An Analysis of Water Use Efficiency of Staple Grain Productions in China: Based on the Crop Water Footprints at Provincial Level

Abstract

This article analyzed the water use efficiency of China’s staple grain productions (maize, rice, and wheat). This research calculated the water footprints of crop production using the CROPWAT model based on reported water use in 2000 and during 2015–2019, and both green and blue water footprints were calculated. The results showed that compared to 2000 water use efficiency of crop production for maize, rice and wheat during 2015–2019 were increased by about 12.4%, 10.8%, and 2.5% respectively. The current spatial structure of the stable grain industry that makes up grain production regions are concentrated in northern China, then grains are distributed across China (North-to-South Grain Transportation). This has advantages in the increase of agricultural water productivity. This research indicated that agricultural policies should further consider an advance of regional grain production, along with the optimization of transportation channels for stable grains to increase irrigation water use efficiency. The establishment of staple grain production in designated areas not only ensures China’s food security, but also promotes the sustainable use of irrigation water resources.

1. Introduction

Over the past 20 years, the global freshwater inventory has declined by 20%. The world is currently facing severe water shortages; about 1.2 billion people now live in areas with severe water shortages, of which about 520 million people live in agricultural areas and 650 million people live in small cities surrounded by agricultural lands [1]. It is imperative to improve the efficiency of water resource utilization. As the world’s largest consumption area, the use and evaporation path of agricultural water deserves more attention.

In 1949, China had a weak agricultural base and low food production capacity. After 70 years of effort, China has basically achieved self-sufficiency in food, solved the food shortage problem affecting nearly 1.4 billion people, and guaranteed the nutritional level and quality of life for its residents. In the early days of the People’s Republic of China, natural conditions in major cities, such as Beijing and Tianjin in the northern part of China and industrial areas such as Heilongjiang, Jilin, and Liaoning in the northeastern region, saw far more food shortages than places in the South, which led to a long period of low grain yields and a tight self-supply of grains in the North. Ultimately, the southern provinces, with superior natural environmental conditions, produced grains for the northern ones.

Most of the oldest EU countries have a more efficient and optimized crop production process in terms of resource savings and output maximization [2], so as to China. As the economic development of the southern region expanded, the usage of cultivated land gradually shifted in the south of China. The northern hemisphere farming belt gradually grew even more northward, which was caused by climate change. Therefore, China’s grain cultivation gradually decreased in the South and stabilized in the North. From 1984 to 1994, the grain cultivation region in southern China decreased by about 50 million hectares. In order to ensure food self-sufficiency and security, China drew a red line over the protection of cultivated lands, strictly controlled the occupation of cultivated lands for development and construction, and implemented a policy to ensure that cultivated lands could reach more than 120 million hectares, thereby consolidating a foundation for food planting.

Currently, with the need for economic development increasing and the changes affecting regional divisions of labor, grain continues to be distributed from northern to southern regions. In recent years, Guangdong, Zhejiang, and Fujian were ranked as the top three provinces for grain shortage. Guangdong’s grain self-sufficiency rate was about 32%, with 12 million tons of grain purchased from other provinces each year, accounting for 5% of the national grain distribution. Zhejiang’s annual grain sales gap was 13 million tons, and Fujian’s annual grain import was also 12 million tons. At the same time, China’s grain production in northern regions in 2020 accounted for a total of 59.22%. The Chinese government introduced a system of responsibility concerning food security for provincial governors, which has strengthened the governor’s supervision of food production safety, efficiency, and preservation as well as promoting food security to a prominent position in economic and social development. Provincial food security has been equally important. The government has required all provinces to ensure a certain amount of major crop production and interprovincial self-sufficiency in grain.

In order to consolidate the basis of agricultural production and solve the contradiction between industrialization, urbanization development, and agricultural land use, China has enacted a policy to establish crop production areas, which defined the production of certain types of agricultural products in designated areas and maintained the self-sufficiency level of important agricultural products to ensure food security. Would North-to-South Grain Transportation be possible for China’s agricultural production to not only meet developmental needs but to also save water resources effectively? Could the food production areas ensure food security while accounting for sustainable development?

This article analyzed the efficiency of China’s agricultural water usage from a new perspective of its water footprint and answered the following questions by calculating and analyzing the water footprint of the staple grains (maize, rice, and wheat) in 31 provinces in China:

(1) What is the status of the provincial water footprint of staple grains in 2015–2019?

(2) From the perspective of the water footprint of production, specifically the blue and green water footprints of grain crops, is China’s utilization of agricultural water efficient in the production of staple grains?

(3) From the perspective of regional production, can the North-to-South Grain Transportation and the food production areas both meet the food demand of various regions and save water resources at the same time?

The current article evaluated the efficiency of China’s agricultural water usage and analyzed the effectiveness of two agricultural policies. The article begins with the introduction of the water footprint theory and the current stage of water footprint-related research, and introduces the data sources and the calculation methods of the water footprints in this article. Then, we calculated the production water footprints as well as the blue and green water footprints of staple grain production in various provinces in China in recent years and provided the calculation results. Next, according to the calculation results of the water footprint, we discussed the efficiency of China’s agricultural water resource utilization and analyzed the effectiveness of the current phenomenon of North-to-South Grain Transportation in China and the policy of establishing grain production areas. This article ends with an analysis of the possibility of efficiency optimization.

