefore, the RSEI is a feasible and reliable tool for assessing the changing environmental quality in the QTP region. Additionally, the acquisition of long time-series and high-resolution remote sensing datasets, coupled with the complicated data preprocessing and RSEI calculation, is often time-consuming and costly. However, the cloud platform Google Earth Engine (GEE) offers high-performance resources for processing large geospatial datasets, providing a significant advantage for large-scale and long time-series remote sensing research.

In this study, the SPEI was selected as the indicator for monitoring meteorological aridity, while the RSEI data from the GEE platform was used to monitor ecological environment quality. The analysis focused on the spatial and temporal variations of aridity and ecological environment quality in the QTP from 2000 to 2020. By combining multi-scale meteorological aridity data, this study aims to elucidate the impact mechanism of long-term aridity on ecological environment quality in the QTP. This study will provide a theoretical foundation and scientific basis for the restoration and construction of vegetation cover, as well as the management and utilization of water resources in the QTP. Meteorological aridity affects regional ecosystems in various ways, significantly impacting vegetation growth and recovery, land degradation, and forest fire incidence.

2. Materials and Methods

2.1. Study Area

The QTP is located in the southern part of the Asian continent and covers a total area of about 2.62 million km², encompassing Tibet, Qinghai, Xinjiang, Gansu, Sichuan, and Yunnan Provinces. The average elevation of the entire plateau is greater than 4000 m above sea level, showing a trend of terrain tilting from north-west to south-east. The complexity and diversity of the terrain form the basic geological structure of the QTP (Figure 1). The natural environment of the QTP is very complex, with a rich and diverse vegetation. The main vegetation types include alpine grasslands and meadows, which form the unique ecological landscape of the plateau [24].

As a result of the high average altitude, the temperature on the plateau is relatively low, with an overall distribution characteristic of “low in the middle and high on both sides” that decreases from north-east to south-west. At the same time, precipitation shows large spatial differences, gradually decreasing from over 2000 mm in the south-eastern Shannan region to even less than 25 mm in the north-western Qaidam Basin. The QTP region is densely populated with lakes and crisscrossed with glaciers, and natural landscapes such as wetlands and swamps are widely distributed. These lakes, glaciers, and wetlands are important carriers of the country’s water resources storage. The water systems of the QTP play a vital role in supplying water resources in the region, maintaining ecological balance and supporting the livelihoods of the local population. The QTP is not only a geographically important region in China but also plays an important role in the global ecosystem [25]. Therefore, an in-depth study of the geographic, climatic, and ecological characteristics of the QTP, with particular attention to the effects of meteorological aridity on vegetation and ecosystems, is essential to the scientific understanding and effective conservation of the ecosystems of this region [26].

2.2. Data Sources

The dataset for evapotranspiration in the Chinese region is derived from the Hargreaves formula, which calculates potential evapotranspiration using monthly average, minimum, and maximum temperatures [27]. Spanning the years 1901 to 2022, this dataset boasts a spatial resolution of about 1 km by 1 km.

In contrast, the precipitation dataset for China is crafted through the Delta spatial downscaling method, utilizing the 0.5-degree global climate data from the Climate Research Unit (CRU) and the high-resolution climate data supplied by WorldClim. This dataset mirrors the temporal extent of 1901 to 2022 and maintains a similar spatial resolution of roughly 1 km by 1 km [28]. The temperature dataset was generated using the Delta spatial downscaling program. This dataset is based on the 0.5 global climate dataset published by CRU (Climatic Research Unit) and the global high-resolution climate dataset published by WorldClim. In addition, this dataset has been validated by 496 independent meteorological observation sites, and the validation results are credible [29]. The total length of the dataset is 1901~2021, and the spatial resolution is about 1 km × 1 km.

The RSEI assessment system encompasses four critical indicators: vegetation cover (greenness), soil moisture (humidity), surface temperature (heat), and building cover (desiccation). Xu et al. [20] reported that vegetation cover is measured by the Normalized Difference Vegetation Index (NDVI) [30], and soil moisture is indicated by the Wet Component of the Coma Cap Transformation (WET) [31,32]. Building cover is evaluated using the Normalized Differential Impervious Surface Index (NDBSI) [33], incorporating the Index-based Built-up area Index (IBI) and Soil Index (SI) [33]. Surface temperature is indicated by the Land Surface Temperature (LST). These indices are extensively used to evaluate the ecological status across various regions [34,35,36].

The construction of the RSEI indices was facilitated by utilizing a suite of MODIS (Moderate Resolution Imaging Spectroradiometer) products, encompassing MOD13A1, MOD11A2, and MOD09A1 datasets. These datasets were sourced from the United States Geological Survey (USGS) and have been seamlessly integrated into the Google Earth Engine (GEE) cloud computing platform, covering a period from 1 January 2001 to 31 December 2020. The MODIS satellite imagery selected for this study was specifically captured during the peak growing season, which spans from July to September. The spatial and temporal resolutions of these images are depicted in Figure 1.

2.3. Methods

2.3.1. Standardized Precipitation Evapotranspiration Index

The Standardized Precipitation Evapotranspiration Index (SPEI) is an Aridity Index based on the Standardized Precipitation Index (SPI). The SPEI can be calculated on different time scales (n = 1, 3, 6, 12), where the 1-month time scale (SPEI-1) is calculated using the precipitation and potential evapotranspiration data of the current month, and the n-month time scale is calculated using the sum of the cumulative precipitation and potential evapotranspiration data of the current month and the forward continuation of the n-1 months. The SPEI values for spring (March–May), summer (June–August), autumn (September–November), and winter (December–February) in the study area reflect the seasonal aridity conditions. The SPEI values for the 6-month time scale (SPEI-6) reflect the bi-annual aridity conditions in the study area, and the SPEI values for the 12-month time scale (SPEI-12) reflect the inter-annual variations in aridity.

