detailed water market mechanisms may be more appropriate and politically more feasible than a rush to markets (see Bruns et al., 2005).

Encourage water-saving irrigation practices and technology.Farmers in most industrialized countries have only recently begun to adopt water saving practices, whereas in developing countries they have been relying on traditional water saving practices for a long time. Low levels of adoption of water-saving may be because the knowledge and incentives are not in place for farmers to benefit directly by saving water. There is an important role for the private sector in making low-cost agricultural water management technologies such as treadle pumps, small power pumps, and bucket and drip kits more widely available. Such technologies can be readily acquired and used by individual small-scale farmers, both men and women, and in many situations can substantially improve nutrition and incomes (Shah et al., 2000; Namara et al., 2005; Mangisoni, 2006; Merrey et al., 2006). Restrictive policies in some sub-Saharan African countries are retarding the wider use of these technologies, in marked contrast to South Asian countries.

Reform of irrigation management to involve local stakeholders. The establishment of Water User Associations and contracting the management of lateral canals to individuals can improve water management by providing incentives for users and managers to conserve water and improve fee collection to increase irrigation revenues. However, pilot projects to transfer management from the state to user groups on government-built schemes have rarely been scaled up effectively to cover larger areas. Many governments were reluctant, even when project documents promised to do so. Another reason was the failure to recognize the critical differences between government- and farmer-managed irrigation systems. Management transfer programs in countries as diverse as Australia, Colombia, Indonesia Mali, Mexico, New Zealand, Senegal, Sri Lanka, Turkey, and the United States have demonstrated some positive results from involving farmers and reducing government expenditures, but they have rarely shown improvements in output performance or quality of maintenance (Vermillion, 1997; Vermillion and Garcés-Restrepo, 1998; Samad and Vermillion, 1999; Vermillion et al., 2000). The few notable exceptions are middle-income developing countries such as Mexico and Turkey and high income countries such as New Zealand and the United States. Research in the 1990s on irrigation management transfer processes and outcomes produced many case studies and some useful guidelines for implementation (e.g., Vermillion and Sagardoy, 1999). There is broad agreement on the necessary conditions, but very few cases where they have been met on a large scale (Merrey et al., 2007).

      Further "coping" strategies proposed for addressing water scarcity (see also Chapter 6 for more details on the options) need attention at policy levels to incorporate their potential into water management agendas to optimize the use of limited water resources:

1. Desalinization: Currently, the costs of desalinated water remain too high for use in irrigated agriculture, with the exception of intensive horticulture for high-value

cash crops, such as vegetables and flowers (mainly in greenhouses), grown in coastal areas (where safe waste disposal is easier than in inland areas), but recent advances in membrane technology are reducing costs. At the global level the volume of desalinated water produced annually (estimated at 7.5 km3) is currently quite low, representing about 0.2% of the water withdrawn for human use (FAO, 2006b).

2. Urban wastewater:Two features complicate policies pertaining to wastewater use in agriculture: most wastewater is generated outside the agricultural sector, and many individuals and organizations have policy interests pertaining to wastewater use (Qadir et al., 2007). Millions of small-scale farmers in urban and peri-urban areas of developing countries use wastewater for irrigating crops or forest trees or for aquaculture, reducing pressure on other freshwater resources. Surveys across 50 cities in Asia, Africa and Latin America have shown that wastewater irrigation is currently a common reality in three-quarters of cities (IWMI, 2006). Most domestic wastewater generated in developing countries is discharged into the environment without treatment but the dominant trend is for more wastewater treatment as countries develop national integrated water resources management plans or improved environmental policies, for example in Mexico, Brazil, Chile and Costa Rica (UNCSD, 2005). Israel currently uses 84% of its treated sewage effluent in agricultural irrigation and in a few cities, such as Windhoek in Namibia, the water is treated to a very high standard so that it can even be used as drinking water (UNIDO, 2006).

3. Virtual water and food trade: The import of food from water-rich countries allows water-poor countries to save water they would have used to grow food (equivalent to the import of "virtual water"), and scarce water reserves can be used for more valuable domestic, environmental and industrial purposes. Countries with limited water resources might also change their production patterns to prioritize production of agricultural commodities requiring less water and to import those requiring more water (FAO/IFAD, 2006). Whereas the strategy of importing virtual water is appealing from a water perspective, political, social and economic issues, rather than water abundance or scarcity, drive much of the current world food trade.

4. Improving the productivity of water use in agriculture (see Chapter 6 for detailed options). Productivity gains could improve overall water use efficiency in irrigated and rainfed agriculture. Agronomic improvements to improve overall productivity will also reduce the global "water footprint" of agriculture. This could be achieved by, for example, improving the efficiency of fertilizer use; improving soil moisture retention capacity through buildup of organic matter; preventing crop productivity losses due to insects, diseases and weeds; or reducing post-harvest losses due to insects, fungi and bacteria. Each of these is an area for research and technology development, or even for the reintroduction of older management systems, to promote water use efficiency gains which places a high demand on AKST (CA, 2007; Hsiao et al., 2007).