2. Literature Review

The use and evaporation path of agricultural water, as the world’s largest water consumption area, has always been a focus of researchers. There are many ways to improve the efficiency of agricultural water resource usage. Barsha et al. analyzed reliable ways to increase water resource productivity from the perspective of the interaction between water and fertilizer, suggesting that by increasing nutrient supply in fertilizers through improving fields, farms, and irrigation conditions, yields can be improved, and, therefore, the productivity of water resources can be increased [3]. With a limited land supply and a significant lack of water resources, increasing water resource utilization efficiency is the only way to solve the problem. Koech et al. explored ways to improve agricultural water use efficiency from a regulatory perspective, suggesting that farmers were reluctant to use new water-saving irrigation methods because of the elevated costs of new irrigation technologies, and the government should introduce incentives, oversights, and appropriate regulations for the use of irrigation water and irrigation equipment to promote the effectiveness of new technologies in improving water efficiency [4]. Razzaq et al. conducted the economic analysis of high efficiency irrigation systems in the Punjab province of Pakistan. They found that the users of high-efficiency irrigation systems (sprinkler and drip irrigation) earned higher gross margins, which made use of available water resources more efficiently [5]. Data envelopment analysis (DEA) is a common method of efficiency analysis. Razzaq et al. demonstrate how groundwater markets can improve efficiency in agricultural production. They used a DEA model to estimate the water use efficiency of water buyers, self-users cum sellers, and self-users in groundwater markets, and found that participation in water markets and access to extension services can improve water use efficiency [6]. Gadanakis et al., using the United Kingdom as a research object, analyzed the efficiency of water use by building a DEA model and found that there was a large amount of excess agricultural water in the UK’s farm operations that could be improved by modifying irrigation methods such as employing drip irrigation or adding recycling systems [7]. Deng et al. built the SBM-DEA model to estimate the efficiency of water usage in China from 2003 to 2013 using panel data and explored the factors affecting the efficiency of water usage. They showed that economically developed areas such as Beijing, Shanghai, and Tianjin have higher water use efficiency. They also found that increased agricultural productions, water consumptions per capita, and the proportion of sewage per unit of output had negative impacts on water resource efficiency while import and export dependence have positive impacts [8].

Currently, as the largest grain producer, China’s total grain production in 2020 was about 669.49 million tons, which consumed 361.24 billion cubic meters of agricultural water. In the face of such massive water consumption, improving the efficiency of agricultural water use has always been a focus of researchers. The water footprint of crop consumption and the import of virtual water increased during 1960–1979 and decreased during 1980–2010. They fluctuated but tended to increase during this period and were influenced mainly by agricultural factors such as crop yield, irrigation efficiency, and area sown [9]. In a study at the regional level, Geng et al. constructed a DEA model to analyze the state of China’s agricultural development from the perspective of blue and green water, Wang et al. used the same model to examine agricultural water use efficiency and related issues in the Heihe River Basin from 2004 to 2012. The results showed that the pure technical efficiency of northern China is higher than that of the southern region. Improving the scale effect can effectively improve China’s water use efficiency [10,11]. Cao et al. quantified the water use and productivity in grain production for 31 Chinese provinces, autonomous regions, and municipalities from 1998 to 2010. The study showed a higher increase in crop yield on irrigated land compared to rain-fed land from the perspective of blue water and green water. The northeast provinces urgently need to improve irrigation efficiency, and the North China Plain should promote rain-fed crop yield to increase grain production and control water use in the future [12]. Wang et al. used random boundary analysis to evaluate the efficiency of agricultural water use in 31 provinces of China from 2000 to 2015. In the past 15 years, China’s agricultural water use efficiency had improved; however, a clear spatial correlation and an imbalance in the development of the water use efficiency at the provincial level still existed [13]. Sun et al. came to similar conclusions. They analyzed the efficiency of water resource utilization and its drivers in China over the past three decades. They found that the water use had increased in northeast, southwest, and central China over the past three decades while the utilization in the western region had declined. The annual precipitation, the average annual temperature, and the average leaf area index were the main factors affecting water resource utilization [14]. In crop research, Xu et al. have explored the efficiency of water resource utilization in wheat production in the North China Plain by comparing six different irrigation methods. Experimental results showed that, combined, irrigation methods and 150 mm of irrigation could maximize the utilization of water usage for wheat cultivation in the North China Plains [15]. Wang et al. conducted a study on the water efficiency of rice cultivation in northeast China and found that the “evaporation pressure”, caused by crop and soil evaporation, had a nonlinear effect on the efficiency of water resource utilization. When the pressure was insufficient, a higher water resource utilization efficiency was available [16]. Greaves et al. used the AquaCrop model to examine the impact of different irrigation scheduling options on yields to identify viable strategies to enhance water use efficiency for irrigated maize. They found that depletion levels of 40–50% of total available water at water depths of 20–40 mm could be used to obtain high water use efficiency without a significant yield penalty [17].

According to the facts, the North-to-South Grain Transportation and the food production areas both strengthen food security, but production efficiency was not in an optimal state. Zhang et al. conducted research into China’s grain production efficiency using a spatial measurement model and a Markov chain. The results revealed two shortages in China’s current production efficiency: growth momentum and inadequate space [18]. Zhang et al. established the DEA global Malmquist productivity index model to analyze China’s grain production efficiency from 2001 to 2017. They found that China’s grain production efficiency has been declining from the central region to the eastern and western regions [19]. The phenomenon of North-to-South Grain Transportation was a product of China’s rapid economic development, and it has been the best means to adapt to the current economic growth and to meet the population’s needs. However, China’s South-to-North Water Diversion, a water infrastructure framework, can overcome the challenges of water shortages by meeting long-term social, economic and environmental goals for water resource systems in northern China [20].

