The Yellow River Basin (YRB) is facing a serious water shortage. How to effectively alleviate the water crisis and achieve sustainable development in the YRB has become a widespread concern. By using the interregional input–output tables of China in 2002, 2007, 2012 and 2017, we analysed the transfer of virtual water and value-added and the inequality embodied in trade between the YRB and other regions. Results demonstrated that: (1) for the YRB, the pressure on water resources was alleviated through the net inflow of virtual water after 2007. However, the economic situation deteriorated due to the net outflow of value-added in interregional trade after 2012. (2) There existed a serious inequality in virtual water consumption and economic benefits embodied in trade between the YRB and Beijing, Shanghai, etc., with regional inequality (RI) index exceeding 1. Meanwhile, agriculture faced the most serious inequality among all sectors in the YRB. Accordingly, the YRB should aim to optimise its industrial structure and improve water use efficiency to achieve a win-win situation for both economic development and net virtual water inflow. In addition, policymakers should take measures to flexibly adjust the trade scale between the YRB and other regions based on the RI index.

  • Water consumption and value-added were calculated from two perspectives.

  • The spatial characteristics of virtual water and value-added flows were analysed.

  • The regional inequality index was calculated.

  • Agriculture faced the most serious inequality in the Yellow River Basin.

  • Strategies were proposed from sectoral and regional trade perspectives.

Graphical Abstract

Graphical Abstract
Graphical Abstract

The Yellow River Basin (YRB), with an area of 795,000 km2, flows through Qinghai, Sichuan, Gansu, Ningxia, Inner Mongolia, Shaanxi, Shanxi, Henan and Shandong (Yellow River Conservancy Committee, 2020). The past 40 years have witnessed a sharp increase in water demand of various industries in the YRB with the rapid industrialisation and urbanisation (Wang et al., 2018a), especially in industries that rely on natural resources such as coal-based electricity industry (Zhang et al., 2017) and agriculture. However, water resources in the YRB are reducing because of climate change and human activities (Wang et al., 2006, 2017; Chen et al., 2020; Omer et al., 2020). From 2004 to 2019, the proportion of per capita water resources in the YRB decreased from 40.2 to 37.9%. The shortage of water resources is increasingly serious in the YRB (Xu et al., 2005; Zhang et al., 2009; Ringler et al., 2010; Liu & Yang, 2012; Cai et al., 2017b), which is not conducive to the sustainable development of the ecological environment and society (Zhang et al., 2011). In this case, the State Council issued ‘The Opinions on the Implementation of the Strictest Water Resources Management System’ and ‘The Assessment Method for the Implementation of the Strictest Water Resources Management System’ in 2012 and 2013 and clearly proposed the red line of total water consumption for each region. However, it only managed water resources from the perspective of production but ignored the important role of trade, restricting the economic development of the YRB to a certain extent. In fact, the YRB exports many water-intensive products through interprovincial trade (Feng et al., 2012), most of which are located in the upstream of industrial chain and contain low value-added. The YRB is facing serious inequality between virtual water consumption and economic benefits embodied in trade. Therefore, how to achieve sustainable economic development while mitigating the water crisis in the YRB is an urgent issue to be addressed. The inequality analysis in this paper helps to expound the asymmetric relationship between the YRB and other regions in terms of the virtual water and value-added transfer embodied in trade, providing a scientific basis for the win-win situation of water crisis mitigation and economic development in the YRB.

At present, studies on regional inequality (RI) are mainly carried out from three aspects. From the perspective of research objects, many scholars mainly analysed ecologically unequal exchange between countries. They believed that although international trade follows the principle of equivalent exchange, the equivalent trade volume measured in monetary terms has different environmental costs in developed and developing economies (Hornborg, 1998). Environmental improvement in developed countries comes at the cost of environmental degradation in developing countries (Roberts & Parks, 2009; Hornborg & Martinez-Alier, 2016; Frey et al., 2019). As the largest developing country, China faces a serious ecological inequality exchange (Feng & Liu, 2019). Actually, with the increase of interprovincial trade and the expansion of regional economic disparities, there also exists the inequality embodied in trade among different provinces in China. Numerous scholars have also studied this (Feng et al., 2013; Zhang et al., 2018; Wei et al., 2020). We can find that the above studies mainly focus on the inequality between different countries in the world and between different provinces in China. However, due to the increasing water scarcity in the YRB, it is necessary to study the inequality of virtual water consumption and economic benefits between the YRB and other regions.

In terms of research perspectives, some studies focused on RI in actual water use (Babuna et al., 2020). Some studies focused on the inequality in resources and pollutants transfer embodied in trade. The research scope included carbon dioxide transfer (Guo et al., 2012; Wang et al., 2018b), air pollutant transfer (Liang et al., 2014; Zhao et al., 2015), virtual water transfer (Zhao et al., 2010; Chen et al., 2017a; Liao et al., 2018) and energy transfer (Chen et al., 2017b) embodied in trade. The above studies have demonstrated that plenty of embodied resources and pollutants are transferred through trade from developed to underdeveloped regions, from coastal to inland regions as well as from eastern to central and western regions, and that the final demand of developed regions exacerbates the burden of resources and environment on underdeveloped regions. Zhang & Anadon (2014) found that the water footprint (WF) of municipalities directly under the central government is heavily dependent on virtual water inflows from other provinces by calculating the virtual water trade and WF between provinces in China. Considering the impact of virtual water flows on water-scarce regions in China, Feng et al. (2014) incorporated water consumption, water scarcity and ecosystem impacts into the analysis of interregional virtual water transfer. They found that net virtual water inflow provinces sacrifice water resources from underdeveloped water-scarce provinces and that exports from coastal provinces sacrifice the ecosystem quality of underdeveloped regions. Considering the water quality impacts of virtual water flows, Cai et al. (2017a) incorporated the blue WF related to water quantity and the grey WF related to water quality into the analysis of virtual water transfer, finding that the virtual grey water flows are 8.65 times larger than the virtual blue water flows. However, actually, economic benefits gained by underdeveloped regions in interregional trade are an appropriate compensation for their resource consumption and environmental costs. Zhang et al. (2018), Wei et al. (2020) and Xiong et al. (2021) analysed the net transfer between atmospheric pollutant equivalents, carbon dioxide, water pollution and value-added embodied in trade and found that there exists a serious inequality between developed and underdeveloped provinces in China. It can be found that there have been more studies on greenhouse gas emissions, virtual water and air pollutants embodied in trade but fewer studies on the inequality between economic benefits and virtual water consumption embodied in trade.

