Spatial–Temporal Dynamics and Drivers of Crop Water Footprint in Xinjiang, China

Abstract

Efficient allocation and utilization of water resources are critical for the sustainable development of agriculture in arid regions, particularly those heavily reliant on irrigation. Xinjiang, one of China’s major agricultural regions, faces significant challenges in managing water resources due to its arid climate and dependence on irrigation. This study investigates the spatial–temporal dynamics of crop water footprint (CWF) and its driving factors in Xinjiang. Unlike previous studies on Xinjiang that primarily focus on total water footprint, this research emphasizes the crop blue water footprint (CWFB) to provide a more precise assessment of agricultural water allocation and consumption. Using the CROPWAT 8.0 model, the CWF of 14 prefectures in Xinjiang were analyzed for the period 2000–2020. Focusing primarily on the crop blue water footprint (CWFB), the study employed the Logarithmic Mean Divisia Index (LMDI) model to identify key drivers and their mechanisms. Results reveal that Xinjiang’s average annual CWF is 179.02 Gm³, with CWFB contributing 90.22% and the crop green water footprint (CWFG) accounting for. 10.05%. The CWFB showed an initial increase followed by stabilization, with Southern Xinjiang being the largest contributor, trailed by Northern and Eastern Xinjiang. Among the 14 prefectures, the top seven accounted for 90.46% of CWFB. Cotton, wheat, and maize were the major crops, comprising 47.80%, 23.14%, and 21.45% of the total blue water footprint, respectively. This study identifies the dominant role of economic effect and water use efficiency effect in driving changes in CWFB through its analysis of the driving factors. Understanding the spatial–temporal changes and key drivers of blue water consumption helps regions adjust cropping structures and agricultural water resource allocation patterns to ensure sustainable agricultural development. The findings not only offer valuable implications for policymakers and stakeholders in Xinjiang but also provide references for other arid and semiarid regions facing similar challenges in agricultural water resource management.

1. Introduction

Agriculture is the largest water-consuming sector globally, accounting for more than 70% of global water use [1]. Meanwhile, the spatial and temporal distribution of water resources is uneven worldwide, and this issue is more pronounced in arid and semi-arid regions [2]. According to relevant studies, the water shortage in the agricultural sector is expected to reach approximately 7000 Gm³ by 2030 and may worsen further [3]. China has abundant agricultural resources, with more than 120 M hm² of arable land, but it is also one of the most water-scarce countries in the world [4,5]. Water scarcity has become a significant factor constraining agricultural development in arid regions of China. Therefore, scientifically and reasonably evaluating agricultural water resources is of great practical importance for improving the efficiency of agricultural water management and promoting sustainable agricultural development.

Hoekstra [6] introduced the concept of the water footprint in 2003. Since then, the water footprint has become a crucial research method for the rational management and evaluation of regional agricultural water resources, offering a valuable perspective for the scientific assessment of water resource use in agricultural production [7]. In agriculture, the crop water footprint is classified into three components: blue, green, and grey. The blue and green water footprint represent irrigation water and natural precipitation consumed by the crop through evapotranspiration from the field, respectively. In contrast, the grey water footprint refers to the volume of water used by the crop to absorb pollutants under natural background concentrations and current ambient water quality standards [8]. Since blue and green water footprints are directly linked to crop growth, the grey water footprint, which is associated with water pollution, remains relatively underexplored in agricultural water use research [9]. Due to the direct association between blue and green water footprint and crop growth, research on the grey water footprint in agricultural water studies has been relatively limited. Currently, numerous researchers have extensively studied the water footprint of crop production across various spatial and temporal scales using different models [10,11,12,13].

In terms of the decomposition of drivers, Kim et al. [14] used econometric panel data techniques to assess the impact of macroeconomic, climatic, and agronomic drivers on agricultural crop water footprints in 43 African countries (2002–2016). When assessing the sustainability of regional agricultural water usage, Tamea et al. [15] identified key drivers of virtual water flows in international trade, including population, GDP, and geographical distance, and found that the explanatory power of these factors, as indicated by the adjusted R-squared coefficient of the gravity-law model, increased over time. Zhuo et al. [16] revealed the impact of changes in the agricultural water footprint of the Yellow River Basin on the scarcity of local water resources. Fu et al. [17] used the Logarithmic Mean Divisia Index (LMDI) model to decompose the grey water footprint efficiency of the Yangtze River Basin into six key contributing drivers (the economic effect, the industrial structure effect, the efficiency effect of capital output, the capital deepening effect, the economic activity effect, and the per capita grey water footprint intensity effect).

Currently, both Structural Decomposition Analysis (SDA) and the Index Decomposition Analysis (IDA) are widely employed by scholars to explore the drivers [18,19,20]. SDA necessitates data from input–output tables and is primarily employed in the study of energy-related CO₂ emissions, though its application in water resource research remains limited [21,22,23]. In contrast, IDA is more flexible and can be applied more broadly, as it allows for decomposition upon acquiring comprehensive driver data [24]. Among the IDA methods, the LMDI, developed by Ang and Choi [25], effectively distributes residuals evenly across sectors, producing decomposition results without residual terms, thereby enhancing interpretability [26].

A great deal of research has been conducted on the drivers of regional water footprints; however, current research on the water footprint of arid zones primarily focuses on the total water footprint, with relatively limited research specifically targeting the blue water footprint and its drivers. Xinjiang Uygur Autonomous Region (Xinjiang), situated in the arid northwestern part of China, serves as a crucial agricultural production base. However, despite having the highest agricultural water consumption in the country [27], its water resources account for only 4% of the national total [28]. In addition, Xinjiang has vast desert areas with limited precipitation and intense evaporation, with an average annual rainfall of 146.4 mm and an average annual evaporation ranging from 1600 to 2300 mm [29]. The scarcity of water resources and their uneven spatial and temporal distribution significantly constrain agricultural development in Xinjiang. In comparison to other provinces, Xinjiang, characterized by intense sunshine and high evapotranspiration rates, heavily depends on irrigated agriculture. Therefore, research and management of agricultural irrigation water in this sector is of paramount importance.

This study investigates the spatial–temporal dynamics of the crop water footprint in Xinjiang. In contrast to previous studies that primarily focus on total water footprint, this research places particular emphasis on blue water footprint, offering a more accurate assessment of agricultural water allocation and consumption in the region. The CROPWAT 8.0 model was utilized to calculate the crop water footprint at regional and prefectural scales, using meteorological and crop statistics from 2000 to 2020. The objectives of this study are to (1) elucidate the temporal changes in crop water footprint, particularly blue water footprint, and crop composition in Xinjiang from 2000 to 2020; (2) analyze the spatial and temporal variations in blue water footprint across the three major regions and fourteen prefectures of Xinjiang; and (3) decompose the blue water footprint drivers using the LMDI model. This study can, on the one hand, provide a reference for agricultural water allocation and inter-regional water transfers in various regions of Xinjiang, and offer a basis for water conservation strategies and the formulation of regional irrigation policies, thus promoting the long-term sustainability of agricultural water management in Xinjiang and similar arid regions. On the other hand, analyzing the influencing factors can provide a foundation for the development of agricultural policies.