China enacted policies, i.e., the North-to-South Grain Transportation and the food production areas, to consolidate the basis of agricultural production and to achieve self-sufficiency in food. The theory of water footprints combined the real and virtual water usage in daily life, which represent both natural water use efficiency and irrigation water use efficiency. Previous studies have focused on estimating China’s agricultural water use efficiency or evaluation of policies effects, but they did not relate the water footprints theory to these two policies. From the perspective of water footprint, this paper estimated the water use efficiency of two policies and proposes ways to improve the policies.

3. Methodology

3.1. Virtual Water and Water Footprints According to Allan and Hoekstra

The theory of water footprints derived from the theory of virtual water, which was proposed by J.A. Allan in 1993. He defined virtual water as water resources that are invisible in the production of goods or services [21]. The research of virtual water was initially concentrated in the field of agricultural production [22], and, over time, the theory of virtual water was gradually applied to the research of nonagricultural products [23]. Based on the concept of virtual water and the theory of ecological footprints, Hoekstra proposed the water footprint theory, which defined the water footprint as the amount of water required to consume all products and services over a period of time, which usually included food, daily necessities, water used in daily life, and the living environment of any known entity (a person, a region, or a country). The theory of water footprints included not only the direct use of water by an individual, but also their indirect water use, which represented the amount of water required for all products or services consumed by an individual over a defined period of time as well as the amount of water required for the production of the consumed products or services [24]. It combined the real and virtual water usage in daily life, which provided a new perspective for research on the possession and the consumption of water resources as well as a new way to explore the efficiency of water resources utilization and, perhaps, to improve it.

3.2. Classification of Water Footprints

According to different evaluation systems, water footprints can be divided into the water footprints of production, the water footprints of consumption, the regional water footprints, and the industrial water footprints. Depending on the source of the water resources, the water footprint can be divided into blue and green. Blue water refers to the surface and groundwater in the aquifer of the land as well as the water resources in rivers and lakes. The blue water footprint includes the consumption of blue water in the production and consumption process. In order to distinguish water sources, Falkenmark proposed the concept of green water, which refers to water that does not come from a river or lake, is stored in unsaturated aquifers of soil, or is consumed in the form of evaporation. The green water footprint includes the consumption of green water during production or consumption [25].

3.3. Methods of Calculating Crop Water Footprints

Crop water footprints refer to the water footprint of the production of crops, and it is divided by the crop consumption of blue and green water resources in the production process based on crop yield. The sum of the consumption of blue and green water is equal to the evaporation of water resources in crop production, wherein blue water refers to the evaporation of irrigation water, and green water refers to the evaporation of effective precipitation. Various methodological approaches for water footprints have been suggested, which differ significantly regarding scope, information value, relevance, and data requirements [26]. Due to the high cost of directly measuring the evaporation of water resources, the theoretical method of measuring the water footprint of crops has generally been applicable. To facilitate research, therefore, researchers have suggested two common methods for calculating crop water footprints based on empirical formula models. The first was the irrigation scheduling method proposed by Mekonnen and Hoekstra. This method uses CROPWAT to calculate the total water storage and the local effective precipitation (green water) during crop growth, and then obtains the water demand (blue water) for crop irrigation during production by calculating differential results according to the principle of field water balance, and, finally, obtains the combined water footprint of the production and yield of the crop [27]. This method determines the distribution of the components of the water footprint and the trends in the water footprint over time, and can make simple predictions of future trends, if needed. The irrigation scheduling method has been a commonly used method of water footprint calculation. The second is the calculation of crop water demand proposed by Sun and Cao et al., which assumes that the amount of water vapor during crop growth can be fully satisfied, and there is no lack of moisture regardless of the actual soil water content [28,29]. The disadvantage of this method is that it does not consider the possibility of other water loss in crop growth other than evaporation, which leads to a deviation from the calculation results to the reality, and the method cannot distinguish between blue and green water consumption or whether the source is surface water or groundwater.

3.4. CROPWAT Model and Calculation

Using the irrigation scheduling method, we calculated and analyzed the water footprint of China’s staple grains (maize, rice, and wheat) with the CROPWAT model. Developed by the Land Rainwater Development Division of the Food and Agriculture Organization of the United Nations, the CROPWAT model’s main function is to calculate the amount of evaporation of water during crop growth, and it allows users to improve irrigation methods and plans based on the simulation results as well as to study the impact of precipitation, artificial irrigation and other non-irrigation conditions on crop growth. Using the CROPWAT model to calculate crop evaporation requires meteorological parameters including temperature, humidity, wind speed, sunshine time, etc., as well as soil parameters, and crop parameters as reference variables, all of which are combined with a study of the longitude and latitude of the cropland. Then the CROPWAT model calculates the results using the embedded Penman–Monteith formula. The meteorological parameters we used were derived from the National Greenhouse Data System and the CLIMWAT Meteorological Database [30,31], and the water footprint of the crop was calculated in combination with the data provided in the China Statistical Yearbook and the China Agricultural Yearbook (e.g., crop yield, planting area, etc.) [32,33].