From the perspective of research methods, on the one hand, when only the inequality in actual water and virtual water is involved, scholars analyse it by using the Gini coefficient and the Lorenz curve. Babuna et al. (2020) used the Gini coefficient and the global Moran index to calculate the urban inequality and spatial distribution of inequality in the Yangtze River Delta. It was found that the urban water inequality in the Yangtze River Delta is decreasing and that no city shows high inequality. According to the Moran index, cities are divided into nine types. Seekell et al. (2011) plotted the Lorenz curve of the WF and calculated its Gini coefficient, finding that virtual water use is highly unequal and that current virtual water transfer is not enough to reduce inequality. Wang et al. (2019) evaluated the spatial and temporal inequality of WF in Jilin Province by using the Gini coefficient and imbalance index. The WF in Jilin Province is in a state of ‘relative equality’. The influence of human factors on WF inequality is increasing, while the influence of natural factors is decreasing. The spatial difference in WF inequality is obvious. On the other hand, when scholars' studies combine economic benefits with resource consumption or environmental costs in trade, scholars usually use the regional environmental inequality (REI) index to quantify (Zhang et al., 2018; Wei et al., 2020). Chen et al. (2021) calculated the water inequality index and found that the water inequality index is significantly higher than the land inequality index. Global trade exacerbates both the unequal distribution of water resources and the environmental and economic inequality between countries. The above analyses found that some statistical indicators such as the Gini coefficient are suitable for studying the inequality of actual water distribution and virtual water, and the REI index is applicable to studies of inequality in economic benefits and environmental costs in interprovincial or international trade. However, for the specific region, the REI index does not accurately show the region's position in trade and the inequality the region faces.

The main contributions of this paper are to analyse the transfer of virtual water and value-added embodied in trade in the YRB, to improve the REI index and to quantitatively assess the inequality by using the RI index. Through the RI index, we can explain the status of the YRB in interprovincial trade and divide it into four types. Meanwhile, focusing on the inequality of various industries in the YRB, we find more about the spatial transfer of virtual water and value-added in various industries and propose corresponding solutions for specific industries in the YRB by the comprehensive consideration of both.

Accounting for production-based and consumption-based water consumption and value-added

The research basis of this paper is the multiregional input–output (MRIO) model (the basic form is shown in Supplementary Material, Table A1). The MRIO model portrays the trade relationship between regions and is widely used to systematically analyse the sources and destinations of virtual water embodied in trade from both regional and sectoral perspectives (Zhang & Anadon, 2014; Chen et al., 2017a; Wang et al., 2021). The MRIO model also establishes a complete interregional production and consumption relationship (Liu & Wang, 2017), and we can obtain the complete water consumption required for final use, including both direct and indirect water consumption. Besides, the interprovincial trade in our study is bidirectional. Net transfer can be calculated based on the MRIO model (Fan et al., 2019). Thus, we use this model to track the net transfer of virtual water and value-added embodied in trade between the YRB and other regions.

The MRIO model in this paper ignores international trade because we only focus on the impact analysis of interprovincial trade in the YRB. Meanwhile, 28 sectors are retained in this paper in order to avoid computational errors caused by the merger of too many sectors. Although the publication interval of interregional input–output tables is long, the various coefficients calculated based on the tables are valid and stable in the long term (Guan & Hubacek, 2007). Therefore, the latest table used in this paper is the one in 2017. In general, the MRIO model is suitable for our study.

Assuming that there are m regions and n sectors in the model, denotes the intermediate input from the sector i in region r to the sector j in region s, is the final product supplied by sector i in region r to region s and is the total output of sector i in region r.

We can derive the following equation:
formula
(1)
where denotes the coefficient of direct consumption. Write Equation (1) in matrix form as , and the deformation yields:
formula
(2)
where is a matrix; and are both vectors. is the Leontief inverse matrix.
Production-based water consumption and value-added are water resources consumed and value-added realised in the production of products in the region. Consumption-based water consumption and value-added are water resources and value-added included in products that meet the final demand of the region. First, we calculate the direct water use coefficient and the direct value-added coefficient. The direct water use coefficient is the direct water consumption by producing one unit of output in sector j of region r and is calculated through Equation (3). The direct value-added coefficient reflects the value-added obtained by producing one unit of output in sector j of region r and is calculated by Equation (4), where and denote water consumption and value-added of sector j in region r.
formula
(3)
formula
(4)
Let and be the diagonal matrixes with the corresponding sectoral water intensity and direct value-added coefficients of region r but zero for all other regions, respectively. is the vector composed of , and is the diagonal matrix of . and are vectors consisting of and Then, we can calculate production-based water consumption and value-added in the region r by Equations (5) and (6). Consumption-based water consumption and value-added can be calculated by Equations (7) and (8)
formula
(5)
formula
(6)
formula
(7)
formula
(8)
where consumption-based water consumption and value-added of sector i in region r is the sum of s from region 1 to region m.

Accounting for the transfer of virtual water and value-added

Virtual water, first introduced by Allan (1993), is the water required in the production of products (Yang & Zehnder, 2007). Let and be the diagonal matrixes of end use in region s and region r, respectively, and and be the diagonal matrixes of water intensity in the corresponding sectors in region s and region r, respectively, but zero for all other regions. We can calculate virtual water transfer by the following equations
formula
(9)
formula
(10)
formula
(11)
where denotes water consumption in region r pulled by the final demand in region s, i.e., the virtual water transferred from r to s. In contrast, denotes the transfer of virtual water from s to r. indicates the net transfer of virtual water from r to s, which can be positive or negative. If is greater than 0, virtual water net flows from r to s; if is less than 0, virtual water net flows from s to r.
Similarly, let and be the diagonal matrixes of direct value-added coefficients in the corresponding sectors in s and r, respectively, but zero for all other regions. We can obtain the following equations
formula
(12)
formula
(13)
formula
(14)
where denotes the value-added in region s pulled by final demand in region r, i.e., the value-added transferred from r to s. In contrast, denotes the value-added transferred from s to r. is the net transfer of value-added from r to s.

Regional inequality index

We slightly improve on the REI index and use the RI index to quantify the inequality of virtual water consumption and economic benefits embodied in trade. It is worth noting that economic benefits here are value-added.

Let and be the elements in and , respectively. For the sake of comparative analysis of the data, it is assumed that for any matrix , we can normalise any element b in it to range between 0 and 1 according to . Then, RI index between region r and region s can be calculated using Equation (15).
formula
(15)

Through Equation (15), the RI index can be divided into four types. Type I, calculated by equation ①, denotes that virtual water net transfers from r to s and value-added net transfers from s to r. Type II, calculated by equation ②, denotes that value-added net transfers from r to s and virtual water net transfers from s to r. Type III, calculated by equation ③, denotes that virtual water and value-added net both transfer from r to s. Type IV, calculated by equation ④, denotes that virtual water and value-added net both transfer from s to r. And the greater the absolute value of , the more serious the inequality of virtual water consumption and economic benefits between region r and region s.

Data

As most of the areas in Sichuan Province belong to the Yangtze River Basin, the YRB mentioned in this paper only includes eight provinces, namely, Qinghai, Gansu, Ningxia, Inner Mongolia, Shaanxi, Shanxi, Henan and Shandong. To ensure that the analysis is more convincing, we use interregional input–output tables of China in 2002, 2007, 2012 and 2017. And the table in 2017 is the latest one. Given the lack of data for Tibet, Hong Kong, Macau and Taiwan, only 30 provinces are included in tables. Furthermore, we combine them into 23 regions, and each region includes 28 sectors (Supplementary Material, Table A2).