SPEI has been extensively applied across various fields, including climate change analysis, environmental quality assessment, and aridity evaluation [37]. The core principle of SPEI is to gauge the level of aridity in a region by determining the difference between precipitation and potential evapotranspiration. The deviation of this difference from the mean value indicates the degree of aridity in the area. The formulation of SPEI typically involves three main steps: the calculation of potential evapotranspiration, the calculation of the difference between precipitation and actual evapotranspiration, and the normalization of the data, which are calculated using the following Formulas (1)–(12).

$$PET=16.0\times {\left({\displaystyle \frac{10T_i}{H}}\right)}^A$$

(1)

$$H=\sum _{i=1}^{12}{\left({\displaystyle \frac{T_i}{5}}\right)}^{1.514}$$

(2)

$$D_i=P_i-PE{T}_i$$

(3)

where PET is the potential evapotranspiration; $T_i$ is the mean monthly air temperature; $T_i$ is the Annual Heat Index; and A is a function of $H$, $A$ = 0.49 + 0.179$H$ − 0.0000771$H^2$ + 0.000000675$H^3$. Potential evapotranspiration in this study uses the data sets that have been obtained (

Table 1. Indices for computing RSEI.

Index	Satellite	Resolution		Level
NDVI	MOD13A1	1000 m	16 d	L3
LST	MOD11A2	1000 m	8 d	L3
WET	MOD09A1	500 m	8 d	L3
NDBSI	MOD09A1	500 m	8 d	L3

). $D_i$ is the difference between precipitation and potential evapotranspiration. The normalization of the sequence $D_i$ can be calculated by Equations (4)–(7):

$$F\left(x\right)={\left[1+{\left({\displaystyle \frac{\alpha }{x-\gamma }}\right)}^{\beta }\right]}^{-1}$$

(4)

$$\alpha ={\displaystyle \frac{\left({\omega }_0-2{\omega }_1\right)\beta }{\mathsf{\Gamma }\left(1+\frac{1}{\beta }\right)\mathsf{\Gamma }\left(1-\frac{1}{\beta }\right)}}$$

(5)

$$\beta ={\displaystyle \frac{2{\omega }_1-{\omega }_0}{6{\omega }_1-{\omega }_0-6{\omega }_2}}$$

(6)

$$\gamma ={\omega }_0-\alpha \mathsf{\Gamma }\left(1+{\displaystyle \frac{1}{\beta }}\right)\mathsf{\Gamma }\left(1-{\displaystyle \frac{1}{\beta }}\right)$$

(7)

where $F\left(x\right)$ is the probability distribution function (log-logistic); G is the factorial function; and ${\omega }_0$, ${\omega }_1$, and ${\omega }_2$ are the probability-weighted moments of $D_i$, computed as Equations (8) and (9):

$${\omega }_s={\displaystyle \frac{1}{N}}\sum _{i=1}^N(1-F_i{)}^s{D}_i$$

(8)

$$F_i={\displaystyle \frac{i-0.35}{N}}$$

(9)

where $N$ denotes the count of months included in the calculations. Following this, the subsequent stage is the normalization of the accumulated probability density, a crucial step for ensuring the data’s standardization:

$$P=1-F\left(x\right)$$

(10)

when the cumulative probability $P$ ≤ 0.5:

$$\omega =\sqrt{-2\ln \left(P\right)}$$

(11)

$$\begin{array}{c}\\ SPEI=\omega -{\displaystyle \frac{c_0+c_1\omega +c_2{\omega }^2}{1+d_1\omega +d_2{\omega }^2+d_3{\omega }^3}}\end{array}$$

(12)

where the SPEI is the Standard Precipitation Evapotranspiration Index. $c_0$, $c_1$, $c_2$, $d_1$, $d_2,$ and $d_3$ have the values of 2.515517, 0.802853, 0.010328, 1.432788, 0.189269, and 0.001308, respectively. In this study the SPEI was constructed using Arcgis 10.8 and Matlab 2021b. The criteria for classifying the degree of aridity according to the SPEI values are shown in

Table 2. Aridity classification based on Standardized Precipitation Evapotranspiration Index (SPEI).

Level	Aridity Degree	SPEI Value
1	No aridity	−0.5 < SPEI
2	Mild aridity	−1.0 < SPEI ≤ −0.5
3	Moderate aridity	−1.5 < SPEI ≤ −1.0
4	Severe aridity	−2.0 < SPEI ≤ −1.5
5	Extreme aridity	SPEI ≤ −2.0

.

2.3.2. Remote Sensing Ecological Index

Over the last 20 years of the study period, data from 2000, 2005, 2010, 2015, and 2020 were selected for this study and analyzed to produce the results. The RSEI was calculated in two main steps [35].

Firstly, the NDVI, WET, LST, and NDBSI are calculated, which represent easily interpretable greenness, humidity, heat, and dryness, respectively. Finally, the RSEI is calculated as Equation (13):

$$RSEI=f\left(NDVI,WET,LST,NDBSI\right)$$

(13)

A higher RSEI indicates higher environmental quality. The RSEI methodology improves the practicality of environmental assessment indicators by reducing the impact of subjective human judgment in assigning factor values. It also utilizes a significant amount of medium-resolution satellite data, effectively broadening its scope.