The calculation method of the crop’s blue and green water footprints was as follows:

$$W{F}_{Blue}=10\times \frac{E{T}_{Blue}}{Y}$$

(1)

$$W{F}_{Green}=10\times \frac{E{T}_{Green}}{Y}$$

(2)

where $E{T}_{Blue}$ is the blue water portion of the evapotranspiration during the growth period of the crop, $E{T}_{Green}$ is the green water portion, and they are expressed in millimeters. To calculate the water footprint of production and other indicators, the resultant units of $E{T}_{Blue}$ and $E{T}_{Green}$ are unified as cubic meters per square hectometer (${\mathrm{m}}^3/{\mathrm{hm}}^2$); “10” in the equation is the conversion factor of millimeters to cubic meters per square hectometer. Y stands for the yield per unit area of the crop, the unit for which is kilograms per square hectometer ($\mathrm{kg}/{\mathrm{hm}}^2$). The calculation of $E{T}_{Blue}$ and $E{T}_{Green}$ is as follows:

$$E{T}_{Blue}=\max \left(0,E{T}_c-P_{eff}\right)$$

(3)

$$E{T}_{Green}=\min \left(E{T}_c,P_{eff}\right)$$

(4)

where $E{T}_c$ is the evaporation during crop growth and $P_{eff}$ stands for the effective precipitation. These are expressed in millimeters. $P_{eff}$ is calculated by the effective precipitation calculation formula proposed by the FAO Soil Conservation Division in CROPWAT. $E{T}_c$ is derived from the following calculation method:

$$E{T}_c={\displaystyle \sum }_{d=1}^n\left(K_c\times E{T}_0\right)$$

(5)

where $K_c$ is the crop coefficient and $E{T}_0$ stands for the reference crop evaporation, expressed in millimeters per day (mm/d). The calculation method of $E{T}_0$ is as follows:

$$E{T}_0=\frac{0.408\Delta \left(R_n-G\right)+900\gamma /\left(T+273\right)\times U_2\left(e_s-e_a\right)}{\Delta +\gamma \left(1+0.34U_2\right)}$$

(6)

The formula is the Penman–Monteith formula embedded in the CROPWAT model, which is the FAO-recommended formula for calculating the amount of evaporation in the evapotranspiration of the reference crops [34], which has also been widely used in research around the world [35,36,37]. $\Delta$ stands for the slope of the saturated water pressure and temperature curve, and its unit is kilopascals per degree Celsius (kPa/°C). $R_n$ represents the referenced net radiation on the crop canopy surface, the unit for which is megajoules per square meter per day ($\mathrm{MJ}/\left({\mathrm{m}}^2·\mathrm{d}\right)$). $G$ is soil heat flux and its unit is the same as that of $R_n$. $\gamma$ stands for wet and dry table constants, which shares the same unit as $\Delta$. $T$ is the daily average temperature at a height of 2 m, the unit for which is meters per second ($\mathrm{m}/\mathrm{s}$). $e_s$ stands for saturated water vapor pressure and $e_a$ stands for actual water vapor pressure, and their units are kilopascals ($\mathrm{kPa}$).

By calculating the blue and green water footprints, we can obtain the water footprint of a crop production. Our main research indicators are the water footprints of production $(W{F}_P)$, the blue water footprint ($W{F}_{Blue}$), and the green water footprint ($W{F}_{Green}$), and their units are cubic meters per kilogram (${\mathrm{m}}^3/\mathrm{kg}$). The water footprint of production is calculated as follows:

$$W{F}_P=W{F}_{Blue}+W{F}_{Green}$$

(7)

4. Results

We used the CROPWAT model to calculate the production water as well as the blue and green water footprints of staple grains (e.g., maize, rice, and wheat) in various provinces in China in recent years. The results are detailed in Appendix A with a brief description offered in this section. It is important to note that the maize production in Tibet in 2015 and the wheat production in Jilin in 2018 were much lower than those in other provinces, resulting in abnormal water footprint calculation results; therefore, these two factors were excluded during analysis.

4.1. Crop Water Footprints

Water footprints of crop production indicated the amount of water resources consumed per crop unit during production. According to the water footprint calculation by the irrigation scheduling method, a crop’s green water footprint represented the amount of water resources extracted from the natural environment and used by the crop unit during growth, including the surface and groundwater in the soil environment, and natural precipitation, etc. The blue water footprint represented the irrigation water used per crop unit during production. The higher the water footprint during production, the more water resources were consumed per crop unit during production, and vice versa. The higher the amount of the green water footprint, the more natural water resources were used in the production per crop unit, and inversely, the less natural water resources were consumed. The larger the blue water footprint, the more irrigation water resources were used in the production per crop unit, and inversely, the less irrigation water resources were consumed.