Since we calculate the net transfer of virtual water in the YRB, we need the water use data of each sector in each region. For the primary industry, its water consumption can be directly obtained from the China Statistical Yearbook in 2003, 2008, 2013 and 2018.

For the secondary industry sub-sectors in each province in 2002, 2007 and 2012, we use to denote water consumption of the national secondary sector i (obtained from the China Environment Yearbook in 2003 and the China Environmental Statistics Annual Report in 2007 and 2012, respectively), and is the total output of the national secondary sector i. By dividing by , we can yield water consumption per unit output in secondary sector i nationwide, which is expressed by . The average water consumption coefficient matrix of secondary industry sub-sectors is . Taking region r as an example, its output matrix is . The total water consumption of the secondary industry in region r can be calculated by multiplying with . Meanwhile, the actual total water consumption of the secondary industry can be directly obtained from the China Statistical Yearbook in 2003, 2008 and 2013. There will be a gap between the actual total water consumption and the calculated one in each region due to different water use efficiency among regions. Therefore, we divide by to get the correction coefficient . Then, water consumption of sector i of the secondary industry in region r can be calculated by . This method is scientific and has been used by some scholars (Wang, 2014; Chen et al., 2017a). Since there is no direct data on water consumption of the secondary industry sub-sectors nationwide in 2017, we assume that water consumption per unit value-added of each sector in the secondary industry in 2017 is the same as that in 2015 and then estimate water consumption at comparable prices (S22 is assumed to be the same as in 2012). Value-added is from national input–output extension table in 2015 and national input–output table in 2017. From 2012 to 2017, sub-sector price indices come from the National Bureau of Statistics. Water consumption of sub-sectors in the national secondary industry in 2015 is from the China Environmental Statistics Annual Report in 2015. Then, we calculate water consumption of the secondary industry sub-sectors in each region in 2017 through the above method.

In order to derive the total water consumption of the tertiary industry, the total household water consumption in each region needs to be known first, which can be obtained by multiplying the national total household water consumption by the population ratio of each region (the former can be obtained from the China Urban and Rural Statistical Yearbook in 2002, 2007, 2012 and 2017, and the latter can be obtained from the China Statistical Yearbook in 2003, 2008, 2013 and 2018). The water consumption of the tertiary industry in each region can be calculated by subtracting water consumption of the primary industry, the secondary industry and the households from total water consumption of each region. Meanwhile, we assume that the proportion of water consumption in the tertiary industry sub-sectors in 2007, 2012 and 2017 is the same as that in 2002. Therefore, water consumption of the tertiary industry sub-sectors in each region can be calculated by multiplying water consumption of the tertiary industry in each region by the proportion of water consumption in 2002 (Researching Group of Chinese Input-Output Association, 2007).

Comparison of production-based and consumption-based water consumption and value-added

Production-based and consumption-based water consumption

In 2002, 2007, 2012 and 2017, water consumption in the YRB based on the consumption side was 0.99, 1.18, 1.28 and 1.004 times as much as that based on the production side, respectively (Figure 1). Water consumption in the YRB achieved a great shift from production side over consumption side to consumption side over production side from 2002 to 2017. In general, the YRB alleviated water stress through domestic trade.

Fig. 1.

Production- and consumption-based water consumption in the YRB (2002–2017). Note: I indicates the production side; II indicates the consumption side.

Fig. 1.

Production- and consumption-based water consumption in the YRB (2002–2017). Note: I indicates the production side; II indicates the consumption side.

Fig. 2.

Production- and consumption-based value-added in the YRB (2002–2017). Note: I indicates the production side; II indicates the consumption side.

Fig. 2.

Production- and consumption-based value-added in the YRB (2002–2017). Note: I indicates the production side; II indicates the consumption side.

Specifically, S01 had the highest production-based water consumption, accounting for more than 70% in all years. First, this is because the YRB (especially the North China Plain, Fen-Wei Plain and Hetao Irrigation District) is an important grain gathering area in China, and a large amount of agricultural output has led to high production-based water consumption. Second, partial evaporation of water, serious leakage of water delivery channels and low utilisation of irrigation water have led to large production-based water consumption. However, with the continuous technological progress, the efficiency of water use in agriculture and electricity has been improved. From 2002 to 2017, the proportion of production-based water consumption of S01 and S22 decreased from 79.3 and 8.1% to 70.1 and 5.2%, respectively.

From the perspective of consumption, the sectors with more water consumption in the YRB over the years were S25, S28, S01 and S06. From 2002 to 2017, water consumption in S25 and S28 increased from 9.4 × 109 and 5.6 × 109 tons to 1.57 × 1010 and 1.3 × 1010 tons, respectively, while water consumption in S01 gradually decreased from 3.5 × 1010 tons to 1.98 × 1010 tons.

A comparative analysis shows that production-based water consumption in S01 and S22 was significantly larger than consumption-based water consumption in all years. The YRB intensified the pressure on water resources when supplying agricultural products, electricity and heat. Meanwhile, the YRB's S25, S28 and S06 had more consumption-based water consumption in all years. The YRB relieved local water stress by consuming products of the above-mentioned sectors.

Production-based and consumption-based value-added

In 2002 and 2007, consumption-based value-added was smaller than production-based value-added, indicating that the YRB improved its economic situation through domestic trade by taking advantage of abundant resources such as coal, non-ferrous metals, oil and gas from 2002 to 2007. However, after 2012, consumption-based value-added was greater in the YRB (Figure 2). After the 18th National Congress of the Communist Party of China, some developed provinces in the east of China took the lead in becoming innovation-driven regions (Chen et al., 2017c). The YRB imported a large amount of high value-added products from these regions and exported products with low value-added, leading to the YRB's net outflow of value-added in interprovincial trade.

From the production perspective, the three most profitable sectors in the YRB over the years were S28, S27 and S01, whose production-based value-added increased gradually from 2002 to 2017. S28's products had the highest value-added, so it gained the highest economic benefits. From the consumption perspective, S25 and S28 had higher value-added. From 2002 to 2017, the proportion of total value-added in the above sectors on the consumption side increased from 40.9 to 56.9%.

From 2002 to 2017, consumption-based value-added in S25 and S28 in the YRB was almost always greater than production-based value-added, indicating that the YRB achieved the net outflow of value-added by consuming products from these two sectors. In contrast, value-added in agriculture and resource-based industries such as S02, S03, S14 and S11 on the production side was larger in all years. Taking advantage of the resource endowment, the YRB achieved the net inflow of value-added in the above sectors.

Production-based water consumption and value-added

We construct the PV index to calculate water consumption to obtain one unit of value-added. From 2002 to 2017, with the development of the economy, the PV index in the YRB decreased from 39 to 5.2 kg/RMB, with a drop rate of 86.7% (Figure 3), indicating that the utilisation of water resources in the YRB changed from crude and inefficient to economical and intensive.

Fig. 3.

PV index and CV index in the YRB (2002–2017).

Fig. 3.

PV index and CV index in the YRB (2002–2017).