Since 2002, NDVI has been widely used for vegetation characterization, such as regional vegetation cover, plant biomass, vegetation cover, etc. It is also an important ecological indicator [38,39,40]. The calculation formula is as follows, where ${\rho }_{Red}$ and ${\rho }_{NIR}$ are the spectral reflectance in the infrared and near-infrared bands, respectively.

$$NDVI={\displaystyle \frac{{\rho }_{NIR}-{\rho }_{Red}}{{\rho }_{NIR}+{\rho }_{Red}}}$$

(14)

Temperature plays a central role in influencing plant growth and is also one of the key indicators of environmental change. In this study, grayscale values from remote sensing data were converted to degrees Celsius to determine the distribution of Land Surface Temperature across the study area. To determine this distribution, the Land Surface Temperature (LST) was calculated using the following formula:

$$LST=0.02\times D{N}_S-273.15$$

(15)

where DN_S is the gray value of the surface temperature image. Studies have indicated that the humidity component derived from MODIS’s Tasseled Cap Transformation (TCT) effectively mirrors the combined moisture of soil and vegetation in the QTP. Consequently, this study employs a method that multiplies the Tasseled Cap Transformation’s Humidity Index with surface reflectance to calculate WET, with the formula for this calculation being as follows:

$$\begin{gathered} WET=0.1147{\rho }_1+0.2489{\rho }_2+0.2408{\rho }_3+0.3132{\rho }_4 \\ -0.3122{\rho }_5-0.6416{\rho }_6-0.5087{\rho }_7 \end{gathered}$$

(16)

where ${\rho }_i$ (i = 1, 2, 3, …, n) is the surface reflectance product reflectance for each band.

As urbanization advances, the expanse of impervious surfaces grows annually, exacerbating the environmental damage inflicted by economic activities. The increasing area of exposed natural soils also plays a role in intensifying surface aridity. For this study, we determine the Normalized Difference Built-up Soil Index (NDBSI) by applying a weighted mean of the Bare Soil Index (SI) and the Impervious Surface Index (IBI), as illustrated in the formula presented in Equations (17)–(19).

$$SI={\displaystyle \frac{\left({\rho }_{SWIR1}+{\rho }_{Red}\right)-\left({\rho }_{Blue}+{\rho }_{NIR}\right)}{\left({\rho }_{SWIR1}+{\rho }_{Red}\right)+\left({\rho }_{Blue}+{\rho }_{NIR}\right)}}$$

(17)

$$IBI\&={\displaystyle \frac{\frac{2{\rho }_{SWIR1}}{{\rho }_{SWIR1}+{\rho }_{NIR}}-\left(\frac{{\rho }_{NIR}}{{\rho }_{NIR}+{\rho }_{Red}}+\frac{{\rho }_{Green}}{{\rho }_{Green}+{\rho }_{SWIR1}}\right)}{\frac{2{\rho }_{SWIR1}}{{\rho }_{SWIR1}+{\rho }_{NIR}}+\left(\frac{{\rho }_{NIR}}{{\rho }_{NIR}+{\rho }_{Red}}+\frac{{\rho }_{Green}}{{\rho }_{Green}+{\rho }_{SWIR1}}\right)}}$$

(18)

$$\begin{array}{c}NDBSI={\displaystyle \frac{SI+IBI}{2}}\end{array}$$

(19)

where ${\rho }_{SWIR1}$, ${\rho }_{Red}$, ${\rho }_{Blue}$, ${\rho }_{Green},$ and ${\rho }_{NIR}$ are the spectral reflectance in the short-wave infrared, infrared, extra-blue, extra-green, and near-infrared bands, respectively. In ArcGIS, the RSEI can be classified into five classes by equidistance, which are best, good, fair, poor, and worst, respectively [41].

In this step, all four indices, i.e., the NDVI, LST, WET, and NDBSI, are exemplified by the year 2000. The data sources for these indices are MOD13A1 for the NDVI, MOD11A2 for LST, and MOD09A1 for both the WET and NDBSI. The acquisition intervals for these images are 16 days for MOD13A1 and 8 days for MOD11A2 and MOD09A1. All MODIS satellite imagery is captured during the growing season of the QTP, which spans from July to September. After undergoing noise reduction, geometric correction, image enhancement, and grayscale processing, we obtained 4 NDVI, 7 LST, and 7 each of WET and NDBSI images for the year 2000. The average values of these indices represent the composite measure for that period. Following the acquisition of these averages, Principal Component Analysis (PCA) was employed to compute the RSEI for the year in question. After calculating the mean values for the four indices, the RSEI for each year was determined through Principal Component Analysis (PCA). This process was repeated for the years 2000, 2005, 2010, 2015, and 2020, amounting to a total of five RSEI computations over the study period. During this time, each image was cloud-masked, and the median value was taken for analysis. The cloud cover rates for the respective years were as follows: 2.37% in 2000, 3.47% in 2005, 2.00% in 2010, 0.17% in 2015, and 0.30% in 2020. All steps were re-implemented within the Google Earth Engine (GEE) platform, where the operator is required to input the vector boundaries of the study area. Upon completion of the process, the spatial characteristics and distribution of the RSEI are obtained, along with the proportional contributions of each index to the annual calculation.

