Quantification of water consumption is generally limited to what can be described as direct or apparent consumption, and represents water collected and extracted by the relevant country in order to satisfy various needs, generally shared between consumption for domestic and urban use, agriculture, industry and energy production. This approach, however, does not facilitate understanding or evaluation of the impact of a country on water resources. Indeed, it is likely to import goods and services that will be consumed there by its inhabitants, and export other goods and services that will be consumed by the inhabitants of other countries. A certain amount of water is linked to these goods and services that was consumed during their production. A small amount of this water may have been included in the final product, but the majority has generally evaporated or been disposed of, changed in quality, in the producing country.
A country therefore consumes the water of each of the countries whose goods and services it imports, and each of the countries that it exports to consume its water. This is referred to as the transfer (or flow) of virtual water and is associated with the notion of a ‘water footprint’, defined by Hoekstra et Hung. (1). The term virtual water relates to the indirect flow of water needed for the production of imported or exported goods by a country. Its water footprint can be estimated by taking into account the amount of water collected and extracted nationally, added to the amount of virtual water entering the country, then subtracting the virtual water flow leaving the country.
Globally, the transfer of virtual water linked to the trade of goods and services has risen to 1 625 km3/year (270m3/person/year), while the total water footprint is estimated to be 7 450 km3/year (1 240m3/person/year). The difference from one country to another is significant: the average imprint of the United States is 2 480m3/person/year of which 470m3/person/year is imported virtual water; compared to 980m3/person/year in India, of which 16m3/person/year is imported virtual water. (2)
Currently, the main transfers of virtual water take place from the United States to Central America, Western Europe, North Africa and Central and South Asia; from South America to Western Europe, and from the South West Pacific to Central and South Asia. For the most part, they relate to agricultural products which form 695 km3/year: 43% of virtual water flow. These flows have a tendency to increase linked to the globalisation of trade, changing ways of living, and in particular eating habits: on average, 14 000m3 of water is needed to produce one tonne of beef, while the production of one tonne of rice consumes around 1 400m3, and one tonne of wheat 1 200m3.
There are many issues underlying virtual water analysis. For one, it is thought that the export of virtual water from water-rich countries to those with a water deficit is a way of restoring equilibrium of the availability of this resource. Moreover, the trade of virtual water between countries results in water savings where exporting countries have the advantage of more favourable climates for agricultural production than the importing countries. These global water savings are estimated to be 352 km3/year (3). However, this results not only in dependence of certain countries on others, but also to the long distance transportation of products, and thus a significant energy impact and repercussions relating to greenhouse gases.
Furthermore, the logic behind global trade is not necessarily that relating to the best management of this resource. In other words, virtual water in some cases involves countries importing water resources from producer countries, who are water-poor themselves, depriving their own inhabitants of water. This raises issues of inter-generational solidarity that this trade jeopardises. In an urban sphere, the growing significance of virtual water flows is highlighted in the constantly growing distances water must travel, and shown in the breaking links between urban and rural areas, that can barely supply enough water for them anymore.
Virtual water essentially takes into account the quantities of water involved in international trade. A qualitative analysis (Chatzimpiros et Barles, 2008) that evaluates the degradation of this resource linked to this trade shows that the issues linked to virtual water are considerable, and certain countries exhaust their water resources, both in quantity and quality, to the profit of others.