The sector with the highest PV index in all years was S01, followed by S22, both of which were higher than the average level of industries in the YRB. In 2017, for example, PV values of S01 and S22 in the YRB were 8.38 and 2.35 times as much as the industry average, respectively. Compared with other sectors, S01 and S22 consumed more water resources for obtaining the same value-added, which aggravated water stress and was not conducive to the sustainable development of the YRB. From 2002 to 2017, PV values of all industries (except S15) decreased. In S02, S16, S17 and S25, the decline rates of PV were greater than the average decline rate. The sharp decrease of PV index in the above industries made the YRB face less pressure on water resources in the production of related products.

Consumption-based water consumption and value-added

The CV index can be used to calculate water resources obtained by paying one unit of value-added. From 2002 to 2017, the CV index in the YRB gradually decreased from 39.5 to 5.2 kg/RMB (Figure 3). On the one hand, the YRB imported more and more low water-consuming and high value-added goods and services through domestic trade. On the other hand, all regions continuously improved their water use efficiency and saved a large amount of water resources.

Meanwhile, the CV index of all industries in the YRB decreased from 2002 to 2017. The sector that had the highest CV index over the years was S01, followed by S06, both of which were higher than the average level of industries in the YRB. In 2017, CV values of S01 and S06 in the YRB were 6.4 and 3.1 times as much as the industry average, respectively. The import of products and services by above-mentioned sectors received a lot of water resources, alleviating the pressure on water resources in the YRB.

What is more interesting is that for the YRB, CV values were larger than PV values in most sectors over the years, and only S01 and S22 had PV values greater than CV values, and they need to improve water use efficiency urgently.

Spatial-temporal evolution and industry characteristics of virtual water and value-added transfer embodied in trade

Spatial-temporal evolution of virtual water transfer

From 2002 to 2017, the number of target provinces for net virtual water outflow in the YRB generally showed a fluctuating trend. There were 13 provinces in 2002, 6 in 2007, 7 in 2012 and 10 in 2017. In 2002, the target provinces were mainly located in Northeast China, the eastern coastal region and the municipalities directly under the central government (Figure 4(a)). In 2007, they were further concentrated in the developed south-eastern coastal region and the municipalities (Figure 4(b)). Compared with 2007, the target provinces in 2012 and 2017 expanded to the western region but were still mainly distributed in the developed Beijing-Tianjin region and the south-eastern coastal region such as Guangdong, Zhejiang and Shanghai (Figure 4(c) and 4(d)). For example, in 2017, the YRB achieved a net transfer of 7.22 × 109 tons of virtual water to the above-mentioned regions, accounting for 78.9% of the total net virtual water outflow in the YRB.

Fig. 4.

Spatial-temporal evolution of virtual water transfer between the YRB and other regions (2002–2017).

Fig. 4.

Spatial-temporal evolution of virtual water transfer between the YRB and other regions (2002–2017).

Meanwhile, from 2002 to 2017, the number of target provinces for net virtual water inflow in the YRB generally showed a fluctuating upward. There were 9 provinces in 2002, 16 in 2007, 15 in 2012 and 12 in 2017 (Figure 4). The target provinces were mainly located in the less developed north-western and central regions in 2002 (Figure 4(a)) and further expanded to the north-eastern and south-western regions in 2007 (Figure 4(b)). In 2012 and 2017, the target provinces were mainly concentrated in the less developed central and western regions (Figure 4(c) and 4(d)). In these provinces, Xinjiang contributed the most in alleviating the pressure on water resources in the YRB over the years. In 2017, the YRB achieved a net inflow of 2.47 × 109 tons of virtual water from Xinjiang, accounting for 24.7% of the total net inflow of virtual water.

Overall, both the YRB and the eastern region had larger transfer of virtual water in all years. In 2017, for example, 1.8 × 1010 tons of virtual water flowed from the YRB to the eastern region, accounting for 64.1% of the total outflow of virtual water; meanwhile, 1.23 × 1010 tons of virtual water flowed from the eastern region to the YRB, accounting for 43.0% of the total inflow of virtual water. However, the transfer of virtual water between the YRB and the central and south-western regions were relatively small over the years. In 2017, the outflow of virtual water from the YRB to the central and south-western regions accounted for 6.8 and 17.1% of the total outflow, respectively; the inflow of virtual water from the central and south-western regions accounted for 12.7 and 15.4% of the total inflow, respectively.

Spatial-temporal evolution of value-added transfer

From 2002 to 2017, the number of target provinces for net value-added inflow in the YRB basically showed a decreasing trend. There were 16 provinces in 2002, 13 in 2007, 9 in 2012 and 11 in 2017. The target provinces in 2002 were mainly distributed in the southern region as well as the north-western region (Figure 5(a)) and further concentrated in the north-western, south-eastern coastal and central regions in 2007 (Figure 5(b)). In 2012, the target provinces were further reduced and more dispersed, including both developed provinces such as Zhejiang and Guangdong and underdeveloped ones such as Jilin, Xinjiang and Yunnan (Figure 5(c)). By 2017, the target provinces were mainly concentrated in Northwest China as well as the south of the Yangtze River (Figure 5(d)), from which the YRB benefited through interprovincial trade. From 2002 to 2017, the proportion of net value-added inflow in the YRB from Zhejiang and Guangdong rose from 36.4 to 48.9%. The YRB is a chemical, raw material and basic industrial base, mainly producing upstream and midstream products such as electricity, coke, chemical products and cement, which are essential intermediate inputs of manufacturing products. Therefore, the YRB obtains net economic benefits from Guangdong and Zhejiang by large sales of intermediate products.

Fig. 5.

Spatial-temporal evolution of value-added transfer between the YRB and other regions (2002–2017).

Fig. 5.

Spatial-temporal evolution of value-added transfer between the YRB and other regions (2002–2017).

Meanwhile, from 2002 to 2017, the number of target provinces for net value-added outflow in the YRB basically showed an increasing trend. There were 6 provinces in 2002, 9 in 2007, 13 in 2012 and 11 in 2017 (Figure 5). In 2002, the target provinces were mainly located in Northeast and North China (Figure 5(a)) and further expanded to the south-western region in 2007 (Figure 5(b)). In 2012, the target provinces further moved to the northeast, mainly concentrated in the central and eastern regions and municipalities (Figure 5(c)). By 2017, the target provinces were mainly concentrated in Northeast China, Beijing-Tianjin-Hebei and areas along the Yangtze River (Figure 5(d)). It is the eastern provinces that have the largest net economic benefits from the YRB through interprovincial trade. In 2017, the net value-added outflow from the YRB to the eastern region was 6.32 × 1011 RMB, accounting for 74.3% of the total net value-added outflow.

In general, the transfer of value-added between the YRB and the eastern region was large over the years. In 2017, for example, 3.1 × 1012 RMB of value-added flowed from the YRB to the eastern region, accounting for 64.0% of the total value-added outflow. Meanwhile, value-added from the eastern region was 2.93 × 1012 RMB in the YRB, accounting for 60.8% of the total value-added inflow. The YRB had closer economic connection with the eastern region. However, the transfer of value-added between the YRB and the central, western and north-eastern regions were small over the years. In 2017, the value-added outflow from the YRB to the central, western and north-eastern regions accounted for 11.3, 13.8 and 10.9% of the total outflow, respectively; the value-added inflow from the central, western and north-eastern regions accounted for 10.4, 18.2 and 10.6% of the total inflow, respectively. The YRB had less economic connection with the central, western and north-eastern regions.