In terms of the water footprint during maize production in 2015–2019, disregarding whether it was the maximum, the minimum, or the overall level, it was comparatively smaller, as compared to the data from 2000, indicating that maize per harvested unit consumed less water in recent years compared with the past. According to the relationship between the overall water footprint during production and the green water footprint during production, we suggest that the total amount of water resources needed to harvest maize per area unit in Fujian, Zhejiang, and other regions is larger than the amount of water needed to harvest the same amount in Qinghai, Xinjiang, and Ningxia. Fujian, Zhejiang, and other areas had very high green water footprints. In terms of blue water footprints, Fujian was higher, while Zhejiang was lower, which showed that although maize planting in Zhejiang consumes more water resources, most of the consumption was natural water resources, instead of irrigation water resources. Fujian consumed a lot of both natural water and irrigation water resources. Such results may have been caused by soil structure, climate, temperature, precipitation, and other natural factors in different provinces [38,39]. The blue water footprint in Xinjiang, Shaanxi, and other regions was larger than those of Yunnan, Sichuan, and other regions, indicating that more irrigation water resources are needed to harvest maize per unit area. In addition, as compared to 2000, the footprint of green water has decreased in recent years, indicating that China’s maize production per unit area consumed less natural water resources. The reason for this may have been the soil moisture loss as a result of the increased greenhouse effect which led to higher temperatures, among other reasons [40]. The maximum and minimum blue water footprints over the last five years were both less than those reported in 2000, indicating that the irrigation water needed to harvest maize per unit area was less than the consumption reported in 2000. This may have been due to the water-saving advancements in irrigation technology over the last ten years [41], or it may have been due to the improvements in maize seeds that have reduced water demand during crop growth, enabling natural water resources to meet crop growth needs to a greater extent and, thus, reducing the use of irrigation water [42] (

Table 1. Water footprints of maize.

Statistical Measure	2000	2015	2016	2017	2018	2019
Crop Water Footprint (m3/kg)
Mean	0.887	0.765	0.717	0.791	0.785	0.824
Minimum	0.535 (Qinghai)	0.452 (Sichuan)	0.485 (Jilin)	0.525 (Qinghai)	0.499 (Sichuan)	0.498 (Sichuan)
Maximum	1.614 (Fujian)	2.377 (Guangdong)	0.964 (Fujian)	1.344 (Fujian)	1.121 (Fujian)	1.273 (Jiangxi)
Green Water Footprint (m3/kg)
Mean	0.479	0.422	0.425	0.462	0.436	0.367
Minimum	0.133 (Ningxia)	0.158 (Gansu)	0.147 (Ningxia)	0.079 (Xinjiang)	0.170 (Xinjiang)	0.041 (Ningxia)
Maximum	0.821 (Guangxi)	0.763 (Fujian)	0.923 (Zhejiang)	0.787 (Guangdong)	0.853 (Zhejiang)	0.685 (Zhejiang)
Blue Water Footprint (m3/kg)
Mean	0.407	0.343	0.292	0.329	0.349	0.457
Minimum	0.088 (Hunan)	0.059 (Chongqing)	0.009 (Yunnan)	0.014 (Yunnan)	0.086 (Sichuan)	0.001 (Sichuan)
Maximum	0.950 (Fujian)	1.755 (Inner Mongolia)	0.574 (Shaanxi)	0.772 (Fujian)	0.669 (Shaanxi)	1.024 (Jiangxi)

).

The water footprint of rice was similar to that of maize. The maximum, minimum, and overall levels of the three water footprints in 2015–2019 were comparatively smaller than those in 2000, which also indicated that the rice harvested per unit of yield in recent years consumed less water, including natural and irrigation water resources, than in the past. In addition to the natural factors similar to those for maize, the above results may have also been due to improved rice seeds. At the provincial and national rice cultivation levels, Hainan, Tibet, and Inner Mongolia had larger water footprints during rice production while Sichuan, Hubei, and Liaoning had smaller ones, indicating that the former consumed more water than the latter when harvesting rice per unit of yield. Guangdong, Guangxi, Hainan, and other places had larger green water footprints than Ningxia and Xinjiang, indicating that Guangdong, Guangxi, and Hunan used far more natural water resources than Ningxia, Xinjiang in the cultivation of rice. Provinces such as Inner Mongolia, Tibet, and Xinjiang have high blue water footprints, while Chongqing, Sichuan, and other regions had lower footprints, indicating that growing rice in Inner Mongolia and other regions consumed more irrigation water resources (

Table 2. Water footprints of rice.

Statistical Measure	2000	2015	2016	2017	2018	2019
Crop Water Footprint (m3/kg)
Mean	1.436	1.258	1.253	1.324	1.274	1.299
Minimum	0.977 (Gansu)	0.849 (Sichuan)	0.878 (Sichuan)	0.880 (Sichuan)	0.862 (Sichuan)	0.796 (Sichuan)
Maximum	2.033 (Hainan)	1.949 (Tibet)	1.794 (Hainan)	1.857 (Hainan)	1.671 (Tibet)	1.734 (Hainan)
Green Water Footprint (m3/kg)
Mean	0.574	0.509	0.545	0.555	0.531	0.475
Minimum	0.117 (Ningxia)	0.146 (Xinjiang)	0.179 (Xinjiang)	0.095 (Xinjiang)	0.149 (Xinjiang)	0.076 (Xinjiang)
Maximum	1.393 (Hainan)	1.157 (Hainan)	1.101 (Hainan)	1.254 (Hainan)	1.066 (Hainan)	1.142 (Hainan)
Blue Water Footprint (m3/kg)
Mean	0.862	0.749	0.709	0.770	0.742	0.823
Minimum	0.480 (Sichuan)	0.379 (Chongqing)	0.397 (Sichuan)	0.294 (Guangdong)	0.288 (Sichuan)	0.240 (Sichuan)
Maximum	1.395 (Jilin)	1.522 (Tibet)	1.207 (Tibet)	1.353 (Inner Mongolia)	1.147 (Hebei)	1.275 (Anhui)

).