2. Materials and Methods

2.1. Study Area

Xinjiang is located deep inland on the Asian–European continent and is characterized by a temperate continental arid climate, featuring dry conditions, limited rainfall, and high evaporation rates. Covering approximately one-sixth of China’s total land area, its water resources account for only 2.5% of the national total [30]. Xinjiang is endowed with abundant light and heat resources, receiving an average of 2808.6 h of sunshine annually and experiencing significant diurnal temperature variations, thereby creating favorable conditions for agricultural production. The region’s terrain is diverse, with the Junggar Basin and Tarim Basins situated between the three major mountain ranges: the Altay, Tianshan, and Kunlun Mountains. The distribution of water resources follows a spatial pattern, with higher levels in the north, west, and mountainous regions, and lower levels in the south, east, and plains.

In this study, 14 prefectures of Xinjiang were selected and categorized into three major regions according to the topography of the mountain ranges (Figure 1): Northern Xinjiang (comprising Urumqi City (Urumqi), Changji Hui Autonomous Prefecture (Changji), Tacheng Prefecture (Tacheng), Karamay City (Karamay), Bortala Mongol Autonomous Prefecture (Bortala), Ili Kazakh Autonomous Prefecture (Ili), and Altay Prefecture (Altay)), Southern Xinjiang (comprising Kizilsu Kirghiz Autonomous Prefecture (Kizilsu), Aksu Prefecture (Aksu), Kashgar Prefecture (Kashgar), Hotan Prefecture (Hotan), and Bayingolin Mongol Autonomous Prefecture (Bayingolin)), and Eastern Xinjiang (comprising Hami city (Hami) and Turpan City (Turpan)).

By 2020, agricultural water consumption in Xinjiang is projected to account for 90.9% of the total water consumption. Agriculture is the primary water-consuming sector in Xinjiang, with crop cultivation areas of cotton, corn, wheat (including both spring and winter varieties), alfalfa, and rice constituting 87.3% of Xinjiang’s total crop sown area. Therefore, this study aims to analyze the spatial–temporal dynamics and determinants of the water footprint associated with these five major crops in Xinjiang between 2000 and 2020.

2.2. Data Sources

Meteorological data for this study were obtained from the China Surface Meteorological Data Daily Value Dataset provided by the National Meteorological Information Centre (http://www.nmic.cn (accessed on 1 January 2025)), including sunshine hours, average wind speed, relative humidity, minimum temperature, maximum temperature, etc. These meteorological parameters were provided by over 100 meteorological stations within Xinjiang. Crop area and yield data, economic and population data, and other influencing drivers were sourced from the Xinjiang Statistical Yearbook (https://tjj.xinjiang.gov.cn (accessed on 1 January 2025)), and water resources data were obtained from the Xinjiang Water Resources Bulletin (https://slt.xinjiang.gov.cn (accessed on 1 January 2025)). All data cover the period from 2000 to 2020.

2.3. Crop Water Footprint

This study primarily focuses on the blue and green water used for crop growth, excluding the grey water footprint related to water pollution. The blue and green water footprints of major crops (cotton, maize, wheat, rice, and alfalfa) were calculated for each prefecture using the CROPWAT 8.0 model developed by the FAO, following the recommendations in the Water Footprint Assessment Manual [8]. The crop blue water footprint is classified into five levels, ranging from level 1 to level 5, in increasing order for statistical analysis, based on the study by Xu et al. [31].

(1)$CWFT={\displaystyle \sum _{i=1}^{14}CW{F}_i}$

(2)$CWF=CWFG+CWFB$

where $CWFT$ represents the volume of total water footprint of crops in each prefecture; $CWF$ represents the volume of water footprint of each crop type in each prefecture; $CWFG$ represents the green water footprint of crops, which corresponds to the volume of precipitation consumed by crop growth that is not converted into runoff; and $CWFB$ represents the blue water footprint of crops, which corresponds to the volume of surface water or groundwater consumed during crop production, all in Gm³.

(3)$CWFB=10\times E{T}_b\times A$

(4)$CWFG=10\times E{T}_g\times A$

(5)$E{T}_b=\max (0,E{T}_c-P_e)$

(6)$E{T}_{\mathrm{g}}=\min (P_e,E{T}_c)$

where $E{T}_b$ (mm) represents blue water evapotranspiration, $E{T}_g$ (mm) represents green water evapotranspiration, each representing the portion of crop evapotranspiration derived from irrigation water and effective rainfall; $A$ (kha) represents the area of crop cultivation; $P_e$ (mm) represents effective precipitation, calculated in ten-day intervals, using the formula recommended by the USDA Soil Conservation Service (Equation (7)); and $E{T}_c$ (mm) represents crop evapotranspiration. In this study, all evapotranspiration values are calculated over ten-day periods.

(7)$P_e=\left\{\begin{array}{l}P\times {\displaystyle \frac{125-0.6\times P}{125}}\\ {\displaystyle \frac{125}{3}}+0.1\times P\end{array}\right.\begin{array}{l}P\le 250/3\\ P>250/3\end{array}$

(8)$E{T}_c=K_{\mathrm{c}}\times E{T}_0$

where $P$ (mm) is ten-day precipitation; $K_c$ is the crop coefficient, dimensionless; and $E{T}_0$ (mm) is the reference crop evapotranspiration, calculated using the Penman–Monteith formula:

(9)$E{T}_0={\displaystyle \frac{0.484\Delta (R_n-G)+\gamma \frac{900}{T+273}{U}_2(e_s-e_a)}{\Delta +\gamma (1+0.34U_2)}}$

where $\Delta$ (kPa/°C) is the slope of the vapor pressure curve; $\gamma$ (kPa/°C) is the psychrometric constant; $U_2$ (m/s) is the wind speed measured at 2 m height; $R_n$ (MJ/m² day) is the net radiation at the crop surface; $G$ (MJ/m² day) is the soil heat flux density; $T$ (°C) is the average air temperature; $e_s$ (kPa) is the saturation vapor pressure; and $e_a$ (kPa) is the actual vapor pressure.

2.4. Trend Analysis Methods

The Mann–Kendall statistical test is a non-parametric statistical method [32] that does not require samples to obey a specific distribution and not be disturbed by a few outliers [33]. It requires only independent data to analyze trends and monitor changes in the time coefficient [34]. It can effectively identify whether the sample sequence exhibits a mutation, including the number and timing of the mutation [35].

A positive UF value in the test results indicates an upward trend in the data series, whereas a negative UF value signifies a downward trend [34]. A comparative analysis was performed at the 95% confidence level. If the UF exceeds Z = 1.96, it indicates a significant increase, whereas if it falls below Z = −1.96, it signifies a significant decrease. The intersection of UF and UB indicates a mutation in the sample sequence at that specific time [36].

2.5. Decomposition of Drivers

To further elucidate the impact of various drivers on CWFB, Kaya’s constant equation was used for the decomposition of CWFB drivers and their analysis using the LMDI model, thereby identifying the key drivers influencing CWFB.