Industry characteristics of virtual water and value-added transfer

The three sectors with largest net virtual water outflow in the YRB over the years were S01, S22 and S14 (Figure 6(a)). In 2002, the net virtual water outflow from the above-mentioned sectors accounted for 82.1, 14.3 and 0.7% of the total net outflow in the YRB, respectively. In 2017, the net virtual water outflow from the above-mentioned sectors accounted for 71.4, 9.6 and 8.2% of the total net virtual water outflow, respectively. The production of some crops (e.g. cotton) has gradually shifted from the YRB to other dominant production areas (such as Xinjiang Province) since the 1990s, so the share of the net virtual water outflow in S01 decreased from 2002 to 2017. At the same time, virtual water of S01 mainly net flowed to Guangdong Province, the Yangtze River Delta and the Beijing-Tianjin region, and its share of the total net virtual water outflow increased from 56.4% in 2002 to 68.2% in 2017 (Figure 7(a)).

Fig. 6.

Net transfer of virtual water (a) (100 million tons) and value-added (b) (1000 million RMB) of 28 sectors in the YRB (2002–2017).

Fig. 6.

Net transfer of virtual water (a) (100 million tons) and value-added (b) (1000 million RMB) of 28 sectors in the YRB (2002–2017).

Fig. 7.

Net virtual water and value-added flows of major sectors in the YRB (2002–2017).

Fig. 7.

Net virtual water and value-added flows of major sectors in the YRB (2002–2017).

The sectors with the highest net inflow of virtual water in the YRB over the years were S06, S25, S16 and S28 (Figure 6(a)). In 2017, the net virtual water inflow from the above sectors in the YRB accounted for 80.2% of the total net inflow. From 2002 to 2017, the proportion of S25, S16 and S28 increased from 21.8, 9.6 and 7.6% to 28.2, 10.5 and 15.3%, respectively, but the proportion of S06 decreased slightly. Meanwhile, the virtual water in S25 had a net inflow mainly from Guangdong, Jiangsu, Hebei, Anhui, Heilongjiang and Guangxi (Figure 7(b)).

In 2017, those three sectors with the largest net inflow of value-added were S02, S01 and S14 (Figure 6(b)). Especially, from 2002 to 2017, the net inflow of value-added in S02 increased from 3.11 × 1010 RMB to 5.51 × 1011 RMB, and the proportion increased from 11.5 to 18.7%. Meanwhile, the net inflow of value-added in S02 was mainly from Jiangsu, Zhejiang, Guangdong and Hebei, and the proportion increased from 46.9% in 2002 to 51.1% in 2017 (Figure 7(d)). Overall, the net inflow of value-added in the YRB is mainly in metal smelting, mining and agriculture in the upstream. As an important industrial base for energy and a grain-producing region in China, the YRB has achieved net economic benefits while providing energy products and agricultural products to other provinces.

The major sectors in the YRB with net value-added outflow over the years were S25, S16 and S17 (Figure 6(b)). From 2002 to 2017, the net value-added outflow in S25 increased from 1.09 × 1011 RMB to 1.29 × 1012 RMB, and its share fluctuated from 41.2 to 43.1%, with the net outflow provinces mainly including Guangdong, Jiangsu, Zhejiang, Hebei and Liaoning (Figure 7(c)). Among them, Jiangsu, Zhejiang and Guangdong are large construction provinces, and their construction outputs in 2019 ranked among the top five in China.

Inequality of virtual water consumption and economic benefits between the YRB and other regions

We use the RI index to reflect the relative inequality between the YRB and other regions. The larger the absolute value of RI index we get, the more serious the relative inequality is. From 2002 to 2017, the number of province pairs with the absolute values of index greater than 1 was relatively stable, with 9 pairs in 2002, 10 in 2007 and 2012 and 11 in 2017 (Figure 8(a)). The YRB faces a win-win or lose-lose situation in trade with these provinces. The provinces that make the YRB face a lose-lose situation are mainly the developed eastern region, especially the megacities such as Beijing and Shanghai. These regions not only receive net inflow of virtual water from the YRB but also gain net economic benefits due to their comparative advantages in the industrial structure. In 2017, the highest RI index in this category was the YRB Beijing (RI = 2.55). The YRB suffered the most serious inequality, and Beijing was the only province that received net virtual water and value-added from the YRB over the years. During 2002–2017, the above provinces fluctuated from 3 to 5. The provinces that make the YRB face a win-win situation are mainly in the less developed central, western and north-eastern regions which all achieve a net transfer of virtual water and value-added to the YRB simultaneously. In 2017, the lowest RI index in this category was Xinjiang Province YRB (RI = −3.00).

Fig. 8.

Regional inequality index (a) and inequality index by industry (b) in the YRB (2002–2017).

Fig. 8.

Regional inequality index (a) and inequality index by industry (b) in the YRB (2002–2017).

Additionally, the RI index between the YRB and other provinces ranges between −1 and 1, indicating the opposite direction of net virtual water and value-added flows. Among the provinces with net outflow of virtual water and net inflow of value-added in the YRB (RI index between 0 and 1), the highest RI index in 2017 was the YRB Zhejiang Province (RI = 1.00). The YRB had a net transfer of 2.23 × 109 tons of virtual water to Zhejiang Province, accounting for 23.9% of the total net virtual water outflow but gained 8.23 × 1010 RMB of net value-added from Zhejiang Province, accounting for only 10.3% of the total net value-added inflow. This indicates that the economic compensation received is not equivalent to the net outflow of virtual water. From 2002 to 2017, the number of such provinces fluctuated from 10 to 5. In particular, Zhejiang and Guangdong all belonged to this category over the years. Among the provinces with net inflow of virtual water and net outflow of value-added (RI between −1 and 0), the lowest index in 2017 was Hebei Province YRB (RI = −1.00). The YRB had a net inflow of 160 million tons of virtual water from Hebei Province, accounting for 1.6% of the total net virtual water inflow, but it transferred 1.05 × 109 RMB of value-added to Hebei Province, accounting for only 0.1% of the total net value-added outflow. From 2002 to 2017, the number of such provinces fluctuated from 3 to 6. More and more provinces had net outflows of virtual water to the YRB, which helped to alleviate water stress. However, the YRB also had net outflow of value-added to more provinces.