2.4. Mann–Kendall Test

The nonparametric Mann–Kendall test, developed by Mann and Kendall, first saw publication in the year 1945, as a technique for detecting and forecasting time-series data [42]. It is widely used in environmental science, climatology, and other fields. The Mann–Kendall test can be used to determine whether a mutation has occurred in a climatic sequence and, if so, when such a mutation occurred [43]. The Mann–Kendall test is a widely recognized tool for detecting trends in time series data attributed to climate change. As a non-parametric method, it has the advantage of not assuming any specific distribution for the sample data, thereby being less susceptible to the influence of outliers. Consequently, it is particularly well suited for analyzing both categorical and sequential variables [44].

The formulation of the Mann–Kendall test is based on a set of Equations (20)–(23):

$$\mathrm{s}\mathrm{g}\mathrm{n}\left(A_i-A_j\right)=\left\{\begin{array}{c}1,A_i-A_j>0\\ 0,A_i-A_j=0\\ -1,A_i-A_j<0\end{array}\right.$$

(20)

$$S=\sum _{i=2}^n\sum _{j=1}^{i-1}sgn\left(A_i-A_j\right)$$

(21)

$$Var\left(S\right)={\displaystyle \frac{n\left(n-1\right)\left(2n+5\right)}{18}}$$

(22)

The Mann–Kendall test incorporates a step function, represented by “sgn”, which is applied to the time series data. In this study, the forest carbon stock represents the time series variable, denoted as $A_i$. The test statistic follows a normal distribution, indicated by “$S$”. The “$Var$” term refers to the variance calculation, which is an essential part of the test’s methodology. Lastly, “$n$” signifies the sample size of the time series data under examination.

Once the variance calculation and the normal distribution test have been conducted, the standardized test statistic, denoted as Z:

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

(23)

At a given confidence level α, the original hypothesis is invalid if |$Z$| ≥ $Z$ 1 − α/2. If $Z$ > 0, the extent of fluctuation within the time series expands; if $Z$ < 0, the range of variation of the time series decreases, then it indicates a decrease in the range of variation. In general, if |$Z$| ≥ 2.58, the trend of change is deemed to be of extreme significance. If |$Z$| ≤ 2.58, the trend is deemed significantly pronounced. If 1.96 ≤ |$Z$| < 2.58, the trend is regarded as statistically significant. If 0 ≤ |$Z$| < 1.96, the trend is considered insignificant. In this study, the ecological quality of the QTP region was assessed using the Mann–Kendall method, with calculations performed in ArcGIS Pro version 3.0.

2.5. Pearson Correlation Analysis

Using the Pearson correlation coefficient method, the correlation analysis between the Remote Sensing Ecological Index (RSEI), the Aridity Index, and the SPEI of each pixel was carried out based on the pixel scale [45]. Pearson’s correlation coefficient is employed to quantify the degree of linear association between a pair of independent variables [46]. The formula for its calculation is as follows:

$$r={\displaystyle \frac{\sum _{i=1}^n\left(X_i-\overline{X}\right)\left(Y_i-\overline{Y}\right)}{\sqrt{\sum _{i=1}^n{\left(X_i-\overline{X}\right)}^2\cdot \sqrt{\sum _{i=1}^n{\left(Y_i-\overline{Y}\right)}^2}}}}$$

(24)

In this context, r signifies the correlation coefficient, while $X_i$ and $Y_i$ denote the two variables under consideration, and $\overline{X}$ and $\overline{Y}$ represent the average values of the respective variables. $r$ is distributed in the interval [–1, 1]. In this study, we employed Pearson’s correlation analysis to investigate the relationship between forest carbon stocks and various environmental elements. The strength of a positive correlation is indicated by values approaching 1, while a strong negative correlation is signified by values near −1. Values within the range of −1 to 1 reflect the degree of correlation between the variables [47]. The computational phase of this analysis was conducted using Matlab 2021b, ensuring a rigorous examination of the correlation coefficients. When conducting correlation analysis on long-term time-series raster data with Matlab 2021b, it is crucial to be mindful of the varying resolutions present in the datasets. Data must undergo preprocessing to align the dimensions, ensuring consistency in the number of rows and columns across all frames before the analysis can be initiated.

2.6. Hurst Index Analysis

Self-similarity and long-range dependence are common phenomena in nature and are widely used in hydrology, climate, geology, and other fields. British scholar Hurst proposed the Hurst Index by studying the changes of the Nile flow time series, which is used to quantitatively characterize the continuity or long-range dependence of the time series. The Hurst Index is estimated by a variety of methods, generally using the R/S analysis method, which is more reliable. In recent years, R/S analysis has been widely used in the fields of environmental change and geoscience.

For the time-series RSEI, i = 1, 2, 3, …, n, and the time series is defined as follows:

$${\mathrm{R}\mathrm{S}\mathrm{E}\mathrm{I}}_{\left(\tau \right)}={\displaystyle \frac{1}{\tau }}\sum _1^{\tau }{\mathrm{R}\mathrm{S}\mathrm{E}\mathrm{I}}_{\left(\tau \right)}\tau =1,2,\cdots ,n$$

(25)

$$X_{\left(\tau \right)}={\sum }_{\iota =1}^{\tau }{(\mathrm{NDVI}}_{\left(\iota \right)}-{\mathrm{N}\mathrm{D}\overline{\mathrm{V}}\mathrm{I}}_{\left(\tau \right)})1⩽\iota ⩽\tau$$

(26)

$$R_{\left(\tau \right)}=\underset{1\le t\le \tau }{max}{X}_{\left(\tau \right)}-\underset{1\le t\le \tau }{min}{X}_{\left(\tau \right)}\tau =1,2,\cdots ,n$$