2.5.1. Kaya Identity

Based on the principle of the Kaya identity, the CWFB is decomposed into five drivers, as shown in Equation (10):

(10)$CWFB=t\times ws\times e\times wue\times P$

(11)$CWFB={\displaystyle \frac{CWFB}{CW}}\times {\displaystyle \frac{CW}{WU}}\times {\displaystyle \frac{GDP}{P}}\times {\displaystyle \frac{WU}{GDP}}\times P$

where $CW$ is agricultural water consumption; $WU$ is total water consumption; $GDP$ is Gross Domestic Product; $P$ is population; and $WT$ is total water resources. The specific meanings of the five drivers are:

$t={\displaystyle \frac{CWFB}{CW}}$ is the ratio of CWFB to agricultural water consumption, which indicates the level of irrigation technology.

$ws={\displaystyle \frac{CW}{WU}}$ is the ratio of agricultural water consumption to total water consumption, which reflects the structure of local water consumption.

$e={\displaystyle \frac{GDP}{P}}$ is the ratio of GDP to population, reflecting the level of local economic development.

$wue={\displaystyle \frac{WU}{GDP}}$ is the ratio of total water use to GDP, reflecting local water use efficiency.

$P$ is population size.

2.5.2. LMDI Model

The Logarithmic Mean Divisia Index (LMDI) is one of the Index Decomposition Analysis (IDA) methods that is based on the concept of logarithmic averaging and decomposes the overall change into the contributions of individual factors. This method allows for a complete decomposition of the objectives, and the results of the disaggregation without residual terms are not only easy to interpret, but also maintain a high degree of consistency among the decomposed factors [37]. In the LMDI model, the time step for index decomposition analysis is determined by the time interval of the data used, which in this study is annual. To quantify the impact of the various influencing drivers, the additive form of the LMDI model is applied to Equation (11) to derive Equation (12):

(12)$\Delta CWFB=\Delta CWF{B}^t+\Delta CWF{B}^{ws}+\Delta CWF{B}^e+\Delta CWF{B}^{wue}+\Delta CWF{B}^p$

According to the LMDI model, the base year CWFB is t₁ and the CWFB in year t is t₂, combined with the driver decomposition formula, the drivers influencing the change in water footprint are decomposed as follows:

(13)$\Delta CWFG{B}^t={\displaystyle \frac{CWF{B}_{t_2}-CWF{B}_{t_1}}{\ln CWF{B}_{t_2}-\ln CWF{B}_{t_1}}}-\ln {\displaystyle \frac{t{f}_{t_2}}{t{f}_{t_1}}}=TF$

(14)$\Delta CWF{B}^{ws}={\displaystyle \frac{CWF{B}_{t_2}-CWF{B}_{t_1}}{\ln CWF{B}_{t_2}-\ln CWF{B}_{t_1}}}-\ln {\displaystyle \frac{ws{f}_{t_2}}{ws{f}_{t_1}}}=WSF$

(15)$\Delta CWF{B}^e={\displaystyle \frac{CWF{B}_{t_2}-CWF{B}_{t_1}}{\ln CWF{B}_{t_2}-\ln CWF{B}_{t_1}}}-\ln {\displaystyle \frac{e{f}_{t_2}}{e{f}_{t_1}}}=EF$

(16)$\Delta CWF{B}^{wue}={\displaystyle \frac{CWF{B}_{t_2}-CWF{B}_{t_1}}{\ln CWF{B}_{t_2}-\ln CWF{B}_{t_1}}}-\ln {\displaystyle \frac{wu{f}_{t_2}}{wu{f}_{t_1}}}=WUEF$

(17)$\Delta CWF{B}^P={\displaystyle \frac{CWF{B}_{t_2}-CWF{B}_{t_1}}{\ln CWF{B}_{t_2}-\ln CWF{B}_{t_1}}}-\ln {\displaystyle \frac{p{f}_{t_2}}{p{f}_{t_1}}}=PF$

Equations (13)–(17) represent the technology effect (TF), water use structure effect (WSF), economic effect (EF), water use efficiency effect (WUEF), and population effect (PF), respectively.

We summarized the entire study process in a flowchart (Figure 2):

3. Results

3.1. Temporal Dynamics of Crop Water Footprint

The average CWFT, CWFG, and CWFB in Xinjiang between 2000 and 2020 were 198.49 Gm³, 179.02 Gm³, and 19.47 Gm³, respectively. CWFT increased significantly during this period, rising from 112.80 Gm³ in 2000 to 278.13 Gm³ in 2020, corresponding to a growth rate of 146.57% (Figure 3). The CWFG reached its minimum value of 104.64 Gm³ in 2001 and its peak value of 341.82 Gm³ in 2017. The CWFB largely drove these increases. The average share of CWFB reached 90.19%, while the share of CWFG remained stable at 10.05% over an extended period. Consequently, the subsequent analysis focuses on CWFB.

The annual trend of CWFB in Xinjiang over the study period is comparable to that of CWFT in Xinjiang, demonstrating an increasing trend followed by a decreasing trend. It increased from 104.43 Gm³ in 2000 to 258.58 Gm³ in 2020, with a growth rate of 147.61%. The minimum value was 93.50 Gm³ in 2001, and the maximum value was 313.91 Gm³ in 2017. The year with the largest increase was 2014, showing a 31.86% increase relative to 2000.

Combined with the Mann–Kendall statistical test for CWFB from 2000 to 2020, the UF curve has been demonstrating an upward trend following a decrease in 2001 (Figure 4). The UF value has exceeded 0 since 2003 and has remained above the significance threshold since 2006, which suggests that CWFB has exhibited a significant growth trend during the study period. The UF and UB statistic curves intersect in 2008, indicating a significant change in the sample sequence. That is, the CWFB grew relatively slowly before 2008, while it grew more rapidly after 2008. We therefore divide the study period into three distinct phases: the slow growth phase (2000–2008, T1), the rapid growth phase (2008–2014, T2), and the steady development phase (2014–2020, T3).

A comparison of CWFB contributions among different crop types demonstrated that cotton had the highest CWFB contribution (Figure 5). The crop that primarily contributed to the growth of CWFB in Xinjiang during the T1 stage was cotton, with its water footprint increasing from 42.93 Gm³ to 74.91 Gm³. By 2008, cotton’s share of CWFB reached 49.95%, a change primarily driven by the significant expansion of cotton cultivation area. The share of wheat’s CWFB gradually decreased from 31.29 % to 23.85 % during the same period.

In the T2 phase, cotton’s CWFB fluctuated around 73.97 Gm³ from 2008 to 2013, but in 2014, cotton’s CWFB increased sharply by 92.21% compared to 2013. In contrast, the CWFB of wheat and maize increased steadily from 2008 to 2014, rising from 35.76 Gm³ and 27.50 Gm³ to 51.02 Gm³ and 50.48 Gm³, respectively. This growth was primarily due to a significant increase in the cultivated area of both crops. By 2014, the CWFB of wheat and maize had increased by 19.41% and 19.20%, respectively.