Wheat production had a slightly different outcome from that of maize and rice production. In 2015–2019, the provinces with the largest water footprint during wheat production were Yunnan and Guangxi, and the smallest were in the provinces of Liaoning and Qinghai, which were very different from the regional growth of maize and rice. Regarding green water footprints, Guangxi, Jiangxi, and Yunnan have used more natural water resources when growing wheat, as compared to Guangxi and Jiangxi, which also used more irrigation water resources. The provinces with smaller water footprints during wheat production were Qinghai, Liaoning, and Xinjiang, and their green water footprints were very small. However, the blue water footprint of Xinjiang was in the upper-middle level. Qinghai and Liaoning were at the end. These patterns showed that Xinjiang’s small water footprint during production did not necessarily indicate that the water use efficiency of these provinces were high. These provinces may not have been able to provide sufficient natural water resources due to natural conditions, and therefore, they needed to use irrigation for successful crop production. For ideal water efficiency, planting advantages may exist in these provinces. However, in terms of saving irrigation water, Xinjiang was not as efficient as Qinghai and Liaoning. Another difference was that the largest water footprint during wheat production increased in 2015–2019, as compared to 2000, and the minimum decreased; the results for green and blue water footprints were similar. However, at the overall level with average production, the blue and green water footprints for each year from 2015 to 2019 was still lower than those in 2000, which indicated that the water needed to plant wheat was still less than it had been more than a decade ago, which was consistent with the results for maize and rice production (

Table 3. Water footprints of wheat.

Statistical Measure	2000	2015	2016	2017	2018	2019
Crop Water Footprint (m3/kg)
Mean	2.052	1.637	1.763	2.099	2.455	2.051
Minimum	0.769 (Qinghai)	0.771 (Qinghai)	0.782 (Qinghai)	0.827 (Qinghai)	0.656 (Liaoning)	0.736 (Liaoning)
Maximum	4.506 (Guangxi)	3.599 (Guangxi)	4.300 (Guangxi)	5.264 (Guangxi)	5.399 (Guangxi)	5.194 (Guangxi)
Green Water Footprint (m3/kg)
Mean	1.105	0.936	0.963	1.063	1.349	0.998
Minimum	0.289 (Tibet)	0.277 (Tibet)	0.257 (Tibet)	0.307 (Tibet)	0.332 (Tibet)	0.207 (Ningxia)
Maximum	2.753 (Jiangxi)	2.774 (Guangxi)	2.559 (Guangxi)	3.621 (Guangxi)	3.294 (Jiangxi)	3.298 (Guangxi)
Blue Water Footprint (m3/kg)
Mean	0.947	0.701	0.800	1.035	1.107	1.052
Minimum	0.225 (Jiangsu)	0.177 (Zhejiang)	0.288 (Jilin)	0.254 (Qinghai)	0.197 (Qinghai)	0.320 (Qinghai)
Maximum	2.301 (Guangxi)	1.774 (Ningxia)	2.381 (Ningxia)	2.629 (Yunnan)	2.555 (Yunnan)	3.485 (Yunnan)

).

According to the regional division of China’s economic level proposed by The National Development and Reform Commission of China, we divided China into three parts. The western region included 12 provinces: Inner Mongolia, Guangxi, Chongqing, Sichuan, Guizhou, Yunnan, Tibet, Shaanxi, Gansu, Ningxia, Qinghai, and Xinjiang. The central region included eight provinces: Shanxi, Jilin, Heilongjiang, Anhui, Jiangxi, Henan, Hubei, and Hunan. Eleven provinces defined the eastern region: Beijing, Tianjin, Hebei, Liaoning, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong, and Hainan. Using the water footprints in 2019 as the study object, the calculation results of the water footprint of each province were weighted and averaged according to the regional output to obtain the water footprint of the staple grain crops in the three regions of China. The results are shown in the figure below.

As can be seen in Figure 1, the advantages of water resource utilization varied from region to region, with maize consuming the least amount of water in the western region, and rice and wheat sharing the same in the eastern region, which was consistent with the results of Yu and Liu’s studies [43,44]. However, in maize cultivation, the western region not only had the smallest water footprints during production, but it also had the smallest blue water footprints and relatively small green water footprints, which indicated that natural water resources provided a relatively large amount of the water for maize cultivation and, therefore, consumed a relatively small amount of irrigation water resources. In terms of rice production, the eastern region had a unique advantage compared with the other two regions in terms of the use of natural water resources versus irrigation water resources; that is, the total amount of water consumption was the lowest, and the use of natural water resources was the highest, with irrigation water being consumed least. In terms of wheat production, the water footprint of the eastern region was the smallest, indicating that the wheat harvested per unit area consumed the least amount of water resources. However, the irrigation water resources consumed in the eastern region did not have a significant advantage, as compared to the central region, as they were similar. The difference in the use efficiency of natural water resources led to a lower water consumption in the eastern region, as compared to the central region.

