Providing a comprehensive insight, water footprint (WF) is widely used to analyze and address water-use issues. In this study, a hybrid of bottom-up and top-down methods is applied to calculate, from production and consumption perspectives, the WF for Xiamen city from 2001 to 2012. Results show that the average production WF of Xiamen was 881.75 Mm3/year and remained relatively stable during the study period, while the consumption WF of Xiamen increased from 979.56 Mm3/year to 1,664.97 Mm3/year over the study period. Xiamen thus became a net importer of virtual water since 2001. Livestock was the largest contributor to the total WF from both production and consumption perspectives; it was followed by crops, industry, household use, and commerce. The efficiency of the production WF has increased in Xiamen, and its per capita consumption WF was relatively low. The city faces continuing growth in its consumption WF, so more attention should be paid to improving local irrigation, reducing food waste, and importing water-intensive agricultural products.
Fresh water is a scarce resource globally, especially in growing urban areas because of their high consumption demand, vigorous economic development, and population expansion (Liu & Yang 2012). China faces a severe water scarcity problem, as its per capita water resources are a quarter of the world's average and two-thirds of its 669 cities have water shortages (Liu & Yang 2012). China is also experiencing an historically rapid urbanization process (Wang et al. 2015). Obtaining a comprehensive understanding of the water challenge faced by Chinese cities is therefore important.
Water footprint (WF) represents an opportunity for better water management and use; it measures the volume of water needed to produce the goods and services consumed by an individual person, business, region, or country (Hoekstra et al. 2012). The concept of WF was developed from virtual water (VW), which is defined as actual water used in the production of a commodity or service (Velázquez et al. 2010; Hoekstra et al. 2012). WF is usually used to address the total water consumption of a producer or a consumer, while VW is frequently used when talking about a product or service; VW is also widely used in the context of international or interregional trade. The WF of different areas can be linked by virtual water trade or flow. In addition, WF can be categorized into blue, green, and grey WFs (Chenoweth et al. 2013). The blue WF refers to surface and groundwater consumed during production, which is equal to direct water use in industries and households minus return flows. The green WF is defined as the use of soil moisture, also known as effective or productive precipitation. The grey WF is the volume of water used to assimilate the load of pollutants so that acceptable water quality standards are met. There have been many studies on product (Chapagain & Hoekstra 2003; Gerbens-Leenes et al. 2013), regional, national, and global WFs (Zang et al. 2012; Ercin et al. 2013; Ercin & Hoekstra 2014). The method presented by Hoekstra is developed into the Water Footprint Assessment (WFA) approach (Hoekstra et al. 2012). Besides, another similar approach, namely ISO 14046, was also developed by the Life Cycle Assessment (LCA) community to address environmental impacts (Boulay et al. 2013; Kounina et al. 2013). Rather than LCA, the WFA approach was more widely used in studying water-related hot-spots and was found useful for guiding actions for improving water efficiency and management (Jefferies et al. 2012; Manzardo et al. 2016).
For a certain geographic area, the production water footprint (WFprod) is defined as the sum of direct and indirect water use in local production processes, while the consumption water footprint (WFcons) is the water resources used to meet the consumption of the goods and services in a given area (Ercin et al. 2013). WFcons and WFprod can reach a balance with net imported virtual water (VWni). The WF of a certain area can be calculated by two methods within the WFA method, top-down and bottom-up (Hoekstra et al. 2012). The bottom-up method is the adding up of the direct and indirect water use of the goods and services consumed or produced in a region. In the top-down method, WFcons is equal to WFprod plus imported virtual water minus exported virtual water. Compared to the top-down method, the bottom-up method is applied more widely because of better availability of data (Hoekstra et al. 2012). In addition, an input–output method has recently been widely used to analyze the water interconnections and interdependences of economic units (Wang et al. 2013).
Xiamen city has been facing severe water shortages. Its per capita water resource is only 333.90 m³, which represents only about one-third of the international minimum standard (1,000 m³ per capita) (UNESCO 2016). To address the water problem, one must understand the structure and efficiency of water use and the interdependency of water use among cities. The concept of WF from production and consumption perspectives, using a hybrid bottom-up/top-down method, was therefore applied to analyze Xiamen's water resource needs from 2001 to 2012. This paper is organized as follows: (1) information concerning the study area and an introduction to research scope, methodology, and data, (2) analysis of the WFs and virtual water trade, and (3) conclusion.
MATERIALS AND METHODS
Xiamen is a coastal city in southeast China (118 °04′04″ E and 24 °26′46″ N). The gross domestic product (GDP) of Xiamen reached 281.52 billion CNY in 2012, up from 55.83 billion CNY in 2001. The secondary and tertiary industries are the main economic sectors and represented 48.4% and 50.7% of industrial activity, respectively in 2012. The urbanization rate was 80.9% in 2012. The weather of Xiamen is dominated by a monsoonal humid subtropical climate with abundant rainfall. The total precipitation amounts to 1,144 mm, most of which falls between April and August. Water resources have been under critical pressure due to a lack of water availability, growing population, and economic development (Tang et al. 2013). During the period 2001 to 2012, the population expanded from 2.19 million to 3.67 million. Accordingly, the average per capita water resource decreased from 669.86 m³ in 2001 to 333.90 m³ per capita in 2012. The recent figure represents only 41% of the national level and is much lower than the internationally recognized minimum standard of 1,000 m³ per capita. The city was therefore listed as a ‘water-poor’ city by the Chinese National Ministry of Water Resources.
Methodology and data
WF of crops
The virtual water content for crop consumption (VWC’crops) is the weighted average of local and imported virtual water content of crops according to local crops production and imported crops.