The CWFB of various crops shows stable development during the T3 phase, with the exception of cotton, whose CWFB rose to 165.74 Gm³ in 2017 before stabilizing around 137.11 Gm³. Rice and alfalfa exhibit relatively low CWFB values between 2000 and 2020, with rice’s CWFB declining from 4.38 Gm³ in 2000 to 2.24 Gm³ in 2020, while alfalfa’s CWFB gradually increases from 4.99 Gm³ in 2000 to 11.16 Gm³ in 2020.

3.2. Spatial Dynamics of Crop Water Footprint

By analyzing the CWFB of different prefectures in the four nodal years across the three phases (2000, 2008, 2014, and 2020), as well as the average annual CWFB for each prefecture from 2000 to 2020, we observed significant prefectural variations in CWFB growth (Figure 6). In 2000, CWFB in most prefectures was at level 1, with only Aksu, Kashgar, and Hotan exhibiting CWFB greater than 10 Gm³. Among these, Aksu had the largest CWFB in level 3, at 29.31 Gm³.

In 2008, fifty percent of the fourteen prefectures had CWFB in level 2 or higher, with Kashgar and Aksu both reaching level 3. Kashgar had the largest CWFB, at 34.34 Gm³, accounting for 22.90% of the total CWFB. Aksu, with a CWFB of 29.60 Gm³, exhibited the most enormous rate in CWFB compared to 2000, with an increase of 53.93%. By 2014, eight prefectures had CWFB in level 2 or higher. Among these, three prefectures (Changji, Tacheng, and Bayingolin) had CWFB in level 3, and two prefectures (Aksu and Kashgar) had CWFB in level 4, with values of 49.77 Gm³ and 59.03 Gm³, respectively. Together, these two prefectures accounted for 41.39% of the total CWFB in Xinjiang in 2014.

Compared to 2008, CWFB growth rates of the five prefectures that reached level 3 and level 4 in 2014 were 74.54%, 147.12%, 119.30%, 68.14%, and 71.90%, respectively. In 2020, the CWFB of these prefectures remained largely stable, with only Kashgar and Aksu showing a slight increase, reaching level 5 (62.02 Gm³ and 60.53 Gm³, respectively). Overall, the annual average CWFB of prefectures in Xinjiang with larger areas is generally in level 2 or higher, although half of the prefectures have annual average CWFB in level 1, but these are smaller in area. The reasons for the relatively small CWFB in these perennial leading regions vary: Urumqi and Bortala prioritize economic development, natural resources, and tourism, while Altay and Kizilsu are characterized by high mountainous terrain, which is unsuitable for large-scale agriculture. Meanwhile, Hotan and Turpan are dominated by desert landscapes, with sparse populations and very limited agricultural cultivation.

Figure 7 shows the annual average CWFB and the annual average CWFB for various crops in the three major regions and fourteen prefectures of Xinjiang. From the three major regions, Southern Xinjiang, Northern Xinjiang, and Eastern Xinjiang accounted for 63.65%, 33.94%, and 2.41% of the total annual average CWFB, respectively. The CWFB of the top seven prefectures accounted for 90.46% of the total annual average CWFB in Xinjiang, namely, Kashgar, Aksu, Bayingolin, Changji, Tacheng, Ili, and Hotan.

Considering different crop categories, CWFB for cotton, wheat, maize, alfalfa, and rice account for 47.80%, 23.14%, 21.45%, 5.42%, and 2.81% of the total CWFB, respectively. Of the total CWFB for cotton, 69.02% is contributed by Aksu, Kashgar, and Bayingolin. The CWFB of wheat is distributed at 58.50% in Southern Xinjiang and 38.93% in Northern Xinjiang, with Kashgar being the most prominent contributing prefecture (24.53%). The main contributing prefectures for alfalfa’s CWFB are Kashgar (23.96%), Hotan (22.56%), and Altay (14.98%). For rice, Aksu is the primary contributing prefecture (28.88%), while Hami and Turpan did not grow rice during 2000–2020.

3.3. Analysis of Drivers

Using the LMDI model, we decompose CWFB into five drivers (EF, TF, WUEF, WSF, and PF) and calculate the contribution values (Figure 8) and average contribution values (Figure 8) of these drivers for the period from 2000 to 2020. Among the changes in CWFB, EF contributes a significant positive value during most periods and plays a key role in driving its growth. In particular, it shows a notable positive influence in 2007–2008 (33.9 Gm³), 2009–2010 (36.77 Gm³), 2011–2012 (40.76 Gm³), and 2017–2018 (86.9 Gm³).

WUEF contributed substantially to the reduction of CWFB in most periods, indicating that inefficiencies in water use have been a significant constraint on the growth of CWFB. Especially during the periods 2007–2008 (−31.94 Gm³), 2009–2010 (−41.85 Gm³), and 2017–2018 (−95.46 Gm³), the low water use efficiency effect resulted in considerable decreases in CWFB. The persistently low levels of WUEF are directly hindering the sustainable development of agricultural water footprints and restricting overall agricultural productivity. The contribution of TF varied between positive and negative values throughout the 2000–2020 period, with a significant positive effect on CWFB in most years, particularly in 2013–2014 and 2016–2017, when contributions reached 86.39 Gm³ and 64.52 Gm³, respectively. PF consistently contributed positively in most periods, albeit at a relatively small scale, suggesting that population growth in some regions during this period had a minor effect. The contribution of WSF varied between positive and negative values in most periods, but was relatively small compared to the other drivers, indicating that its effect on CWFB was not significant.

Overall, the mean value of the total contribution to CWFB change from 2000 to 2020 is 7.74 Gm³, with the mean contributions of EF, TF, and P being 17.91 Gm³, 4.94 Gm³, and 1.99 Gm³, respectively. These three drivers played a positive role in driving CWFB (Figure 9). The mean contributions of WUEF and WSF were −16.91 Gm³ and −0.22 Gm³, respectively, indicating that both acted as reverse drivers for CWFB. The contributions of EF and WUEF accounted for 42.67% and 40.29% of the total, respectively, suggesting that one of these two drivers was the primary driver of CWFB change. TF and P were secondary drivers, while the inhibitory effect of WSF was negligible.

4. Discussion

4.1. Spatial–Temporal Heterogeneity of CWFB in Xinjiang

From 2000 to 2020, the CWF of Xinjiang exhibited an increasing trend followed by stabilization. Additionally, the CWFB and CWFG demonstrated similar trends, with the average share of CWFB reaching 90.22% during the study period. This aligns with the trend of the share of blue–green water in Xinjiang from 1996 to 2015, as calculated by Cao et al. [4], and corresponds to the long-term average share of water used for primary production in Xinjiang, which is 93.37%. During this period, cash crops such as fruits and cotton increasingly dominated Xinjiang’s agricultural landscape. Concurrently, the proportions of grain crops like wheat and maize also grew, as documented in studies [12,38,39].

In Liu et al.’s [10] study, it was noted that food crops are predominantly reliant on green water. Specifically, from 2000 to 2020, the CWFG of maize in the northwest region of China accounted for more than 50% of maize’s CWF. However, in our study, the green water footprint in Xinjiang was significantly lower, possibly due to its greater reliance on irrigation compared to other provinces in the northwest region of China. Studies by Cui et al. [40] and Zheng et al. [41] demonstrate that regional water use for crop production is closely related to planted area and total production. Additionally, the rapid expansion of planted areas for crops that require low agricultural labor and irrigation inputs but offer high economic profitability contributes to the growth of the water footprint of crops.