4.2. North-to-South Grain Transportation

Currently, the grain transportation in China has reversed from South-to-North to North-to-South. The three northeastern provinces (Heilongjiang, Jilin and Liaoning) have become China’s largest grain production and grain export regions, and the southeastern coastal provinces have become China’s largest grain marketing areas and export destinations. Interregional cooperation of supply and demand has gradually transitioned to supply-chain cooperation while the northern and southern regions have their advantages in the staple grain producing and marketing arenas, respectively. They have defined economic interests as the anchor point and market demand as the guidance for their divisions of labor and cooperation for common development. The northern regions which have been transporting staple grains from the South to the North are Beijing, Tianjin, Hebei, Shanxi, Inner Mongolia, Liaoning, Jilin, Heilongjiang, Shandong, Henan, Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang; while the export destinations in the South are Shanghai, Jiangsu, Zhejiang, Anhui, Fujian, Jiangxi, Hubei, Hunan, Guangdong, Hainan, Guangxi, Sichuan, Guizhou, Chongqing, Yunnan, and Tibet. In this article, the water footprint calculations of each province were weighted and averaged by regional output.

The water consumption for production in the northern region was less than in the southern region. In addition, maize and wheat had more significant advantages. The division of labor and the cooperation system that enables grain planting in the northern region and grain sales in the southern region has been organized for efficiency based on economic development and water conservation. However, regarding the use of irrigation water, growing wheat and rice in the northern region consumed more irrigation water as compared to the southern region (e.g., the northern region consumes 0.17 m³ more irrigation water per kilogram of rice than the southern region, while it consumes 0.26 m³ more irrigation water per kilogram of wheat, and the production of maize can only save 0.06 m³ irrigation water). The conclusion of saving water resources in the northern region was due to the use of fewer natural water resources and more irrigation water resources. Moreover, in terms of saving irrigation water resources, the current North-to-South Grain Transportation is not the best way to allocate crops in order to save water (Figure 2).

4.3. Crop Production Areas

The establishment of crop production areas was China’s agricultural production policy intended to solve the contradiction between industrialization, urbanization development, and agricultural production land as well as to further consolidate the basis of agricultural production, which would ensure grain production by designating the crop production areas and by focusing on the production of certain types of agricultural products so as to maintain sufficient levels of important agricultural products. At present, the food production areas have been designated for the production of maize, rice, and wheat. The maize production areas include Songnen Plain, Sanjiang Plain, Liaohe Plain, and HuangHuaiHai as well as the Wei River basins (Heilongjiang, Jilin, Liaoning, Shaanxi, Shanxi, Gansu, Beijing, Tianjin, Shandong, Hebei, Henan, Jiangsu, and Anhui). The rice production areas include the Northeast Plain, Yangtze River basin, and southeast coastal areas (Heilongjiang, Jilin, Liaoning, Yunnan, Guizhou, Sichuan, Gansu, Shaanxi, Henan, Hubei, Hunan, Jiangxi, Anhui, Jiangsu, Zhejiang, Guangdong, Fujian, Tibet, Guangxi, Chongqing, and Shanghai). The wheat production areas include the HuangHuaiHai region, the middle and lower reaches of the Yangtze River, and the northwest and southwest regions that have been deemed most advantageous (Beijing, Tianjin, Shandong, Hebei, Henan, Jiangsu, Anhui, Hubei, Hunan, Jiangxi, Zhejiang, Shanghai, Xinjiang, Qinghai, Tibet, and Inner Mongolia). Based on the data reported in 2019, the water footprint calculation results of each province were weighted and averaged by regional output to obtain the water footprint of staple grains in China’s crop production areas, as shown in the figure below.

In comparison, we found that the water footprints of staple grain production areas were smaller than that of non-designated areas. Therefore, China’s agricultural policy could not only be able to ensure food security, but also conserve water resources while producing staple grains. In terms of the blue water footprints, staple grain productions in designated areas had smaller blue water footprints than that in non-designated areas, so a harvest of the same amount of grain in the designated area consumed less irrigation water and effectively saved irrigation water resources. The green water footprints of maize and wheat harvested in designated areas were smaller than those that were harvested in non-designated areas. while that of rice harvested in designated areas was larger than rice harvested in non-designated areas, which indicated that maize and wheat use relatively fewer natural water resources in designated areas, while rice cultivation used relatively larger amounts of natural water resources. This result did not conflict with the conclusion that the establishment of the crop production areas could conserve water resources (Figure 3).

5. Discussion

The purpose of water resource management analysis was to improve the efficiency of water use and promote sustainable development. In response to regional population growth, climate change, water shortage, and other global challenges, agricultural production as the main and indispensable channel of water resource use has been the focus of much research. In traditional agronomy, farmers usually employ improved irrigation, engineering, and seed technologies to upgrade and improve water resource conservation. The analysis of the water footprints did not directly contribute to improving water resource conservation-like techniques, but it may provide a new perspective for water resources management analysis. Considering direct and indirect usage of water resources, it sheds light on the water resource requirements in crop production, and should assist, in future research, with water resource conservation and the consolidation of sustainable development.

We used an irrigation analysis method to analyze the water footprints, where the water footprints of crop production depicted the total amount of water resources consumed by crops from seeding to harvesting per unit area yield. The green water footprint depicted the utilization efficiency of natural water resources and the blue water footprint depicted the utilization efficiency of irrigation water resources. This research will hopefully provoke thought and consideration to improve sustainable development initiatives and water resource conservation.

