Modernization of irrigation used to be seen as an "engineering project" - replacing open channels with pipes, using sensors to trigger water applications, lining canals and levelling land. Engineers were in charge, too. As one Tanzanian irrigation specialist put it: "We used to design water projects in our office and keep the plans there. We thought the villagers couldn't understand such things."
Today, FAO reports, there is general recognition that "where management is incapable of operating and maintaining a system to high standards, restoring its physical infrastructure alone will not lead to production improvements". Needed are fundamental changes in institutional arrangements and regulations that aim at re-orienting public irrigation institutions towards providing services to farmers and improving their performance in both economic and environmental terms.
This means adopting new technologies and upgrading infrastructure, but also applying sound administrative principles and promoting the participation of water users in a system's operation and management. Farmers' sense of ownership is likely to prevent many problems common to irrigation networks traditionally managed from the "top-down" - field channels demolished, gates stolen and field drainage systems in disrepair.
Conditions for success. Studies of recent irrigation modernization in Argentina, Mexico and Peru have produced a set of "conditions for success". First, modernization is more effective if the initial idea comes from the water users - in the cases studied, farmers had become aware that "business as usual" was no longer possible and pressed for greater efficiency in the irrigation delivery system.
While all the modernization projects included improvements to physical infrastructure, success also depended on training water users in good farming practices, water requirements and irrigation scheduling. Those unaccustomed to water usage charges came to appreciate that water is not free - in the long run, the studies found, "reliance on government subsidies is no longer an option".
In other countries, however, plots of 5 ha. proved unmanageable for some farmers. "The allocation of large plots to full-time, maximum-profit farmers was inconsistent with the reality of people generally pursuing diversified livelihoods," FAO says. Experience in Zimbabwe also showed that small plots are often used more intensively. In the early 1990s, the government increased the size of irrigated plots from 0.1 ha to 3-5 ha, and also favoured giving titles to males. However, productivity per unit of land and productivity per unit of water were later found to be higher on the smaller holdings, and women farmers were significantly more likely than men to adopt irrigation as their main source of food and income.
Devolution. In many developing countries, responsibility for irrigation systems is being devolved from central government to private enterprise and local water user associations. While technical organizations are needed to manage reservoirs and large canal systems, user organizations are almost always capable of managing the village-level, final distribution system (this often allows management organizations to be "re-born" as service providers or service companies).
South Africa's Water Act of 1998, for example, created Catchment Management Agencies with the participation of low-income farmers, while in Turkey, irrigation management has been almost entirely handed over to farmer groups. In Mexico, farmers' associations have assumed management of more than 85% of the country's 3.3 million hectares of publicly irrigated land - while this entailed an increase in charges for irrigation water, they do not exceed 8% of total production costs and most associations are now considered economically independent.
Along with devolution, many governments are phasing out subsidies on irrigation water in order to conserve the resource and promote greater economic efficiency. FAO cautions, however, that the transition phase to modernized management may still require substantial public investment and that effects of water pricing on poor farmers should be carefully assessed. "Pricing policies can be pitched so that farmers neither pay the full cost of their water nor get it free. For example, charges can be based on a traditional price for one-half of the volume normally used, an increased price for the next quarter and a much higher price for the final quarter. Tiered pricing systems of this kind can produce substantial savings."
They can also be used to protect aquifers that are being overpumped (see box below). Once a study has assessed the rate at which an aquifer is naturally replenished, rights to extract that amount can be distributed among the farmers who use the aquifer. Farmers who pump more than their allowance can then be either charged very high prices or forced to buy pumping rights on an open market from others not using their full allowance.
Difficult priorities. Irrigation management must play its part in ensuring that water use throughout a river basin is optimized for benefit of all users, including urban consumers, industry, agriculture, hydro-electric authorities, wetland areas and downstream fishing communities. "During the next few decades, difficult priorities will have to be set," FAO says. "In many parts of the world, urban centres have simply appropriated water in peri-urban areas which they regard as their property, depriving local farmers of their livelihoods. Elsewhere, intensive trading occurs between municipal authorities and rural landowners."
Where water is scarce, these diverse claims to water may need to be be rationalized in logical priorities - the purest water being used for domestic purposes, less pure water (e.g. treated domestic effluent) for irrigation of crops such as cereals, and the poorest quality water for the irrigation of forestry plantations and pasture land.