Between 2000 and 2020, support from various policies led to a significant increase in agricultural productivity and water usage in Northwest China [39]. During the T1 phase, starting after 2000, China implemented the Western Development Strategy, which resulted in a series of financial supports for agricultural cultivation, the construction of water conservancy infrastructure, and the rapid advancement of crop cultivation technologies [42,43]. During the T2 phase around 2010, China’s provincial and municipal support initiatives in Xinjiang further promoted the expansion of arable land and the development of agriculture-related technologies and facilities [44]. During the T3 period, to optimize the structure of agricultural planting and reduce the cultivation of rice and corn, which have large stocks, the General Office of the Ministry of Agriculture of China issued the ’Key Points of Planting Industry Work’ in 2018. This document outlined the national planting industry’s work for that year and explicitly proposed a target to reduce the rice planting area by 10 million mu by 2020, while also aiming to reduce the cultivation areas of corn, wheat, and other crops. These policies have significantly impacted crop acreage changes in Xinjiang, with both CWF and CWFB declining in 2018 compared to 2017.

There are notable regional variations in CWFB. Southern Xinjiang exhibits the highest average CWFB, with Aksu and Kashgar accounting for 44.18% of the total. Zhang et al. [45] calculated the three types of agricultural water footprint across five regions in Southern Xinjiang, revealing that the extensive adoption of irrigation-based agriculture has led to the most notable increase in blue water. The main reason for this increase is the planting of large areas of wheat and cotton, which is consistent with the results of the present study. Southern Xinjiang is characterized by a vast area of arable land, yet it faces water scarcity. In contrast, Northern Xinjiang, while endowed with relatively abundant water resources, has limited arable land. Economic growth driven by industrial development in this region provides favorable conditions for agricultural expansion [46]. The main contributors to CWFB in Northern Xinjiang are Changji, Tacheng, and Ili. Altay and Ili, the two regions richest in total water resources in Xinjiang, account for 12.45% and 18.76% of the region’s total water resources, respectively, yet consume only 1.97% and 8.12% of CWFB. Eastern Xinjiang has the lowest CWFB among the three major regions, with both limited water resources and the smallest area of crop cultivation.

4.2. Drivers of CWFB

We analyzed the influence of five drivers on CWFB in Xinjiang using the LMDI model, with the following ranking of their contributions: economic effect > water use efficiency effect > technology effect > population effect > water use structure effect. The study by Cui et al. [47] identified the economic effect as a key driver of crop water footprint growth in Northwest China. In our study, the economic effect accounted for nearly 50% of CWFB in Xinjiang, suggesting that indicating that the growth of the economy substantially contributes to the increase of CWFB.

The water use efficiency effect remains negative for a significant portion of the study period, suggesting that low water use efficiency serves as a constraint on the growth of the CWFB in most years. On one hand, although Xinjiang possesses a substantial number of energy and modern service industries, the economic output per unit of water use in agriculture is relatively low, which partially diminishes the population’s incentives to engage in agricultural activities and hampers CWFB growth. On the other hand, this may also be attributed to limitations in resource efficiency and environmental constraints. Zhao et al. [48] also indicated that the water use efficiency effect is the largest negative factor contributing to the reduction of the overall water footprint in China’s agricultural sector from 1990 to 2009. Additionally, the impact of water-saving technologies is also noteworthy, as the relatively low level of irrigation efficiency contributes to the prolonged dominance of agricultural water use, which aligns with the relatively small positive contribution from the technology effect. The population of Xinjiang has been steadily increasing; however, the positive contribution of the population effect is relatively small, which may be attributed to Xinjiang’s vast area and low population density, and Zhang et al.’s [49] study of the Yangtze River Delta of China, also showed that the increase in population corresponded to the increase in water footprint.

The technology effect contributed approximately 10% to the overall value. From 2000 to 2020, the adoption and expansion of water-saving irrigation technologies led to an increase in blue water use per unit of agricultural water consumption, rising from 0.31 to 0.52. This trend suggests that technological advancements are crucial for the increase in blue water use for crops. Liu et al. [10], in their study on crop yield water footprint in northwestern China, found that the blue water footprint per unit of yield decreased annually from 2000 to 2020, indicating a consistent improvement in irrigation water use efficiency effect. This finding is further supported by the fact that the technology factor t (the ratio of blue water footprint to agricultural water use) also showed an annual increase.

4.3. Recommendations and Limitations

On the whole, Xinjiang’s agricultural development is stable, but given the uneven spatial and temporal distribution of water resources and limited water availability, the region faces certain development risks. Based on the conclusions of this study, the following recommendations are made:

• 1.. Economic development should be prioritized, with a focus on expanding not only traditional agriculture but also high value-added agricultural sectors such as organic farming, specialized crop production, and agricultural processing industries. For example, the cotton industry, which plays a pivotal role in Xinjiang’s agricultural economy, can be further developed by promoting value-added products like cotton textiles, high-quality cottonseed oil, and cotton-based bioproducts. This approach would not only enhance the economic value derived from cotton but also foster a more diversified agricultural economy, improving both profitability and sustainability. Additionally, leveraging relevant policy support, efforts should continue to adjust and optimize the industrial structure. This includes reducing the share of agricultural output within the overall economic composition, which in turn will reduce agricultural water consumption relative to economic water use. It is crucial to enhance the development of industries, manufacturing, and services, thereby achieving coordinated development across all sectors. Furthermore, economic growth should stimulate technological advancements and population dynamics, creating a virtuous cycle that promotes mutual reinforcement. A strong emphasis should be placed on developing water-saving irrigation technologies and constructing related infrastructure, while strengthening agricultural water management. The focus should shift from expanding cultivated land area to improving production efficiency, thus fostering the sustainable development of agriculture in Xinjiang.

• 2.. The uneven spatial and temporal distribution of water resources is a critical constraint on agricultural development in Xinjiang. Consequently, water transfer initiatives within the region should be vigorously promoted. Given the varying water resource stocks and agricultural development conditions across different regions, it is essential to implement regional resource scheduling and coordination to achieve the alignment of spatial–temporal water use and ensure the sustainable development of agriculture [41]. For instance, Aksu and Kashgar, which account for only 7.82% and 9.08% of Xinjiang’s total water resources, respectively, generate 20.23% and 23.86% of the crop blue water. In contrast, Yili, which holds an average of 18.76% of Xinjiang’s total water resources, produces exactly 18.76% of the crop blue water. On one hand, this suggests significant potential for agricultural development in Yili. On the other hand, the surplus water resources in Yili could be transferred to southern Xinjiang’s prefectures and states, thereby promoting a more balanced agricultural development across the region.