(27)

$$S_{\left(\tau \right)}=\sqrt{{\displaystyle \frac{1}{\tau }}\sum _{t=1}^{\tau }{\left({\mathrm{R}\mathrm{S}\mathrm{E}\mathrm{I}}_{\left(\tau \right)}-{\mathrm{R}\mathrm{S}\mathrm{E}\mathrm{I}}_{\left(\tau \right)}\right)}^2}\tau =1,2,\cdots ,n$$

(28)

For the ratio $R_{\left(\tau \right)}/S_{\left(\tau \right)}\cong R/S$, a Hurst phenomenon exists in the time series if the following relationship exists $R/S\propto {\tau }^H$. H is the Hurst exponent, which can be obtained by using least-squares fitting according to $\mathrm{l}\mathrm{o}\mathrm{g}(R/S{)}_n=a+H\times \mathrm{l}\mathrm{o}\mathrm{g}(n)$. The size of the H value can judge the persistence of the RSEI sequence, including three forms: 0.5 < H < 1, suggesting that the time series exhibits characteristics of long-term correlation within a persistent sequence, with higher values of H approaching 1 indicating greater persistence; H = 0.5, then the RSEI sequence is a stochastic sequence; 0 < H < 0.5, then the RSEI sequence has an anti-persistence, and the closer H is to 0, the stronger the anti-persistence.

3. Results

3.1. Characterization of the Spatial and Temporal Distribution of SPEI

The annual mean Standardized Precipitation Evapotranspiration Index (SPEI) values exhibit varying patterns across different time scales as depicted in Figure 2. For the SPEI-1 (as shown in Figure 2a), the values oscillate between −0.17 and 0.12, averaging at −0.07. Moving on to the SPEI-3 (Figure 2b), the range of fluctuation is from −0.12 to 0.30, with an average of 0.03. The SPEI-6 (Figure 2c) shows a broader fluctuation, spanning from −0.09 to 0.43, and has an average value of 0.20. Further extending the time scale, the SPEI-9 (Figure 2d) demonstrates a wider range, oscillating between −0.16 and 0.65, with the mean settling at 0.18. The SPEI-12 (Figure 2e) follows suit with a fluctuation from −0.25 to 0.75, averaging at 0.16. Finally, the SPEI-24 (Figure 2f), significantly influenced by long-term precipitation and potential evapotranspiration, displays the most extensive variation, ranging from −0.45 to 1.03, with an average value of 0.43.

Overall, as illustrated in Figure 3, the SPEI values of the six different time scales show more differentiated characteristics. One- and three-month-scale SPEI values show a more consistent trend, showing that the regional value of the Qaidam Basin is significantly higher than that of other regions, and the SPEI value of some regions in the south-east Hengduan Mountains is higher. Nine- and twelve-month-scale SPEI values show a more consistent trend, showing that the regional value of the Qaidam Basin is significantly higher, and some regions in the south-west are higher than other regions. The SPEI values at the 6-month scale show a more uniform distribution, with lower values in the eastern region, and the SPEI values at the 24-month scale are higher in the whole, with slightly lower values in the Qaidam Basin than in other regions.

3.2. Characteristics and Trends of Spatial and Temporal Distribution of RSEI

3.2.1. RSEI Components

The RSEI consists of four evaluation indicators, NDVI, LST, WET, and NDBSI, which represent the ease of greenness, heat, humidity, and dryness, respectively. Figure 4 presents the four indicators in time and space for the years 2000, 2005, 2010, and 2020, respectively. Over the past two decades, various environmental indices have shown distinct spatial and temporal trends. The NDVI has demonstrated a stable trend, with higher values consistently observed in the south-east, which decrease as one moves towards the north-west [48]. Similarly, the WET values have maintained a general stability, with a decline from the south-east to the north-west [49]. On the other hand, the LST has shown higher values near the northern and south-western regions, with a decrease towards the south-east. The NDBSI has exhibited a general decrease over the 20-year period, with lower values in the south-east and higher values in the north-west.

3.2.2. RSEI Spatial and Temporal Distribution

The RSEI, categorized into five classes of best, good, normal, bad, and worst according to the equidistance values in ArcGIS (Figure 5a–e), shows the spatial and temporal distribution characteristics of the RSEI in the QTP region. Figure 5f–j is the total area of best, good, normal, bad, and worst levels in each study period.

Overall, the trend analysis, along with the spatial and temporal attributes of the RSEI, revealed that the ecological status of the QTP was predominantly characterized by normal and best. The normal and best land areas were 2,214,760 km² in 2015 and reached a maximum of 2,365,378 km² in 2000 (Figure 5f). In the rest of the years, the proportion of land characterized by normal and best land area was stable at about 70%. The analysis implies that the ecological state of the QTP is generally favorable.

Over the past two decades, the spatial distribution of the RSEI across the QTP has exhibited a relatively consistent pattern, demonstrating a gradient of decline from the south-eastern to the north-western regions (Figure 5). The areas with “bad” and “worst” RSEI ratings are mainly located in the south-western and northern parts of the QTP, and the areas with “good” and “normal” RSEI ratings are mainly located in south-eastern Tibet, Central Yushu Prefecture, and northern Golog, while the RSEI “best” regions are mainly located in the eastern and southern parts of the Plateau.