Figure 8(b) reflects the inequality index of each industry in the YRB. From 2002 to 2017, the number of industries with inequality index less than −1 fluctuated from 8 to 4. Especially, S10 belonged to this category over the years. It achieved the net inflow of virtual water and value-added through interprovincial trade, which not only relieved water stress but also brought net economic benefits. Additionally, from 2002 to 2017, the number of industries with net virtual water inflow and net value-added outflow (inequality index between −1 and 0) decreased from 13 to 11, and mainly concentrated in S06, S08, S09, S25 and S16–S20 (high-end manufacturing). From 2002 to 2017, the number of industries with net virtual water outflow and net value-added inflow (inequality index between 0 and 1) fluctuated from 7 to 13, and mainly concentrated in S01, S22, S14, S02–S05 (mineral extraction), etc. These industries are the pillar industries in the YRB and have high water consumption. Especially, agriculture suffered the most serious relative inequality, with the inequality index value of 1 in all years. In 2017, S01 had a net outflow of 1.65 × 1010 tons of virtual water to other regions, accounting for 71.4% of the total net virtual water outflow, while it had a net inflow of 3.5 × 1011 RMB of value-added, accounting for only 11.9% of the total net value-added inflow. It is urgent to take measures to reduce agricultural water consumption in the YRB.

Strategies for alleviating water crisis in the YRB from the perspective of sector classification

All sectors in the YRB have inconsistent net transfer of virtual water and value-added in interprovincial trade. Therefore, the industrial structure of the YRB should be optimised to reduce the net outflow of virtual water while maintaining economic development as much as possible. Therefore, the sectors are divided into three categories with different industrial policies based on the RI index in 2017.

Specifically, the first category is industries with net inflow of both virtual water and value-added (inequality index less than −1). In 2017, such industries included S07, S10, S23 and S24. The YRB should focus on the development of the above industries and the implementation of certain preferential measures to guide these industries to increase investment and expand their market share in interprovincial trade. The second category is industries with net virtual water outflow and net value-added inflow (inequality index between 0 and 1). In 2017, such industries included S01, S02, S03, S04, S05, S11, S12, S13, S14, S21, S22, S26 and S27 which are extremely dependent on natural resources. The YRB should guide such industries to reduce investment appropriately, to reduce the scale of interprovincial trade and to increase the amount of imports from other relatively advantageous regions. However, such industries are the pillar industries in the YRB, and they need to improve water use efficiency urgently. In particular, agriculture faced the most serious inequality. The YRB should popularise water-saving irrigation techniques and devices as well as reduce the irrigated area of the Yellow River Irrigation District. Meanwhile, the YRB should actively optimise agricultural structure, extend the agricultural industry chain, improve the value-added of products and give full play to the original comparative advantage on the basis of reducing virtual water consumption. The third category is industries with net virtual water inflow and net value-added outflow (inequality index between −1 and 0). In 2017, such industries included S06, S08, S09, S15, S16, S17, S18, S19, S20, S25 and S28. The YRB can guide the development of such industries, appropriately increasing the trade scale of provinces with relatively low technology levels and reducing those with relatively high technology levels.

Strategies for alleviating water crisis in the YRB from the perspective of regional trade

The inconsistency of the net transfer of virtual water and value-added has caused inequality between the YRB and other regions. In fact, the inequality is closely related to the imbalance of regional economic development that has emerged in China since the reform and opening up. Talents, capital and technology flow more freely within the national market with the acceleration of the regional economic integration process, enhancing the specialisation of each region but aggravating the unbalanced regional economic development. The YRB's advantageous industries are at the low end of the industrial chain and face great inequality in participating in interprovincial trade. Nowadays, when interprovincial trade is becoming more frequent and water resources in the YRB are increasingly scarce, the trade scale should be taken seriously to alleviate the water resource crisis in the YRB.

For the provinces with RI index less than −1, the YRB should implement trade preferential policies and actively adjust to expand trade with such provinces in order to achieve more net inflow of virtual water and value-added. In 2017, such provinces included Heilongjiang, Jiangxi, Hunan, Guangxi, Guizhou and Xinjiang. Meanwhile, because the RI index between Xinjiang and YRB is the lowest, the YRB should actively lead to expand the trade scale with this province. For the provinces with RI index between −1 and 0, the YRB can guide moderate expansion of trade with them. In 2017, such provinces included Hebei, Jilin, Jiangsu, Anhui, Hubei and Sichuan. In particular, the YRB should more moderately increase trade with Hebei Province (RI = −1.00). For the provinces with RI index between 0 and 1, they exchange lower import price costs for higher water inputs in the YRB, which makes the YRB also face certain inequality. Furthermore, backward regions in the upstream of the YRB (such as Qinghai, Ningxia and Gansu), which are originally short of water, have a strong will for economic development and are more likely to promote economic growth at the expense of water resources. However, they are key areas for ecological protection in the YRB with greater pressure for water control. Therefore, the YRB should perfect the trading market of water rights and guide the moderate reduction of the trade scale with such provinces. In 2017, such provinces included Zhejiang, Guangdong, Fujian, Hainan and Yunnan. For the provinces with RI index greater than 1, the YRB has the net outflow of virtual water and value-added to them. In 2017, such provinces included Beijing, Tianjin, Shanghai, Chongqing and Liaoning, which were basically developed regions. The YRB can moderately adjust to reduce the scale of trade with them. However, due to the increasing demand for products from these regions, they should share the responsibility for water use with the YRB under the premise of a comprehensive consideration of the net transfer of virtual water and value-added. Specifically, national or provincial governments can establish a joint interregional mechanism, with the above developed regions providing finance, technology and advanced devices to the YRB in order to improve water use efficiency in the YRB.

Optimisation strategies for specific industries and major target provinces

In order to better guide the YRB to mitigate the water crisis, we further propose optimisation strategies for specific sectors and major target provinces. Since S01 and S25 are the sectors with the highest net virtual water outflow and inflow in 2017, we will further discuss them.

The optimisation measures of S01 (agriculture)

The sector with the largest net virtual water outflow in the YRB was S01 over the years. In 2017, the top five target provinces of net virtual water outflow from S01 were Guangdong, Zhejiang, Beijing, Chongqing and Shanghai, accounting for 18.9, 15.0, 10.3, 8.7 and 8.1%, respectively. Meanwhile, the above provinces were just the top five provinces of net value-added inflow in the YRB, with their contribution rates of 18.7, 15.1, 9.9, 8.2 and 7.0%, respectively (Figure 9(a)). S01 in the YRB had a net inflow of value-added in trade with Guangdong, Beijing, Chongqing and Shanghai but had relatively more net outflow of virtual water, which exacerbated the water crisis of the YRB. Therefore, the YRB should be guided to moderately reduce the trade scale in S01 with the above provinces. Additionally, S01 faces a similar situation in trading with 11 provinces including Jiangsu, Tianjin, Hebei, Liaoning, Jilin, Anhui, Sichuan, etc. Meanwhile, S01 should also actively improve water use efficiency to reduce the net outflow of virtual water. Nevertheless, in the agricultural trade between the YRB and Yunnan, Guangxi, Fujian, Zhejiang, Heilongjiang, the proportions of net virtual water outflow were smaller than those of net value-added inflow. The YRB can moderately increase agricultural imports from these regions.

Fig. 9.

Proportion of the net transfer of virtual water and value-added embodied in trade between S01, S25 in the YRB and other provinces in 2017.

Fig. 9.

Proportion of the net transfer of virtual water and value-added embodied in trade between S01, S25 in the YRB and other provinces in 2017.

