Pressures on fresh water
Pressures on land resources
Pressures on the environment
Although this subject has been considered in depth by Alexandratos (1995), some aspects relating to the results of this study will be briefly discussed. The more regional constraints on agricultural production are those imposed by available water and land resources. Some pressures on the environment are of a local nature, others are of global importance. Local environmental effects are, for example, leaching of nutrients from fertilized soils, and acidification caused by re-deposited NH3 from animal wastes. More global environmental effects are caused by, for example, radiatively active gases causing global warming, or compounds (such as NH3) involved in aerosol formation in the atmosphere. In this chapter effects of agrochemical use will not be discussed.
A number of attempts have been made to estimate the irrigation potential in developing countries. For example, FAO (1979) indicated that close to 400 Mha of land is suitable for irrigation on the basis of the land's physiographic position, slope and soil characteristics. Alexandratos (1988) estimated that there is potential for about 300 Mha of irrigation in all developing countries. More recent information is available for Africa. According to preliminary FAO estimates there is still the potential for an additional 2 Mha in North Africa, while the current irrigated area and the projected increases for Sub-Saharan Africa do not exceed the irrigation potential for that region. Preliminary estimates for a few countries in the Near East (Jordan, Lebanon, Syria, Turkey) indicate a potential expansion of the irrigated area of close to 5 Mha. Based on the current climatic conditions and current demand of water from the other sectors in society, both in North Africa and Near East, the agricultural water demand will be testing the limits of the irrigation potential. Based on the estimates of FAO (1984), the irrigated areas in the other regions will not exceed the potential.
It is very difficult to estimate future water availability for agriculture, since both the demand and supply may change (Lins et al., 1992). The causes of changes in water demand are numerous, and many of them are even unpredictable. Non-agricultural water use is expected to increase rapidly along with continued urbanization and industrialization. Competition between agriculture and other sectors for limited water resources will become more intense and in many cases may only be accommodated by increasing water use efficiency. Agricultural water demand is highly sensitive to variations in climate. The areas that can be cropped and the chance of failure can change significantly if temperatures rise or precipitation reduces, and the variation of the impact of change through the year is important. Additional parameters determining the water requirements and irrigation potential include the CO2 effect on plant transpiration, hours of sunshine and wind speed.
Changes in the water resources may be caused by climatic changes because river discharges are very sensitive to climatic characteristics (Nemec and Schaake, 1982; Nemec, 1983). The smaller the catchment area, the higher the sensitivity to climatic changes. The major hydrological processes and their interrelationships are complex as they involve seasonal variation of precipitation, evapotranspiration, storage and runoff, and differ from one catchment to the other.
This study has not considered possible negative effects of land degradation on the land's productivity. The extent and degree of degradation has been estimated globally (Oldeman et al., 1992), although it is not known whether these areas are at present stable, in a process of degradation or in a process of rehabilitation. Degradation of land may also affect crop yields and grassland production, but in temperate regions productivity losses are thought to be only marginal1. There are few studies for tropical regions. For example, loss of productivity and the resulting economic damage is significant in South Asia (FAO, 1994). With the available data it is not possible to estimate effects of the rate and status of the degradation process on the land's production potential. This information is crucial for studies such as this one. If yields are negatively influenced by degradation, the crop production scenarios may not be realized; as yields are important factors in determining the areas of arable land required for a certain volume of production, declining productivity may also cause arable land expansion needed to replace the lost productivity, and indirectly deforestation rates.
1 For the USA it has been estimated that continuation of current rates of soil erosion for the next 100 years would reduce crop yields by between 3 and 10% (Crosson, 1986; Alt et al., 1989).
The process of deforestation is a very important form of land degradation. If the information on global land use changes currently available is combined, it is clear that there is a large discrepancy between the estimates of deforestation (about 15 Mha per year in all developing countries) and the expansion of arable land and pastures (about 4 Mha per year in all developing countries). Alexandratos (1995) concluded that agriculture is not the sole cause of deforestation. However, analysis of the results of the forest resources assessment project reveals that a very important part of the forest conversions are long fallow and short fallow shifting cultivation2. Assuming that the rate of land conversion to urban land is of the order of 2 million ha per year (Alexandratos, 1995), the major part of the annual 15 million ha permanent forest clearing is, in fact, involved in shifting cultivation. Increasing land pressure and shortening rotation periods may cause a degradation and loss of productivity of the land, which is an important cause of deforestation2.