The calculation of this sector contained main crops, including grains, oil crops, vegetables, fruits, and tea. The data for the production and consumption of the crops, climate, crop types, cultivation times, and soil conditions were taken from the Yearbook of Xiamen Special Economic Zone (2002–2013) (Xiamen Bureau of Statistics 2002-2013).
WF of livestock
The animal products quantified in this paper include pork, beef, lamb, eggs, and dairy products. The data for the production and consumption of livestock products are taken from the Yearbook of Xiamen Special Economic Zone (2002–2013) (Xiamen Bureau of Statistics 2002-2013).
WF of industry
The data for the WF of industrial production were obtained from the Xiamen Water Resource Bulletin (2001–2012) (Xiamen Bureau of Water Resources 2002-2013). Virtual water flows were calculated as the virtual water content of industry (VWCindustry, m3/CNY) multiplied by the gross industrial output value (CNY). Virtual water content of imports was equal to industrial water consumption divided by the national gross industrial output value, both obtained from the China Statistical Yearbook (2002–2013) (National Bureau of Statistics of the People's Republic of China 2002-2013). The virtual water content of exports was calculated as a weighted average of local and national virtual industrial water content.
WF of household use and commerce
The WF of household use and commerce (WFhousehold and WFcommerce) were considered as equal to water consumption of household use and commerce (Ercin et al. 2013; Vanham & Bidoglio 2014), for which WFhousehold,prod=WFhousehold,cons and WFcommerce,prod=WFcommerce,cons. WFhousehold included water supply for permanent and non-permanent residents (e.g. tourists). WFcommerce included water consumed by hotels, commercial enterprises, restaurants, etc. These data were both taken from Xiamen Water Resources Bulletin (2001–2012) (Xiamen Bureau of Water Resources 2002-2013).
RESULTS AND DISCUSSION
WF of Xiamen city
WF of agricultural products
Among the agricultural sectors, pork contributed most to the total agricultural WF from both production (35.63% of total WF) and consumption (32.18% of total WF) perspectives (Figures 3 and 4). It was followed by vegetable (8.95% of total) and grain (5.52% of total) in terms of production, and by grain (18.88% of total) and chicken meat (6.59% of total) in terms of consumption.
Water efficiency in Xiamen city
Per capita consumption WF was also calculated to evaluate citizen consumption levels from 2001 to 2012. This coefficient shows a relatively stable trend during the study period, ranging from 440.28 m3/capita/year to 539.89 m3/capita/year. The value of per capita WF in 2007 (495.25 m3/cap/year) was lower than those of Fujian Province (717.27 m3/cap/year) and China as a whole (648.11 m3/cap/year) (Ge et al. 2011). To keep the calculation boundary coincident with later research, grey WF was included in the calculation. The low water consumption of Xiamen might reflect higher local production water efficiency because its main agricultural products were consumed locally.
Water conservation from trade
As Xiamen city has been short of land for agriculture, importing crops and livestock products to meet local demand could help reduce both water consumption and water pollution from agriculture. Although it may be considered as a strategy for water conservation, trade is always driven by factors other than water. Only when water availability is below a certain threshold can the link between per capita water resource availability and food imports be established (Yang et al. 2003).
To understand the current water situation in Xiamen city, the WFs from production and consumption perspectives were evaluated from 2001 to 2012. The efficiency of the WF and the dependency on outside sources were also presented in this paper. Our main conclusions are as follows.
From a production perspective, the WF of Xiamen city was 881.75 Mm3 on average and showed a relatively stable trend. The consumption WF, however, increased from 979.56 Mm3/year in 2001 to 1,664.97 Mm3/year in 2012. Livestock contributed most to the WF of production and consumption; it was followed by crops, industry, household use, and commerce.
Agriculture was the largest sector in terms of WF, and pork contributed most to the WF of agriculture from production and consumption perspectives. There was a difference in the agriculture consumption WF between urban and rural areas due to different diet structures; urban residents consume more water-intensive products, such as pork and fruit, rather than less water-intensive products such as grain.
On average, the per-GDP WF of agriculture (248.96 m3/103CNY) was much higher than those of industry (2.76 m3/103CNY) and commerce (0.69 m3/103CNY). The latter two decreased by 69.82% and 52.95%, respectively, during the study period, while that of agriculture dropped by only 16.64%. The per capita WF in Xiamen city averaged 482.48 m3/year and remained relatively stable.
In terms of virtual water trade balance, water resources have been conserved by Xiamen city, in which the net imported virtual water increased from 2001. Among the sectors, livestock was the largest contributor to net imported virtual water (338.25 Mm3/year); it was followed by crops (248.03 Mm3/year). Xiamen was a net exporter, however, in terms of industrial virtual water trade (47.59 Mm3/year).
In the future, we expect significant growth in the consumption WF due to population growth, economic development, and changes in resident diet pattern. To solve the water shortage in Xiamen city, the main measures lie in improving the efficiency of water production, decreasing per capita consumption WF, and importing more virtual water through trade. For improving the efficiency of water production, water conservation measures must be adopted in each sector and the industrial structure must be adjusted to become less water intensive. Irrigation improvement in Xiamen has a large potential for water conservation, which would improve the efficiency of water production. To decrease the per capita water consumption footprint, a balanced and healthy diet should be promoted, especially in terms of reducing food waste. Finally, Xiamen should strengthen its trade links to areas with abundant water resources and increase its imports of virtual water; this would reduce stress on local water resources, especially those used for water-intensive agricultural products.
This study was supported by the National Natural Science Foundation of China (71273252 and 71573242).