In 2017, the proportions of net virtual water outflow and net value-added inflow in trade between the YRB and Jiangxi Province in S01 were −0.08 and −0.07%, respectively (Figure 9(a)). The YRB can moderately expand trade with Jiangxi. In 2017, in the agricultural trade between the YRB and Xinjiang, the proportion of net virtual water outflow was −3.6%, and the proportion of net value-added inflow was 0.3% (Figure 9(a)). The YRB had the net inflow of both virtual water and value-added from Xinjiang, which not only gained net economic benefits but also alleviated the pressure on water resources. Therefore, the YRB should take preferential measures to expand the trade scale with Xinjiang actively and increase the import of agricultural products from Xinjiang.

The optimisation measures of S25 (construction)

In 2017, the largest net inflow of virtual water in the YRB was S25. The three provinces with the largest net outflow of virtual water to S25 in the YRB were Jiangsu, Guangdong and Xinjiang, with contribution rates of 14.3, 10.9 and 6.9%, respectively. Three target provinces with the largest net outflow of value-added in the YRB were Jiangsu, Zhejiang and Guangdong, with contribution rates of 10.9, 10.4 and 9.0%, respectively (Figure 9(b)). The YRB had the larger share of net inflow of virtual water than that of net outflow of value-added in construction trade with Jiangsu and Guangdong. The YRB receives relatively more net virtual water inflow at a smaller trade cost, which is helpful to relieve the water crisis. Therefore, the YRB should be guided to moderately expand the trade scale with Jiangsu and Guangdong and similarly with 10 other provinces, including Guangxi, Jilin, Sichuan, Heilongjiang, etc. However, in 2017, the proportion of net virtual water inflow was 6.2% in construction trade between the YRB and Zhejiang, but the proportion of net value-added outflow was much larger than that of net virtual water inflow in the YRB (Figure 9(b)). This means that the technical level of the YRB is relatively backward, and the YRB is in a disadvantageous position. Nine provinces, including Beijing, Tianjin, Shanghai, Hebei, etc., also fall into this category. Therefore, the YRB should be guided to appropriately reduce the import of related products from these provinces.

Using the interregional input–output tables of China in 2002, 2007, 2012 and 2017, we analyse production-based and consumption-based water consumption and value-added in the YRB, calculate the net transfer of virtual water and value-added, and further reveal the resulting inequality. The conclusions are as follows.

First, from 2002 to 2017, the YRB achieved a shift in virtual water and value-added from production side over consumption side to consumption side over production side. Generally speaking, its water stress was alleviated through domestic trade, but its economic situation deteriorated. Meanwhile, only S01 and S22 in the YRB had PV values greater than CV values over the years, and their water use efficiency needs to be improved urgently.

Second, from 2002 to 2017, the number of target provinces for net outflow of virtual water in the YRB dropped from 13 to 10, with a greater concentration in the Beijing-Tianjin region and developed south-eastern coastal provinces. Among all sectors in the YRB, agriculture had the largest net outflow of virtual water in all years, mainly to Beijing-Tianjin, Yangtze River Delta and Guangdong. Meanwhile, some underdeveloped central and western regions have net outflow of virtual water to the YRB in interregional trade. From 2002 to 2017, the number of targeted provinces for the net inflow of the value-added in the YRB reduced from 16 to 11. The YRB profits from both developed provinces such as Zhejiang and Guangdong and underdeveloped regions such as Xinjiang. The sector with the largest net outflow of value-added over the years was S25, and the net outflow of S25's value-added was mainly to Guangdong, Jiangsu, Zhejiang, Hebei and Liaoning.

Third, according to the difference of the net transfer of virtual water and value-added, there are four types of inequality between the YRB and other regions. Type I refers to the RI index which is over 1, and the YRB faces a lose-lose situation. From 2002 to 2017, the number of provinces in this category showed a fluctuating upward trend, which mainly included megacities such as Beijing and Shanghai. Type II refers to the RI index which is less than −1, and the YRB faces a win-win situation. Provinces in this category are mainly concentrated in underdeveloped central and western regions such as Xinjiang and Jiangxi. Type III refers to the RI index between 0 and 1, and the YRB is in a relatively inferior position due to more net outflow of virtual water and less net inflow of value-added. From 2002 to 2017, the number of provinces in this category showed a fluctuating downward trend, which mainly included Zhejiang, Guangdong and so on. Type IV refers to the RI index between −1 and 0. The YRB, with more net inflow of virtual water and less net outflow of value-added, is in a relatively dominant position. The number of provinces in this category fluctuated from 3 to 6. Furthermore, in terms of sectors, from 2002 to 2017, more and more sectors in the YRB had net virtual water outflows and net value-added inflows. Among them, agriculture had the most serious inequality.

This paper aims to reveal the inequality faced by the YRB in interprovincial trade, which can provide a certain scientific basis for alleviating the pressure on water resources in the process of rapid economic development in the YRB. However, the paper also has the following shortcomings: first, the study of inequality between the YRB and other regions only involves one element which is water resources. In fact, the ecosystem contains many elements, and there are complex interactions among them. The equality represented by a single element is limited and one-sided (Wu et al., 2009). In the future, a coupled analysis of multiple elements including water resources can be conducted to comprehensively evaluate the inequality between the YRB and other regions. Second, the YRB engages in interprovincial trade as well as participates in global trade, but we only focus on domestic trade and do not involve the analysis of virtual water and value-added transfer between the YRB and the rest of the world. The YRB can be included in the world input–output tables in future studies to analyse the inequality.

The authors are grateful for financial support from the National Natural Science Foundation of China under Grant No. 71773118.

All relevant data are included in the paper or its Supplementary Information.