2 Results of the forest Resources Assessment Project show that in Latin America the conversion of tropical forest to short and long fallows makes up about 10% of total changes, but in Africa (-20%) and Asia (-20%) it is much more important. In Africa the conversion of closed forest to short fallow (small-scale subsistence farming) alone is close to 15% of the total changes; the other transitions in Africa with similar frequency are closed forest - open forest - fragmented forest - other land cover, representing the various progressive stages of forest depletion. In Asia the typical pattern is closed forest - long fallow - other land cover (mainly agriculture and waste land) and closed forest - short fallow - other land cover. There is an imbalance between the high rate of transition of closed forest to long fallow and regrowth of long fallows to closed forest. In Latin America the major type of conversion is closed forest to large-scale cattle grazing or to small-scale permanent cropping.
If land degradation plays a major role in the dynamics of land use of the developing countries, future scenarios have to account for this process. However, on the basis of the data currently available, it is not possible to make projections of land degradation and its effects on crop yields and grassland productivity.
Land scarcity, bought-in feed, manure disposal problems, concerns with water quality, disease and odours, and animal welfare, define the basic parameters of livestock-environment interactions in much of the EU and to some extent the USA. In intensive production systems the primary environmental concerns arise from animal excreta and associated problems of water pollution, gaseous pollutants and odours. Pig, cattle and poultry production generate both solid and liquid slurry manure of varying physical and chemical properties. The elements nitrogen, phosphorus and potassium are of particular concern. In some developing regions the nitrogen excreted by the animals exceeds the amount of fertilizer nitrogen used in crop production (Figure 9).
Because animal manure has a high biological oxygen demand (BOD), higher than human waste, a small leak into a water source can cause a major pollution problem. Water pollution from animal excreta can arise from direct runoff from farms into surface waters and from leaching of nutrients from the soil in periods of excess rainfall. As livestock production becomes more intensive, the incidence of pollution rises. High levels of BOD waste, nitrate and phosphate are major causes of pollution, with phosphorus generally the major cause of eutrophication of inland fresh water systems and excess nitrogen the cause of eutrophication in coastal or marine environments. An additional impact on water quality is waterborne illness. Cryptosporidium, a parasite whose oocysts are common in livestock, has been associated with various outbreaks of human illness in recent years.
Livestock production also generates a variety of gaseous pollutants. The CH4 emissions from enteric fermentation contribute significantly to the global CH4 injection into the atmosphere (Prather et al., 1995), and the scenarios presented in this study indicate that the contribution from the developing countries may increase substantially in the future. The CH4 emissions from the waste are not significant at present, but they may increase by one order of magnitude if the waste management changes into a lagoon/liquid storage system.
It is difficult to prevent emissions of ammonia from animal waste. Most of the ammonia is re-deposited again contributing to soil acidification. Part of the NH3 plays a role in atmospheric aerosol formation and chemical reactions where N2O is formed in the atmosphere. Odours from animal waste may become a particular problem in densely populated areas. Animal wastes contain as many as 60 volatile compounds, of which about a dozen contribute to bad odours.
One way to avoid environmental problems related to livestock production is by reducing livestock production, current in some countries in the E.U. This is, however, probably not a realistic assumption for most developing countries. In general, the efficiency of production can be improved by increasing animal productivity. Virtually all efforts that improve animal productivity will reduce methane emissions (Hogan, 1993) and nitrogen use efficiency. The composition of the animal diet can also greatly influence the digestibility and nitrogen use effiency1.