This article calculated the water footprints, as well as the green and blue variations of crop production for China’s stable grains (maize, wheat, and rice) in 2000 as compared to 2015–2019, and we discussed the efficiency of water resource utilization at the provincial and regional levels. With advancements and developments in technology and science in recent years, harvesting the same amount of crops consumed less water than it had 20 years ago. Maize production saved about 12.4% of water resources, rice cultivation saved about 10.8%, and wheat production saved about 2.5%. The provinces that consumed the most water in maize cultivation were Fujian and Zhejiang; however, Zhejiang had less irrigation water consumption. Qinghai and Sichuan had the least water consumption, with Chongqing and Sichuan consuming the least irrigation water. Tibet and Hainan consumed the most water due to rice cultivation, and Tibet also had the highest irrigation water consumption. Sichuan and Liaoning had the least water consumption, with Sichuan’s irrigation water use ranking at the bottom of all provinces. The provinces that consumed the most water from wheat cultivation were Yunnan and Guangxi, and the least were Liaoning and Qinghai. The provinces with the most irrigation water consumption were Yunnan and Ningxia, and the least were Qinghai and Chongqing. At the regional level, the total amount of water consumed by maize cultivation, as well as irrigation water resources consumed, were the lowest in the western region, and the total amount of water resources consumed by rice cultivation, as well as irrigation water resources consumed, were lowest in the eastern region. Wheat cultivation in the eastern region consumed less water resources, while less irrigation water was needed in the central region. The results were consistent with the results of Yu and Liu’s studies [43,44].

North-to-South Grain Transportation has advantages in terms of total water consumption in grain production, but there is still room for optimization in terms of irrigation water conservation. In order to save irrigation water to a greater extent, the advantages of regional production should be further considered when grain is transported, and different kinds of crops should be optimized. For example, producing more wheat in areas where wheat cultivation consumes less irrigation water (e.g., Qinghai, Jilin, Heilongjiang, etc.), increasing rice cultivation in Henan and other places where it is most beneficial, and increasing maize cultivation in Shaanxi, could further clarify the division of labor among different types of staple grain cultivation as well as encourage the provinces with less irrigation water usage to produce more grain and participate in the crop allocation in an effort to achieve the goal of water resource conservations. The designation of three staple grain production areas used for the production of specific crops will not only ensure China’s food security and ensure the sufficient supply of staple grains, but also has the potential to conserve water resources during production, reduce irrigation water, and promote sustainable use of irrigation water resources at the same time.

For sustainable development, we should consider conserving water resources during the entire production process. Moreover, in view of the shortage of available water resources, we should also consider saving irrigation water resources as much as possible and make more use of natural water resources. Reconfiguring the crop production policies could be a good way to increase the use efficiency of irrigation water. The research results in this article provide suggestions for grain production quotas according to the benefits available in each province or region. We suggest growing less grain in provinces that consume more water, and multiple grains in provinces that consume less water. For example, it would be appropriate to plant a variety of wheat and corn, instead of rice, in Heilongjiang, and plant less corn and wheat and a variety of rice in Zhejiang to achieve the purpose of conserving water resources and improving water resource utilization efficiency. We also suggest growing fewer grains in provinces that consume more irrigation water, and multiple grains in provinces that consume less irrigation water. For example, in Guangdong, there should be less wheat and more rice and corn production, and Xinjiang should produce less rice and more wheat.

This analysis mode may be one-sided, however. Assuming that the regional demand remains unchanged, if the production supply decreases, it will be necessary to import grain from another province, which may incur purchase and transportation costs. Although reconfiguring the crop production policies could conserve water resources to a certain extent, if higher costs (such as transportation and grain storage costs, etc.) are generated, then other resources may be wasted. If interprovincial trade is added to the analysis, the results of the argument may be more accurate, which could be a direction for future research.

The climate data used in this study were obtained from meteorological observation stations in different provinces, rather than being measured through field experiments, which is another limitation of this study. However, the climate conditions of the planting areas can be replicated to a greater extent by using data from meteorological observation stations closer to the crop planting areas. Although the water footprint calculation results were inaccurate, we thought the calculation of water footprints were still convincible. Field climate observation data of the planting areas can be used to make the calculation results more accurate and the analysis more effective, which is the direction of our future research.

6. Conclusions

The use and evaporation path of agricultural water are always regarded as vital for sustainability aspects, and at the same time, water use efficiency is considered essential. As the largest grain producer, China has enacted policies to consolidate the basis of agricultural production and has maintained self-sufficiency in food production. Facts and studies have proved that the policy effects are significant, but production efficiency was not in an optimal state [18,19,20].

The current article evaluated the efficiency of China’s agricultural water usage and analyzed the effectiveness of two agricultural policies. The results showed that compared to 2000, water use efficiency of crop production for maize, rice and wheat during 2015–2019 were increased by about 12.4%, 10.8%, and 2.5% respectively. North-to-South Grain Transportation and food production areas have advantages in terms of reducing total water consumption in grain production, but reconfiguring the crop production policies could be a good way to increase the use efficiency of irrigation water. Large numbers of crops are planted in provinces that consume more water (both green water and blue water) under the current policies; for example, planting rice in Heilongjiang and corn in Zhejiang. The article provided perspectives to reconfigure the crop production policies, such as growing less grain in provinces that consume more water, and more grain in provinces that consume less water. Furthermore, interprovincial trade can be considered in the analysis to estimate the efficiency of water saving and food supply, which relate economic efficiency to sustainability.