Allan
J. A.
, (
1993
).
Fortunately There Are Substitutes for Water Otherwise Our Hydro-Political Futures Would Be Impossible, Priorities for Water Resources Allocation and Management
.
Overseas Development Administration
,
London
,
UK
, pp.
13
26
.
Cai
B.
,
Wang
C.
&
Zhang
B.
, (
2017a
).
Worse than imagined: unidentified virtual water flows in China
.
J. Environ. Manage.
196
,
681
691
.
Chen
Y.
,
Fu
B.
,
Zhao
Y.
,
Wang
K.
,
Zhao
M.
,
Ma
J.
,
Wu
J.
,
Xu
C.
,
Liu
W.
&
Wang
H.
, (
2020
).
Sustainable development in the Yellow River Basin: issues and strategies
.
J. Clean. Prod.
263
,
121223
.
Feng
K.
,
Davis
S. J.
,
Sun
L.
,
Li
X.
,
Guan
D.
,
Liu
W.
,
Liu
Z.
&
Hubacek
K.
, (
2013
).
Outsourcing CO2 within China
.
Proc. Natl. Acad. Sci. USA
110
(
28
),
11654
11659
.
Feng
K.
,
Hubacek
K.
,
Pfister
S.
,
Yu
Y.
&
Sun
L.
, (
2014
).
Virtual scarce water in China
.
Environ. Sci. Technol.
48
(
14
),
7704
7713
.
Feng
Z.
&
Liu
F.
, (
2019
).
Ecological unequal exchange, value transfer and environmental problems in developing economies
.
J. World Econ.
42
(
4
),
3
28
(in Chinese)
.
Frey
R. S.
,
Gellert
P. K.
&
Dahms
H. F.
, (
2019
).
Introduction: ecologically unequal exchange in comparative and historical perspective
. In:
Ecologically Unequal Exchange
.
Palgrave Macmillan
,
Cham, Switzerland
, pp.
1
10
.
Guan
D.
&
Hubacek
K.
, (
2007
).
Assessment of regional trade and virtual water flows in China
.
Ecol. Econ.
61
(
1
),
159
170
.
Hornborg
A.
&
Martinez-Alier
J.
, (
2016
).
Ecologically unequal exchange and ecological debt
.
J. Polit. Ecol.
23
(
1
),
328
333
.
Liang
S.
,
Zhang
C.
,
Wang
Y.
,
Xu
M.
&
Liu
W.
, (
2014
).
Virtual atmospheric mercury emission network in China
.
Environ. Sci. Technol.
48
(
5
),
2807
2815
.
Liao
X.
,
Zhao
X.
,
Hall
J. W.
&
Guan
D.
, (
2018
).
Categorising virtual water transfers through China's electric power sector
.
Appl. Energy
226
,
252
260
.
Liu
J.
&
Yang
W.
, (
2012
).
Water sustainability for China and beyond
.
Science
337
(
6095
),
649
650
.
Omer
A.
,
Elagib
N. A.
,
Zhuguo
M.
,
Saleem
F.
&
Mohammed
A.
, (
2020
).
Water scarcity in the Yellow River Basin under future climate change and human activities
.
Sci. Total Environ.
749
,
141446
.
Researching Group of Chinese Input-Output Association
(
2007
).
Input-output analysis of water resources consumption and water input coefficient in national economic sectors: the fifth of researching report series on input-output tables of 2002
.
Stat. Res.
24
,
20
25
(in Chinese)
.
Ringler
C.
,
Cai
X.
,
Wang
J.
,
Ahmed
A.
,
Xue
Y.
,
Xu
Z.
,
Yang
E. T.
,
Zhao
J.
,
Zhu
T.
,
Cheng
L.
,
Fu
Y.
,
Fu
X.
,
Gu
X.
&
You
L.
, (
2010
).
Yellow River basin: living with scarcity
.
Water Int.
35
(
5
),
681
701
.
Seekell
D. A.
,
D'odorico
P.
&
Pace
M. L.
, (
2011
).
Virtual water transfers unlikely to redress inequality in global water use
.
Environ. Res. Lett.
6
(
2
),
024017
.
Wang
X.
, (
2014
).
Research on distribution and tendency of virtual water trade in China based on inter-regional input-output model
.
Manage. Rev.
26
(
7
),
46
54
(in Chinese)
.
Wang
G.
,
Zhang
J.
,
Jin
J.
,
Weinberg
J.
,
Bao
Z.
,
Liu
C.
,
Liu
Y.
,
Yan
X.
,
Song
X.
&
Zhai
R.
, (
2017
).
Impacts of climate change on water resources in the Yellow River basin and identification of global adaptation strategies
.
Mitig. Adapt. Strat. Gl.
22
(
1
),
67
83
.
Wang
X.
,
Zhang
J.
,
Shahid
S.
,
Yu
L.
,
Xie
C.
,
Wang
B.
&
Zhang
X.
, (
2018a
).
Domestic water demand forecasting in the Yellow River basin under changing environment
.
Int. J. Clim. Chang. Strateg. Manag.
10
(
3
),
379
388
.
Wang
Z.
,
Yang
Y.
&
Wang
B.
, (
2018b
).
Carbon footprints and embodied CO2 transfers among provinces in China
.
Renew. Sustain. Energy Rev.
82
,
1068
1078
.
Wang
F.
,
Cai
B.
,
Hu
X.
,
Liu
Y.
&
Zhang
W.
, (
2021
).
Exploring solutions to alleviate the regional water stress from virtual water flows in China
.
Sci. Total Environ.
796
,
148971
.
Wei
W.
,
Hao
S.
,
Yao
M.
,
Chen
W.
,
Wang
S.
,
Wang
Z.
,
Wang
Y.
&
Zhang
P.
, (
2020
).
Unbalanced economic benefits and the electricity-related carbon emissions embodied in China's interprovincial trade
.
J. Environ. Manage.
263
,
110390
.
Wu
C.
,
Yao
Z.
,
Li
Y.
&
Zhong
F.
, (
2009
).
An overview and progress of the study of environmental equity
.
Adv. Earth Sci.
24
(
11
),
1268
1274
(in Chinese)
.
Xu
Z.
,
Takeuchi
K.
,
Ishidaira
H.
&
Liu
C.
, (
2005
).
An overview of water resources in the Yellow River Basin
.
Water Int.
30
(
2
),
225
238
.
Yang
H.
&
Zehnder
A.
, (
2007
).
‘Virtual water’: an unfolding concept in integrated water resources management
.
Water Resour. Res.
43
,
W12301
.
Yellow River Conservancy Commission
. (
2020
).
Yellow River Water Resources Bulletin 2019
.
Zhang
Q.
,
Xu
C.
&
Yang
T.
, (
2009
).
Variability of water resource in the Yellow River basin of past 50 years, China
.
Water Resour. Manag.
23
(
6
),
1157
1170
.
Zhang
J.
,
Chen
G.
,
Xing
S.
,
Shan
Q.
,
Wang
Y.
&
Li
Z.
, (
2011
).
Water shortages and countermeasures for sustainable utilisation in the context of climate change in the Yellow River Delta region, China
.
Int. J. Sustain. Dev. World Ecol.
18
(
2
),
177
185
.
Zhang
X.
,
Liu
J.
,
Tang
Y.
,
Zhao
X.
,
Yang
H.
,
Gerbens-Leenes
P. W.
,
van Vliet
M. T. H.
&
Yan
J.
, (
2017
).
China's coal-fired power plants impose pressure on water resources
.
J. Clean. Prod.
161
,
1171
1179
.
Zhang
W.
,
Liu
Y.
,
Feng
K.
,
Hubacek
K.
,
Wang
J.
,
Liu
M.
,
Jiang
L.
,
Jiang
H.
,
Liu
N.
,
Zhang
P.
,
Zhou
Y.
&
Bi
J.
, (
2018
).
Revealing environmental inequality hidden in China's inter-regional trade
.
Environ. Sci. Technol.
52
(
13
),
7171
7181
.
Zhao
H.
,
Zhang
Q.
,
Guan
D.
,
Davis
S. J.
,
Liu
Z.
,
Huo
H.
,
Lin
J.
,
Liu
W.
&
He
K.
, (
2015
).
Assessment of China's virtual air pollution transport embodied in trade by using a consumption-based emission inventory
.
Atmos. Chem. Phys.
15
(
10
),
5443
5456
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial purposes with no derivatives, provided the original work is properly cited (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Supplementary data