5. Alleviating the environmental impact of agricultural water development

Dealing with externalities

Most production systems, agriculture included, can cause both positive and negative side-effects, or externalities that are not accounted for in markets. Agriculture’s positive and negative environmental services are unintended consequences of market activities that have an impact on people other than the producer of the effect. These by-products tend not to be priced in the market and, hence, their economic values are unknown or difficult to assess. Consideration of all the positive externalities of agriculture is not readily possible. There are cases where the same phenomenon may be positive in certain circumstances and negative in others, or it may be valued positively by some observers and negatively by others. A positive externality may reduce a negative one, and vice versa. In addition, positive and negative externalities are often linked closely, e.g. soil salinity and improved employment opportunities in irrigated agriculture.

Moreover, positive externalities are often ignored whereas negative ones tend to be reported widely. A well-known example of a negative externality is the loss of biodiversity as a result of draining wetlands for agriculture (FAO, 2002d). Such losses are accelerating as human settlement continues to impinge upon wetlands and forests (Box 10).

Many agricultural systems have become efficient transformers of technologies, non-renewable inputs and finance. They can produce large amounts of food, but have substantial negative impacts on capital assets (Pretty, 1999). These assets comprise not only the natural resources of soil and water per se but also nutrient cycling and fixation, soil formation, biological control, carbon sequestration and pollination. The issue raises concerns about what constitutes success in agricultural production if large yield increases come at the cost of environmental and health problems. One problem is that the benefits and costs accrue to different people and are not measured in the same units. In the 1970s and 1980s, some people considered energy to be such a common measure. Indeed, sustainable systems are much more energy efficient than modern high-input systems. Low-input rainfed rice in Bangladesh and China can produce 1.5-2.6 kg of cereal per megajoule of energy consumed. This is some 15-25 times more efficient than irrigated rice produced in Japan and the United States of America. On average, sustainable systems produce 1.4 kg of cereal per megajoule compared with 0.26 kg/MJ in conventional systems. Modern agricultural systems depend heavily on external inputs, largely derived from fossil fuels. In most industrialized countries, energy is cheaper than labour. Hence, it seems rational to overuse natural resources and underuse labour. The result has been adverse, long-term effects on the environment (Pretty, 1999). Although labour is cheaper than energy in many developing countries, agriculture often has negative effects on the environment. In relation to their policy implications, the environmental externalities of agriculture operate at different geographic scales, e.g. carbon sequestration (a positive externality) on a world scale, but salinization of a watershed on a local scale (a negative externality).

Plate 12 Tuaregs and Bellas preparing soil for planting bourgou (Mali)

FAO/11604/J. VAN ACKER

Applying concepts such as the ‘polluter pays’ principle, cost recovery and cost sharing may prove unrealistic, impractical or politically disastrous to governments in countries where millions of people are poor and small-scale farmers are trying to make a living on marginal lands. A common concern in developing countries is how agricultural production in marginal areas can fulfil its primary function without depleting the natural resource base. For these reasons, developing appropriate technologies, assigning individual or common property rights, and the promotion of alternative employment outside the agricultural sector will be key strategies.

The salinity and drainage question

Much of the environmental impact of irrigated agriculture is linked to the management of water and salt balances of irrigated lands. This includes both minimizing the amount of water required to remove salt from the root-zone, and minimizing the land area required to store the salt temporarily or permanently. Good management has proved difficult. Although human-induced salinity problems can develop swiftly, solutions can be time consuming and expensive. Various improvements in irrigation and agronomic practices can be introduced depending on the type of salinity and on the cause of the accumulation of salts to harmful levels in the rootzone. The fact that saline waters have been used successfully to grow crops shows that under some conditions, e.g. in Mediterranean climates with winter rains, saline water can be used for irrigation. Experience in other locations, where negative long-term effects from irrigating with saline or sodium-rich waters have been observed, indicates that more permanent interventions in the water and salt balance are generally required.

All arid-zone rivers have natural salt profiles, attributable to mobilization of salts in the catchment area and saline seeps. An additional cause of river salinity is irrigation-induced transport of fossil salts owing to pumping from the groundwater into drains that discharge into the river. Figure 3 shows the salinity profiles for four rivers. It illustrates the various degrees to which salts are returned to the river or remain in the land and the groundwater (Smedema, 2000). Increases in the salinity of rivers and streams in many dry parts of the world pose an ecological hazard that has been largely overlooked. The ecological impact of increased salinity in inland waters warrants greater attention in view of the vulnerability of aquatic ecosystems to increased salt levels.

Figure 3 Salinity profiles in four major rivers

Source: Smedema, 2000

Most of the drainage water from agricultural land in Punjab, Pakistan, is reused, either from surface drains or pumped up from shallow groundwater. In fact, in some systems in Punjab, one-half to two-thirds of the irrigation water is pumped from the groundwater. Therefore, the leached salts are returned to the land rather than disposed of in the river system or in evaporation ponds. The average salt influx for the Indus River water is estimated to be about twice the amount that flows out to sea. Hence, half of the annual salt influx remains in the land and the groundwater. Most of the accumulation takes place in Punjab. A more extensive drainage system is needed in order to maintain a sustainable salt balance in the irrigated lands. Worldwide, only 22 percent of irrigated land has a drainage system (less than 1 percent of irrigated land has subsurface drainage). This makes it inevitable that more land will go out of production because of waterlogging and salinity. In general, those people who will lose their land are already very poor farmers.