This study has several limitations, primarily related to the calculation of water footprints and the construction of the model. First, the crop types considered in the calculation of water footprints do not cover all crops grown in the study area, leading to some statistical errors. Additionally, the calculation of water footprints in this study excludes the grey water footprint, which introduces potential inaccuracies in the proportion analysis. Regarding the model construction, due to data availability constraints, the selection of driving factors may not fully capture all the aspects influencing CWFB. Future research should expand the range of crops considered in the calculation and incorporate the grey water footprint. Furthermore, future studies should adopt more comprehensive and suitable approaches to assess the sustainability of irrigated agriculture in Xinjiang, providing more targeted recommendations for the region’s high-quality development.

5. Conclusions

This study calculates the water footprint of major crops in Xinjiang from 2000 to 2020 based on water footprint theory and specifically examines the spatial and temporal characteristics of the blue water footprint using the Mann–Kendall statistical test. The results show that both CWFT and CWFB in Xinjiang followed a pattern of increase followed by stabilization from 2000 to 2020. The average annual CWFT for Xinjiang was 198.49 Gm³, with CWFB and CWFG accounting for 90.22% and 10.05%, respectively. The changes in CWFB can be categorized into three phases: the 2000–2008 period, characterized by slow growth at an average rate of 4.10%; the 2008–2014 period, marked by faster growth at an average rate of 8.35%; and the 2014–2020 period, characterized by stability, with minimal changes in the blue water footprint across regions, except for a sudden increase in 2017. Spatially, the southern border had the largest CWFB, followed by the Northern Xinjiang and Eastern Xinjiang. The top seven regions—Kashgar, Aksu, Bayingolin, Changji, Tacheng, Ili, and Hotan—accounted for 90.46% of the total CWFB in Xinjiang. Among the crops contributing to CWFB growth, cotton was the dominant driver, followed by wheat and maize. The economic effect was the primary driver of CWFB growth, while the water use efficiency effect acted as the main counteracting driver. The findings of these studies not only revealed the changing patterns of agricultural irrigation water usage and crop distribution characteristics in Xinjiang but also provided a robust reference for the rational planning of water resource allocation in arid areas, enhancing the efficiency of water resource utilization, as well as an important basis for the development of scientific water resource management strategies. Future research could explore the impact of the grey water footprint, as well as the effects of various measures on the sustainability of irrigated agriculture in Xinjiang, potentially through scenario analysis.

Author Contributions

Conceptualization, Z.W. and H.Y.; methodology, X.Z., Z.W., J.L. and X.G.; software, X.Z., H.L. (Huadong Liu), Y.L. and H.W.; validation, X.G.; formal analysis, X.Z. and H.L. (Haixia Lin); investigation, X.Z.; resources, X.Z.; data curation, X.Z., H.L. (Huadong Liu) and H.W.; writing—original draft preparation, X.Z.; writing—review and editing, H.Y., Z.W., J.L., W.L., H.L. (Haixia Lin), X.G., G.Q. and T.J.; visualization, Y.L.; supervision, Z.W.; project administration, J.L.; funding acquisition, Z.W. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

Restrictions apply to the datasets: The datasets presented in this article are not readily available because the data are part of an ongoing study. Requests to access the datasets should be directed to the corresponding author.

Conflicts of Interest

The authors declare no conflicts of interest.

Footnotes

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.

Figures

Figure 1. Map of the study areas.

Figure 2. Diagram of the steps of the analysis process.

Figure 3. Yearly crop water footprint of Xinjiang (2000–2020). The blue bar representing (CWFT) indicates the total crop water footprint, with the scale on the left Y-axis. The red square line representing (CWFB) and the red triangular line representing (CWFG) indicate the crop blue water footprint and the crop green water footprint, respectively, with the scale on the right Y-axis. The units are Gm3.

Figure 4. Mann–Kendall statistical test analysis of crop blue water footprint from 2000–2020. The blue line with square markers represents UF, the black line with circle markers represents UB, and the two dashed lines above and below the Y = 0 axis represent the ±1.96 significance level lines.

Figure 5. Crop blue water footprint of major crops from 2000 to 2020. The bars, colored differently from left to right, represent the CWFB in Gm3 for cotton, wheat, corn, alfalfa, and rice for each year, respectively.

Figure 6. Spatial distribution of CWFB for 2000, 2008, 2014, and 2020, along with the 2000–2020 annual average crop blue water footprint. The CWFB is classified into five levels, with different colors ranging from lighter to darker. The letters a, b, c, and d denote Urumqi, Changji, Karamay, and Kizilsu, respectively.

Figure 7. Annual average CWFB for the three major regions, fourteen prefectures, and major crops in Xinjiang from 2000 to 2020. Values in parentheses represent the corresponding annual average crop blue water footprint in Gm3.

Figure 8. Contributions of the five drivers of CWFB from 2000 to 2020. Total indicates the amount of change in CWFB, PF indicates the amount of change in CWFB due to changes in the population effect, WUEF indicates the amount of change in CWFB due to changes in the water use efficiency effect, EF indicates the amount of change in CWFB due to changes in the economic effect, WSF indicates the amount of change in CWFB due to changes in the water use structure effect, and TF indicates the amount of change in CWFB due to changes in technology effect. WSF denotes the change in CWFB caused by the change in water use structure effect, and TF denotes the change in CWFB caused by the change in technology effect.

Figure 9. Average contribution of drivers, 2000–2020. The value for each factor indicates the average amount of change in CWFB caused by its change.

References

1. Hoekstra, A.Y.; Mekonnen, M.M. The Water Footprint of Humanity. Proc. Natl. Acad. Sci. USA; 2012; 109, pp. 3232-3237. [DOI: https://dx.doi.org/10.1073/pnas.1109936109] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22331890]

2. Mekonnen, M.M.; Hoekstra, A.Y. Four Billion People Facing Severe Water Scarcity. Sci. Adv.; 2016; 2, e1500323. [DOI: https://dx.doi.org/10.1126/sciadv.1500323] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26933676]

3. He, Y.; Lin, K.; Zhang, F.; Wang, Y.; Chen, X. Coordination Degree of the Exploitation of Water Resources and Its Spatial Differences in China. Sci. Total Environ.; 2018; 644, pp. 1117-1127. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2018.07.050] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30743825]

4. Cao, X.; Shu, R.; Ren, J.; Wu, M.; Huang, X.; Guo, X. Variation and Driving Mechanism Analysis of Water Footprint Efficiency in Crop Cultivation in China. Sci. Total Environ.; 2020; 725, 138537. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2020.138537]

5. Van Oel, P.R.; Mekonnen, M.M.; Hoekstra, A.Y. The External Water Footprint of the Netherlands: Geographically-Explicit Quantification and Impact Assessment. Ecol. Econ.; 2009; 69, pp. 82-92. [DOI: https://dx.doi.org/10.1016/j.ecolecon.2009.07.014]

6. Hoekstra, A. Virtual Water Trade: Proceedings of the International Expert Meeting on Virtual Water Trade; IHE Delft: Delft, The Netherlands, 2003; pp. 1-244.

7. Kayatz, B.; Baroni, G.; Hillier, J.; Luedtke, S.; Heathcote, R.; Malin, D.; van Tonder, C.; Kuster, B.; Freese, D.; Huettl, R. et al. Cool Farm Tool Water: A Global Online Tool to Assess Water Use in Crop Production. J. Clean Prod.; 2019; 207, pp. 1163-1179. [DOI: https://dx.doi.org/10.1016/j.jclepro.2018.09.160]

8. Aldaya, M.M.; Chapagain, A.K.; Hoekstra, A.Y.; Mekonnen, M.M. The Water Footprint Assessment Manual: Setting the Global Standard; Routledge: London, UK, 2012.