The northern region has the largest and longest duration of worst in 20 years. The region is the Qaidam Basin, with high surface temperatures and scarce surface water resources. The RSEI indicates that the area is ecologically vulnerable and necessitates the implementation of suitable conservation strategies to safeguard against ecological decline. In terms of temporal characteristics, the proportion of bad area decreased from 160,422 ${\mathrm{k}\mathrm{m}}^2$ in 2000 (Figure 5f) to 146,297 km² in 2005 (Figure 5g) and has remained below 5% since then, which indicates that the management of the area in the past 20 years has been effective.

The RSEI reveals that areas classified as normal and bad are largely concentrated in the south-west. Between 2000 and 2005, there was a noticeable decrease in the proportion of bad areas, from 15.1%, as illustrated in Figure 5f, down to 13.92%, as shown in Figure 5g. This suggests a positive trend, reflecting an improvement in the regional ecological conditions over the specified timeframe. Nevertheless, the share of undesirable areas rose to 20.11% by 2010, as depicted in Figure 5h, and has since remained relatively constant. This stabilization suggests a period of ecological decline in the region commencing from 2010. The areas rated as best and good for land quality are predominantly situated in the south-eastern part of the region. In 2010, the best category constituted the smallest proportion at 0.02%, as indicated in Figure 5h. By 2015, the proportion of the best land category reached its zenith at 6.67%, as illustrated in Figure 5i. Conversely, the good category hit its lowest point in 2010 with 27.58%, according to Figure 5h, yet peaked in 2005 at 43%, as documented in Figure 5g. The higher RSEI values in the south-eastern part of the QTP indicate that the local ecological conditions are better; this can be attributed to the lower altitude in the region possibly. The area experiences limited human activity and boasts favorable hydrothermal conditions that are conducive to vegetation growth. The good growth of vegetation can in turn nourish water and better feed back to the ecosystem [50]. In general, the ecological quality of the QTP has been improving over the past 20 years, but deterioration exists in some areas.

3.2.3. RSEI Spatial Trends

Figure 5 shows that there are large spatial and temporal differences in the RSEI (Figure 5a–j), and changes in ecological conditions cannot be obtained by analyzing only its multi-period status. Therefore, GIS pro was introduced to treat long time series of remote sensing images as trend analysis. Consequently, the spatio-temporal variations of the RESI were delineated and are depicted in Figure 6.

From the spatial trend of the RSEI in Figure 6a, the QTP can be divided into three parts, i.e., western, central, and eastern. The western region encompasses the largest expanse, and, over the last two decades, the RSEI has exhibited a consistent increasing trend while most of the eastern region showed a decreasing trend. Based on the contents of

Table 3. Statistics on the trend of remote sensing ecological indices within the QTP region.

Z Value	Trend of RSEI	Percentage
−2.20 < Z ≤ −1.96	Significant decline	0.3%
−1.96 < Z ≤ 0	Nonsignificant decline	61.45%
0 < Z < 1.96	Nonsignificant increase	36.62%
1.96 ≤ Z < 2.20	Significant increase	1.55%

, it can be seen that the areas where the RSEI has shown a very significant increase account for 1.55%, predominantly located in the western region. In contrast, areas with non-significant growth constitute 36.62% and are found in the same regions as those with significant increases. The sum of the area with a non-significant decrease and highly significant decrease in RSEI was about 61.73%, which was mainly located in the eastern region. Overall, the ecological environment quality of the QTP is exhibiting a trend of divergence. From the perspective of the significance of the results, most areas do not show significant changes, while a minority of regions in the west have demonstrated a clear upward trend in the RSEI.

During the study period, the area around Qinghai Lake in the north-eastern part emerged as the region with the most severe decline in the RSEI within the entire study area, which is part of the Qaidam Basin with high surface temperatures and scarce surface and groundwater resources. Groundwater resources are scarce; thus, vegetation is scarce, and most of the land is barren [51], resulting in a declining trend of RSEI values in the region over the past 20 years. Furthermore, given the region’s ecological fragility, it necessitates the implementation of appropriate conservation measures to prevent further ecological degradation.

3.3. SPEI and RSEI Correlation at Different Scales

The annual correlation and significance of the SPEI (SPEI-1, SPEI3, SPEI-6, SPEI-9, SPEI-12, and SPEI-24) and RSEI at different time scales, based on the likelihood meta-reflecting the degree of the ecological quality of the QTP region affected by meteorological aridity, were studied.

The correlation and significance test results of different time-scale SPEIs (SPEI-1, SPEI3, SPEI-6, SPEI-9, SPEI-12, and SPEI-24) with RSEI based on the image metrics reflecting the degree of ecological environmental quality conditions affected by meteorological aridity are reflected in Figure 7. The results show that the spatial distribution of the correlation between SPEI and RSEI at 1, 3, 6, 9, 12, and 24-month time scales is relatively consistent, short-term aridity has less impact on ecological environment quality due to the presence of certain moisture in the soil, and vegetation can take up water from the soil through the root system, while long-term aridity has a greater impact on ecological environment quality.

Furthermore, Figure 7 shows that the 1, 3, 6, 9, 12, and 24-month time-scale SPEI and RSEI, which showed E-S-N-C and S-N-C, accounted for 18.83%, 21.35%, 18.18%, 19.88%, 17.12%, and 18.27% of the total study area, respectively, and the negatively correlated areas were more distributed in the south-western part of the study area. Additionally, Figure 7 shows that the area is most of the ecological quality classified as bad and worst, indicating that the ecological quality of these areas is negatively correlated with aridity conditions, which may be due to the fact that there is more glacier snow in the area. Furthermore, it shows that the occurrence of meteorological aridity will increase the glacier snow melt, which will be conducive to the growth of vegetation to promote the improvement of the quality of the ecological environment. The positively correlated regions are smaller and more scattered, and the positively correlated regions of the SPEI and RSEI at 1, 3, 6, 9, 12, and 24 time scales are mainly concentrated in the south-eastern, south-western, and northern regions of the QTP, where the sum of the S-P-C and the E-S-P-C accounted for 20.89%, 17.41%, 13.44%, 11.53%, 12.08%, and 15.56%, respectively, suggesting that the ecological integrity of these areas is significantly compromised by meteorological aridity.