1 The following example illustrates the reason for low nitrogen use efficiencies in pig production: 1 kg of maize containing 15 g N is fed to a pig. At an efficiency of 20%, 3 g N will be found in the pork, and 12 g N in the excreta. For the maize crop N fertilizer was used with a recovery rate of 40%, the average for Europe (Van Duivenbooden, 1995). The losses of 20% of the nitrogen as NH3 from the excreta, 5% as NH, from fertilizer, 20% of the N in excreta + fertilizer lost by leaching and denitrification, yield a total of 15 g N. Hence, the leaching and gaseous N losses exceed the N found in the meat by a factor of 5. This situation may be common in current European pork production. However, considerable reduction of the environmental effects could be achieved by improving the digestibility of the feed. If the digestibility is improved to, for example, 40%, the N excretion could be brought back to half of that in the above example. If, in addition, the animal waste is used to fertilize the maize crop, part of the gaseous and leaching losses from synthetic fertilizer use could be avoided.
There are some examples of intensive livestock production where animal wastes are recycled, and crop and animal production integrated. Most pork in China comes from individual home producers. The production is concentrated in relation to land area. The primary constraint is the feed availability. In the face of chronic shortages of animal feed, improvements in yields is a priority, enhancing the value of animal waste as a fertilizer. Accordingly, Chinese pig breeders receive an allotment to purchase feed grains in exchange for manure. This is in marked contrast to the practice in the EU. In addition to the use as fertilizer, slurry and waste from pork production have been employed as a source of biogas from home use, further increasing its value.
The projections presented in Chapter 6 indicate that fertilizer in the developing countries will grow more slowly than in the past three decades. This can be explained in part by the relatively high levels of synthetic fertilizer use in many countries, particularly China, the Near East in Asia and North Africa. In the last two regions the fertilizer dressings are close to reaching their ceilings at current crop yields. In addition, the projected growth of total agricultural production is slower than the historical growth, and based on the method presented in Chapter 6, this will automatically lead to a slowing down of the growth of fertilizer use.
Many countries are experiencing local environmental problems related to intensive use of fertilizers. This seemingly contradicts the statistical data presented in Chapter 6. The presentation of average fertilizer use per unit of area for regions is misleading, however. For example, in Africa the average annual fertilizer use may be in the order of 10 kg NPK/ha, but estimates of fertilizer use by crop show that in many countries fertilizers are used by a small group of farmers only (FAO/IFA/IFDC, 1994), with application rates not different from those in developing countries (IFA, 1992). If the scenarios are realistic, this implies that both the percentage of fertilized fields and the fertilizer dressings will increase because application rates will have to increase along with the growth in crop yields.
Fertilizer use in Sub-Saharan Africa is limited to a very small group of farmers. Yield levels are very low compared to other regions. The fact that yields are low, but stable may be explained by an efficient system of organic recycling and animal waste use as fertilizer. As indicated in Chapter 7, the nutrient excretion by animals exceeds synthetic fertilizer use by more than a factor of 10.
Duxbury et al. (1993) estimated that roughly about 25% of nitrogen fertilizer may be lost by leaching. According to the estimates of fertilizer use by crop (FAO/IFA/IFDC, 1994) about half of the fertilizer is allocated to cereals, with rice consuming about one-third. The fertilizer dressings common in high yield rice varieties are much higher than those common for other cereals. Percolation rates in rice fields are generally high (FAO, 1986), and leaching losses for nitrogen may be in the order of 10%. In addition, denitrification and ammonia losses from rice fields are also high (30% and higher), as discussed in Chapter 8. Leaching losses from fertilizer depend on the soil and climatic conditions, as well as fertilizer application rates, and timing and mode of application. With split application the fertilizer recovery may be increased, and leaching losses and all gaseous losses may be reduced when compared to one single dressing.
FAO is currently involved in experiments in rice fields to test coated urea as a slow-release fertilizer, which has been shown to cause a significant reduction in both NH3 loss and N2O emission. Another option is the use of encapsulated calcium carbide, which, by producing acetylene, inhibits nitrification and has been shown to cause a reduction of N2O and CH4 emissions in rice fields (Banerjee and Mosier, 1989). Calcium carbide addition to fertilized fields has also been shown to lead to higher fertilizer recovery rates and higher yields in irrigated wheat, maize, cotton and flooded rice (various references quoted in Mosier et al., in press). Many other nitrification inhibitors are available, but these need to be tested in the field for their effectivity and their environmental side-effects (e.g. on soil life) (Kroeze, 1994).