9. Koley, S.; Jeganathan, C. Estimation of the Green and Blue Water Footprint of Kharif Rice Using Remote Sensing Techniques: A Case Study of Ranchi. Proceedings of the 2020 IEEE India Geoscience and Remote Sensing Symposium (INGARSS); Ahmedabad, India, 1–4 December 2020; pp. 1-4.

10. Liu, X.; Xu, Y.; Sun, S.; Wu, P.; Wang, Y. Water Resource Efficiency Evaluation of Crop Production in Arid and Semi-Arid Regions Based on Water Footprint and Comparative Advantage. Eur. J. Agron.; 2024; 160, 127310. [DOI: https://dx.doi.org/10.1016/j.eja.2024.127310]

11. Salmoral, G.; Willaarts, B.A.; Garrido, A.; Guse, B. Fostering Integrated Land and Water Management Approaches: Evaluating the Water Footprint of a Mediterranean Basin under Different Agricultural Land Use Scenarios. Land Use Pol.; 2017; 61, pp. 24-39. [DOI: https://dx.doi.org/10.1016/j.landusepol.2016.09.027]

12. Mao, Y.; Liu, Y.; Zhuo, L.; Wang, W.; Li, M.; Feng, B.; Wu, P. Quantitative Evaluation of Spatial Scale Effects on Regional Water Footprint in Crop Production. Resour. Conserv. Recycl.; 2021; 173, 105709. [DOI: https://dx.doi.org/10.1016/j.resconrec.2021.105709]

13. Ansorge, L.; Stejskalova, L. Water Footprint as a Tool for Selection of Alternatives (Comments on “Food Recommendations for Reducing Water Footprint”). Sustainability; 2022; 14, 6317. [DOI: https://dx.doi.org/10.3390/su14106317]

14. Kim, B.F.; Santo, R.E.; Scatterday, A.P.; Fry, J.P.; Synk, C.M.; Cebron, S.R.; Mekonnen, M.M.; Hoekstra, A.Y.; de Pee, S.; Bloem, M.W. et al. Country-Specific Dietary Shifts to Mitigate Climate and Water Crises. Glob. Environ. Change-Human Policy Dimens.; 2020; 62, 101926. [DOI: https://dx.doi.org/10.1016/j.gloenvcha.2019.05.010]

15. Tamea, S.; Carr, J.A.; Laio, F.; Ridolfi, L. Drivers of the Virtual Water Trade. Water Resour. Res.; 2014; 50, pp. 17-28. [DOI: https://dx.doi.org/10.1002/2013WR014707]

16. Zhuo, L.; Mekonnen, M.M.; Hoekstra, A.Y.; Wada, Y. Inter- and Intra-Annual Variation of Water Footprint of Crops and Blue Water Scarcity in the Yellow River Basin (1961–2009). Adv. Water Resour.; 2016; 87, pp. 29-41. [DOI: https://dx.doi.org/10.1016/j.advwatres.2015.11.002]

17. Fu, T.; Xu, C.; Yang, L.; Hou, S.; Xia, Q. Measurement and Driving Factors of Grey Water Footprint Efficiency in Yangtze River Basin. Sci. Total Environ.; 2022; 802, 149587. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2021.149587]

18. Ribašauskienė, E.; Eičaitė, O.; Baležentis, T.; Agnusdei, G.P. Decomposition of the Water Footprint of Food Loss and Waste: The Case of Lithuanian Supply Chains. Ecol. Indic.; 2024; 166, 112255. [DOI: https://dx.doi.org/10.1016/j.ecolind.2024.112255]

19. Montoya, M.A.; Allegretti, G.; Bertussi, L.A.S.; Talamini, E. Domestic and Foreign Decoupling of Economic Growth and Water Consumption and Its Driving Factors in the Brazilian Economy. Ecol. Econ.; 2023; 206, 107737. [DOI: https://dx.doi.org/10.1016/j.ecolecon.2022.107737]

20. Laporte, J.P.; Román-Collado, R.; Cansino, J.M. Key Driving Forces of Energy Consumption in a Higher Education Institution Using the LMDI Approach: The Case of the Universidad Autónoma de Chile. Appl. Energy; 2024; 372, 123797. [DOI: https://dx.doi.org/10.1016/j.apenergy.2024.123797]

21. Hoekstra, R.; van der Bergh, J. Comparing Structural and Index Decomposition Analysis. Energy Econ.; 2003; 25, pp. 39-64. [DOI: https://dx.doi.org/10.1016/S0140-9883(02)00059-2]

22. Cansino, J.M.; Roman, R.; Ordonez, M. Main Drivers of Changes in CO₂ Emissions in the Spanish Economy: A Structural Decomposition Analysis. Energy Policy; 2016; 89, pp. 150-159. [DOI: https://dx.doi.org/10.1016/j.enpol.2015.11.020]

23. Martin, J.-C.; Becuwe, S. Regionalising the Results of a National Structural Decomposition Analysis of Greenhouse Gas Emissions: An Application to Aquitaine Region. Environ. Model. Assess.; 2014; 19, pp. 257-269. [DOI: https://dx.doi.org/10.1007/s10666-013-9396-9]

24. Ang, B.W.; Wang, H. Index Decomposition Analysis with Multidimensional and Multilevel Energy Data. Energy Econ.; 2015; 51, pp. 67-76. [DOI: https://dx.doi.org/10.1016/j.eneco.2015.06.004]

25. Ang, B.W.; Choi, K.-H. Decomposition of Aggregate Energy and Gas Emission Intensities for Industry: A Refined Divisia. Energy J.; 1997; 18, pp. 59-73. [DOI: https://dx.doi.org/10.5547/ISSN0195-6574-EJ-Vol18-No3-3]

26. Ang, B.W.; Xu, X.Y.; Su, B. Multi-Country Comparisons of Energy Performance: The Index Decomposition Analysis Approach. Energy Econ.; 2015; 47, pp. 68-76. [DOI: https://dx.doi.org/10.1016/j.eneco.2014.10.011]

27. Zhang, Y.; Long, A.; Zhang, P.; Deng, X.; Gu, X. Are Water Use Efficiency and Effectiveness Relatively Lower in Arid Zones? Comparative Analyses of the Water Productivity of Typical Crops. Agronomy; 2024; 14, 2153. [DOI: https://dx.doi.org/10.3390/agronomy14092153]

28. Qin, G.; Liu, J.; Lin, H.; Javed, T.; Gao, X.; Tang, Y.; Mu, X.; Guo, M.; Wang, Z. Assessing the Coordination and Spatial Equilibrium of Water, Energy, and Food Systems for Regional Socio-Economic Growth in the Ili River Valley, China. Agronomy; 2024; 14, 2037. [DOI: https://dx.doi.org/10.3390/agronomy14092037]