The drainage situation in Pakistan is in sharp contrast with that in Egypt (Box 11). In Egypt, subsurface drains that take the drainage water back to the river underlie a large portion of the irrigated land. The salts do not stay in the Nile Basin but are discharged into the Mediterranean Sea. During part of the year, the salt content in the lower Indus River is much lower than in the lower Nile River (in the Nile Delta) and more salt disposal into the Indus River could be accepted. However, during critically low-flow periods, such disposals would not be possible. The only option during such periods would be to store the drainage water temporarily for release during high-flood periods. Extending the Left Bank Outfall Drain, now operational in Sindh, into Punjab could provide a more permanent, but quite expensive, solution than the present inadequate number of evaporation ponds.

Wastewater reuse

The reuse of municipal and industrial wastewater in irrigated agriculture is widespread. Some of the wastewater is treated before it is reused. However, much of it is not, and this causes significant environmental and health hazards. In addition, many of the treatment plants in developing countries operate below design capacity, which contributes to the discharge of untreated wastewater into irrigation and drainage canals. Concentrations of heavy metals in canal and drain sediments and in soil samples, as well as faecal coliform bacteria counts in canal and drainage water, often exceed WHO water-quality guidelines. For example, wastewater constitutes 75 percent of the total flow of the Bahr Bagar Drain in the Eastern Delta, Egypt, effectively turning the drain into an open sewer. Soil samples in the Eastern Delta showed cadmium levels of 5 mg/kg, more than twice the natural level. Evidence of uptake of trace elements in crops has also been reported. For example, in the Middle Delta, Egypt, cadmium levels of 1.6 mg/kg (ppm) have been found in rice. Such levels are harmful for human health, and warrant serious attention. Thus, some of the drainage water is unfit for reuse, not because of its high salt content but because of its pollution load. In addition, safe disposal of such polluted wastewater becomes a real problem (Wolff, 2001). Similar cases have been reported for other countries, e.g. Pakistan and Mexico (Chaudhry and Bhutta, 2000).

Plate 13 Dredging irrigation canal (Egypt)

FAO/16222/L. SPAVENTA

The challenge of managing the conjunctive use of groundwater and canal water successfully has been alluded to before. In some areas, over-abstraction of groundwater is evidenced by the rapid dropping of water-table levels. In other areas where the groundwater is too saline for agricultural production, the water table rises as a consequence of over-irrigation and seepage from irrigation canals. Much agricultural land has gone out of production as capillary rise from shallow water tables has ruined soils and poisoned crops. Reversing this process is difficult and expensive (Box 12). In India, the extent of the waterlogged areas is estimated at 6 million ha. In 12 major irrigation projects with a design command area of 11 million ha, 2 million ha are reported to be waterlogged and another 1 million ha salinized (Shah et al., 2000).

It is estimated that salinization alone causes 2-3 million ha/year of potentially productive agricultural land to be taken out of production. How much of this land is reclaimed (to various degrees) and then cultivated again is unknown. Pollution of groundwater by salts and residues of agrochemicals is also a common occurrence. Where slightly saline groundwater is used for irrigation, the repeated cycles of water application to the fields, seepage of the excess water and pumping it up again from the top of the aquifer increases the salt load of the groundwater. If the vertical permeability of the aquifer is restricted, only limited mixing of seepage water takes place and the top of the aquifer from which the water is pumped becomes increasingly saline. This process has been documented for several irrigation systems in Punjab, Pakistan, where conjunctive irrigation with canal water and groundwater takes place (Kijne et al., 1988).

The poorest farmers are those most vulnerable to environmental degradation as most of them farm under difficult growing conditions. A few farmers cultivate the best lands; the vast majority of the farmers cultivate the less fertile and marginal lands. Further degradation is likely to affect the quality of the farmers’ sources of drinking-water and irrigation water, the quality of their land, possibly the quantity and quality of the fish they catch, and ultimately their health. Lack of data on water and salt balances of irrigated land and lack of knowledge on how much water (and of what minimum quality) should be committed to downstream users frustrate attempts to allocate water more equitably to users in order to improve basin-level water productivity in agriculture. The way forward to ending unsustainable practices and reducing the concentrations of salts and agrochemicals that result directly from the degradation of the soil and water resources is a consolidated and long-term effort to improve land and water management.

Generally, agriculture and rural development have not benefited from systematic environmental analysis and management. One reason for their exclusion in the past was probably the very large number of projects (large and small) that could have been referred for an assessment, which would have overwhelmed the environmental assessment agencies. Environmental impact assessment (EIA) is usually applied to physical project planning (e.g. dams, roads, pipelines and industries), but seldom to farm practices and rural development plans. As a result, inadequate planning and inappropriate land-use practices have persisted. In many areas, soil, land and water resources are used inefficiently or are degraded, while poverty and income disparities grow.

With some 30 years of experience, EIA techniques now usually consider not only biophysical impacts but also socio-economic effects on health, human migration in and out of the project area, training of local workforce, local government capacity building, etc. Government and international policies are still needed to establish appropriate legal frameworks and an institutional base for EIA for agricultural projects. These policies should include transfer of the necessary knowledge to the rural poor, e.g. through agricultural extension services, so that they can participate in the environmental assessment of agricultural water resource management and project planning (FAO, 2002d).