3.4. Ecological Quality Sustainability Changes

Figure 8 shows the Hurst Index of the QTP. The mean value is 0.47, and the Hurst Index of the RSEI of the QTP is greater than 0.5 in 52.03% of the regions and less than 0.5 in 47.97% of the regions. This indicates that the future trend of ecological quality on the QTP is stronger in the same direction than in the opposite direction, suggesting that the future change in ecological quality on the QTP is weakly persistent, i.e., the trend of the RSEI change in most regions is the same as that in the past.

The spatial variation of the Regional Sustainable Environment Index (RSEI) and the Hurst Index provides insights into the future ecological trends (Figure 6). Regions showing sustained improvement in ecological environment quality are primarily located in the western part of the Qinghai–Tibet Plateau (QTP), while areas of sustained degradation are found in the eastern and south-eastern parts. The south-west of the study area exhibits a clear persistence in ecological environment quality, which is in stark contrast to the non-continuous characteristic observed in the south-eastern part, where human activities have had a significant impact.

4. Discussion

The evolution of ecosystem patterns in the QTP region has been caused by a variety of factors, with natural and anthropogenic factors being the main contributors to the changes, and the SPEI, which is a meteorological Aridity Index that takes into account the effects of the dispersal of signs, has been shown by many scholars to be suitable for characterizing the aridity in the study area [52,53,54]. It has been found that temperature is the most influential factor on the rate of change in ecological patterns in the QTP; therefore, the SPEI was chosen as a driver of ecological quality in this study to investigate the response of aridity events to changes in ecological quality in the QTP region. Six SPEI with different time scales were selected to reflect the aridity conditions in the QTP region on monthly, quarterly, and annual scales, etc. The study findings are largely in accordance with the conclusions from studies monitoring the ecological environment of the QTP using other aridity indices. This indicates that using the SPEI as an influencing factor to reflect changes in the ecological environment quality of the QTP region holds a high degree of credibility.

The QTP is a unique ecosystem where altitudinal gradients, temperature, and aridity lead to the slow growth and uneven distribution of natural vegetation. The spatial patterns observed in the RSEI indices are influenced by various geographical and climatic factors. The lower elevation in the south-east contributes to the favorable water and heat conditions that promote better vegetation growth, hence the higher NDVI values. As elevation increases towards the north-west, hydrothermal conditions become less conducive to vegetation development, resulting in lower NDVI values. WET values are higher in the south-east due to the influence of warm and humid monsoons, lower terrain, and higher precipitation, which facilitates water vapor accumulation and higher humidity. In contrast, the higher terrain, lower rainfall, and lower temperatures in the north-west result in lower WET values. NDBSI values are lower in the south-east due to the influence of the monsoon, which brings high rainfall, a humid climate, dense vegetation, and good ground cover. In the north-west, the dry westerly winds, low rainfall, and arid climate result in sparser vegetation and poorer surface cover, leading to higher NDBSI values.

As shown in the Results section, the distribution pattern of the RSEI is higher in the south-east, decreasing in the north-west, and is lowest in the Qaidam Basin. The reason for the low RSEI values in the Qaidam Basin region may be that the location is in the north-eastern part of the QTP, with low relief, arid climate, scanty precipitation, and scarce water resources, and the arid climatic conditions make the vegetation growth limited and the soil susceptible to erosion and salinization phenomena. Secondly, the long-term unsustainable development and utilization patterns influenced by human activities have led to serious problems such as land degradation and water resource pollution. In conclusion, the Qaidam Basin’s poor ecological environment is largely the result of a confluence of factors, including aridity, human-induced influences, and natural calamities. The higher RSEI values in the south-eastern part of the QTP indicate that the local ecological conditions are better; this is probably due to the lower altitude in the region. The area experiences less human interference, and the conditions of water and heat are favorable for the growth of vegetation. The good growth of vegetation in turn can contain water and better feedback to the ecosystem. Overall, the ecological quality of the QTP has been improving over the past 20 years, but there is deterioration in some areas, which is consistent with the general knowledge in this study field [53,55,56].

From the spatial scale, the negative correlation between the SPEI and RSEI in the QTP over the past two decades is more distributed in the south-western region, which is due to the fact that the warming and humidification of the climate of the QTP has led to a decrease in the Aridity Index of the region, and the climate is humid and the temperature is higher, which is favorable for the growth of vegetation, so the ecological quality of the environment in the south-west is improved. The positive correlation in the south-eastern part of the region is due to the fact that the eastern and south-eastern parts of the low-altitude region are topographically enclosed by high mountains. The climate in the region, shaped by the south-east monsoon, is characterized by abundant rainfall and extended periods of sunshine. This results in higher precipitation levels that surpass evaporation rates, leading to more effective soil moisture retention. The area benefits from a warmer environment that is propitious for the growth of vegetation. Prior research has established the correlation between climatic factors and the ecological quality, particularly in relation to the flourishing of vegetation in the south-east region. Overall, topographic and climatic factors led to better vegetation recovery, a higher RSEI, better environmental quality, and stronger response to aridity changes in the south-eastern QTP, while vegetation growth in the central and western regions was slow, and the response to aridity changes was weaker.