29. Pei, D.; Wen, Y.; Li, W.; Ma, Z.; Guo, L.; Zhang, J.; Liu, M.; Mu, X.; Wang, Z. Agricultural Water Rebound Effect and Its Driving Factors in Xinjiang, China. Agric. Water Manag.; 2024; 304, 109086. [DOI: https://dx.doi.org/10.1016/j.agwat.2024.109086]

30. Gao, R.; Zhao, D.; Zhang, P.; Li, M.; Huang, H.; Zhuo, L.; Wu, P. Driving Factor Analysis of Spatial and Temporal Variations in the Gray Water Footprint of Crop Production via Multiple Methods: A Case for West China. Front. Environ. Sci.; 2023; 10, 1104797. [DOI: https://dx.doi.org/10.3389/fenvs.2022.1104797]

31. Xu, C.; Liu, Y.; Fu, T. Spatial-Temporal Evolution and Driving Factors of Grey Water Footprint Efficiency in the Yangtze River Economic Belt. Sci. Total Environ.; 2022; 844, 156930. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2022.156930]

32. Gocic, M.; Trajkovic, S. Analysis of Changes in Meteorological Variables Using Mann-Kendall and Sen’s Slope Estimator Statistical Tests in Serbia. Glob. Planet. Change; 2013; 100, pp. 172-182. [DOI: https://dx.doi.org/10.1016/j.gloplacha.2012.10.014]

33. Hamed, K.H. Trend Detection in Hydrologic Data: The Mann–Kendall Trend Test under the Scaling Hypothesis. J. Hydrol.; 2008; 349, pp. 350-363. [DOI: https://dx.doi.org/10.1016/j.jhydrol.2007.11.009]

34. Worku, G.; Teferi, E.; Bantider, A.; Dile, Y.T. Observed Changes in Extremes of Daily Rainfall and Temperature in Jemma Sub-Basin, Upper Blue Nile Basin, Ethiopia. Theor. Appl. Climatol.; 2019; 135, pp. 839-854. [DOI: https://dx.doi.org/10.1007/s00704-018-2412-x]

35. Roshan, G.; Yousefi, R.; Kovacs, A.; Matzarakis, A. A Comprehensive Analysis of Physiologically Equivalent Temperature Changes of Iranian Selected Stations for the Last Half Century. Theor. Appl. Climatol.; 2018; 131, pp. 19-41. [DOI: https://dx.doi.org/10.1007/s00704-016-1950-3]

36. Reddy, K.V.; Paramesh, V.; Arunachalam, V.; Das, B.; Ramasundaram, P.; Pramanik, M.; Sridhara, S.; Reddy, D.D.; Alataway, A.; Dewidar, A.Z. et al. Farmers’ Perception and Efficacy of Adaptation Decisions to Climate Change. Agronomy; 2022; 12, 1023. [DOI: https://dx.doi.org/10.3390/agronomy12051023]

37. Odey, G.; Adelodun, B.; Lee, S.; Adeyemi, K.A.; Cho, G.; Choi, K.S. Environmental and Socioeconomic Determinants of Virtual Water Trade of Grain Products: An Empirical Analysis of South Korea Using Decomposition and Decoupling Model. Agronomy; 2022; 12, 3105. [DOI: https://dx.doi.org/10.3390/agronomy12123105]

38. Fang, H.; Wu, N.; Adamowski, J.; Wu, M.; Cao, X. Crop Water Footprints and Their Driving Mechanisms Show Regional Differences. Sci. Total Environ.; 2023; 904, 1675499. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2023.167549]

39. Li, Q.; Huang, W.; Wang, J.; Zhang, Z.; Li, Y.; Han, Y.; Feng, L.; Li, X.; Yang, B.; Wang, G. et al. Quantitative Evaluation of Variation and Driving Factors of the Regional Water Footprint for Cotton Production in China. Sustain. Prod. Consump.; 2023; 35, pp. 684-696. [DOI: https://dx.doi.org/10.1016/j.spc.2022.11.016]

40. Cui, J.; Sui, P.; Wright, D.L.; Wang, D.; Sun, B.; Ran, M.; Shen, Y.; Li, C.; Chen, Y. Carbon Emission of Maize-Based Cropping Systems in the North China Plain. J. Clean. Prod.; 2019; 213, pp. 300-308. [DOI: https://dx.doi.org/10.1016/j.jclepro.2018.12.174]

41. Zheng, J.; Wang, W.; Liu, G.; Ding, Y.; Cao, X.; Chen, D.; Engel, B.A. Towards Quantification of the National Water Footprint in Rice Production of China: A First Assessment from the Perspectives of Single-Double Rice. Sci. Total Environ.; 2020; 739, 140032. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2020.140032]

42. Li, X.; Chen, Y.; Xu, L.; Li, P.; Zhang, R. Transformation of Farmland Use and Driving Mechanism in Xinjiang since China’s Western Development Policy. Front. Ecol. Evol.; 2022; 10, 942065. [DOI: https://dx.doi.org/10.3389/fevo.2022.942065]

43. Cai, T.; Luo, X.; Fan, L.; Han, J.; Zhang, X. The Impact of Cropland Use Changes on Terrestrial Ecosystem Services Value in Newly Added Cropland Hotspots in China during 2000–2020. Land; 2022; 11, 2294. [DOI: https://dx.doi.org/10.3390/land11122294]

44. Luo, R.; Fan, L. Government Operation Mechanism and Its Effects under Counterpart Assistance-A Case Study of Xinjiang, China. Proceedings of the 2014 International Conference on Public Administration (10TH); Chengdu, China, 24–26 October 2014; pp. 37-44.

45. Zhang, L.; Yu, Y.; Guo, Z.; Ding, X.; Zhang, J.; Yu, R. Investigating Agricultural Water Sustainability in Arid Regions with Bayesian Network and Water Footprint Theories. Sci. Total Environ.; 2024; 951, 175544. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2024.175544] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39151630]

46. Su, K.; He, D.; Wang, R.; Han, Z.; Deng, X. Assessment of Natural Resource Endowment and Urban-Rural Integration for Sustainable Development in Xinjiang, China. J. Clean. Prod.; 2024; 450, 142046. [DOI: https://dx.doi.org/10.1016/j.jclepro.2024.142046]

47. Cui, S.; Zhang, J.; Wang, X.; Wu, M.; Cao, X. Fuzzy Composite Risk Assessment of Water-Energy-Food-Carbon Nexus in the Dispark Pumped Irrigation System. J. Hydrol.; 2023; 624, 129879. [DOI: https://dx.doi.org/10.1016/j.jhydrol.2023.129879]

48. Zhao, C.; Chen, B. Driving Force Analysis of the Agricultural Water Footprint in China Based on the LMDI Method. Environ. Sci. Technol.; 2014; 48, pp. 12723-12731. [DOI: https://dx.doi.org/10.1021/es503513z]

49. Zhang, C.; Chen, J.; Chu, Z.; Zhang, P.; Xu, J. History and Future of Water Footprint in the Yangtze River Delta of China. Environ. Sci. Pollut. Res.; 2024; 31, pp. 25508-25523. [DOI: https://dx.doi.org/10.1007/s11356-024-32757-5]