The persistence of ecological improvement in the south-west of the QTP can be attributed to the region’s susceptibility to climate change. With global warming, the temperature in this high-altitude area is expected to rise, potentially altering precipitation patterns. This climate warming could lead to the melting of alpine ice and snow, thereby increasing water resource availability, which is beneficial for ecological restoration and improvement in the region. Conversely, the non-continuous RSEI trend in the south-eastern part is likely a result of intensified human activities. These activities have disrupted the local ecological balance and biodiversity, contributing to ecosystem fragility and an increase in the Hurst Index value, indicating a less predictable and more erratic ecological state in the area.

Given the results presented previously, it can be concluded that the spatial distribution of the correlation between the SPEI and RSEI in the QTP region at the 1, 3, 6, 9, 12, and 24-month time scales is basically the same, and short-term aridity has a smaller impact on the quality of the ecological environment because there is a certain amount of water in the soil. Moreover, the vegetation can absorb the water from the soil through the root system, whereas long-term droughts have a larger impact on the quality of the ecological environment. This is because, under short-term aridity conditions, a certain amount of water may still remain in the soil, which provides the necessary water source for vegetation to absorb and sustain growth through the root system. This buffering effect of soil moisture mitigates the direct effects of short-term drought on vegetation. At the same time, vegetation may have certain adaptations to drought conditions. For example, some plants may develop deeper root systems or more efficient water uptake and utilization mechanisms in response to arid conditions. Such adaptations allow vegetation to withstand the effects of aridity in the short term.

In contrast, long-term aridity is cumulative and has a greater impact on ecological quality due to its longer duration. Climate change may lead to changes in precipitation patterns, affecting the frequency and intensity of aridity events. Prolonged aridity may be associated with long-term changes in regional climate patterns, which may have far-reaching impacts on ecosystems.

The QTP is a relatively complex ecosystem, with alpine rocks as the main terrain type in the central and western part of the region, which is susceptible to factors such as wind speed and sunshine, and thus the correlation between the RSEI and Aridity Index is relatively weak. Therefore, meteorological factors such as sunshine hours and average wind speed can be added as driving factors in the future study direction so that the changes of ecological environment quality in the QTP region can be analyzed more comprehensively. The study calculated the Aridity Index only from the perspective of meteorological data and did not consider hydrological, agricultural, or socio-economic factors. In order to further strengthen the rigor of the study, a comprehensive Aridity Index that integrates information from multiple sources can be used, which can more accurately monitor the occurrence of meteorological aridity and provide technical support for a more objective and comprehensive assessment of the extent and scope of the impact on the quality of the ecosystem and the reduction in the losses incurred.

In this study, five years of data were used for the M–K test and Hurst Index analyses, given the limitations of time and resolution access. This temporal resolution of the data may have had some impact on the precision and sensitivity of the analyses. Future research will aim to obtain annual data with higher temporal resolution for analysis with a view to obtaining more robust analytical results.

In addition, the composition of ecosystems in different study areas has different ecosystems with different natural characteristics, and different natural factors can be selected to construct ecological environment evaluation models in future studies, so it is suggested that the natural characteristics of land-use types can be combined to select appropriate ecological factors in ecological environment evaluation so as to construct a more diversified ecological environment evaluation model to better analyze the ecological environment. Overall, the use of the SPEI facilitates the evaluation and quantification of ecological quality on the QTP and improves our understanding of aridity change as it relates to the ecosystem.

5. Conclusions

In this study, the response of meteorological aridity to ecological environment quality was studied, and the temperature and precipitation data of the QTP from 2000 to 2020 were selected. By calculating the SPEI indices on the time scales of 1, 3, 6, 9, 12, and 24 months, the spatial and temporal distributions of aridity characteristics on the QTP, as well as the ecological quality of the QTP region and its trends from 2000 to 2020, were analyzed. The results were also analyzed by overlaying the results with the correlation and significance of the SPEI at different time scales, respectively, for evaluating the degree of the response of meteorological aridity to the quality of ecological environment. The main conclusions of the article are summarized as follows.

• The Aridity Index of the QTP shows a differentiated spatial and temporal distribution pattern. At shorter time scales, such as 1, 3, 9, and 12 months, the distribution of the SPEI is higher in the north-east and part of the south-west and lower in the center. Monthly, quarterly, and annual scales of SPEI values show an increasing trend, indicating that the QTP has shown a non-significant trend of becoming drier in the last two decades.
• The ecological environmental quality of the QTP has a strong spatial distribution pattern. The ecological environment quality in the south-east is significantly better, while the Qaidam Basin and the west have lower ecological environment quality indexes, but the overall trend of environmental quality is getting better. During the study period, the changes in ecological environmental quality in the QTP region were the result of a combination of natural and human factors. Therefore, in the future, the implementation of ecological projects in each region should be continued, with a focus on areas where the ecological quality has deteriorated and where it may deteriorate in the future in order to further optimize and improve the ecological protection scheme.
• There is a correlation between aridity conditions and ecological environment quality in the QTP. The areas with significant positive correlation between the RSEI and SPEI are mainly concentrated in the south-east, south-west and north of the QTP, indicating that the ecological environment quality in these areas is more seriously affected by meteorological aridity. At the same time, the correlation between the SPEI and RSEI increased with the increase in aridity time scale.