Previous Page Table of Contents Next Page


Currently there is a concern that activities related to food production in the developing countries may be inadvertently changing the environment. This concern is based on recent experiences of environmental pollution in many developed countries with intensive crop and livestock production. The use of mineral fertilizers1 and pesticides is growing in the developing countries. While excessive fertilizer application rates are a problem in some irrigated areas, there are also serious problems resulting from insufficient inputs of nutrients in other, particularly rainfed, areas. Additional problems in developing countries are associated with the claims laid on land and water by agriculture. The expansion of land in use for agricultural production is one of the major driving forces behind deforestation in developing countries. There is a growing demand for fresh water and agriculture has to compete with other sectors of society for scarce water resources.

1 The term mineral fertilizer in this report is used to denote all synthetic fertilizers, including urea.

Specific indicators can be used for monitoring the environmental effects of agricultural production. An indicator which has proven very useful for livestock production systems in many developed countries is the emission of ammonia (NH3). Other indicators are nitrous oxide (N2O) and methane (CH4). Methane emissions may be due for some 40% to agriculture (including biomass burning); for N2O the proportion is about 25% (Sombroek and Gommes, 1995).

Perhaps more than half of the global emission of ammonia (NH3) stems from agricultural production (Table 1). Most atmospheric NH3 is returned to the surface by deposition (primarily in rain). The re-deposited NH3 plays an important role in soil acidification, and also in the agricultural and general biospheric N cycle through its contribution to soil nitrogen. In soils part of the deposited NH3 is converted to NO and N2O by nitrification and to N2O (and N2) by subsequent denitrification (Firestone and Davidson, 1989). Ammonia is also involved in the Earth's radiative balance by its role in aerosol formation and through its transformation to N2O in the atmosphere (Dentener and Crutzen, 1994).

Methane is associated with many sectors in the society. Major agricultural sources are livestock production (through enteric fermentation and from animal wastes), rice paddies and biomass burning (deforestation, savannah and grassland burning and agricultural waste burning). Methane is not only a reactive gas involved in ozone chemistry in the lower atmosphere, but it is also a greenhouse gas (Sombroek and Gommes, 1995). A major part of N2O from agricultural sources stems from manure and mineral fertilizers. Nitrous oxide is inert in the lower atmosphere, but in the stratosphere it is involved in ozone destruction (Bouwman et al., 1995). In addition, N2O is a greenhouse gas, since it very efficiently absorbs infrared radiation (Sombroek and Gommes, 1995).

Global emissions of atmospheric ammonia (NH3) in million metric tons NH3-N per year

Domestic animals


Synthetic fertilizers


Undisturbed ecosystems




Biomass burning


Human excrement




Biofuel combustion


Total emission


Based on Bouwman et al. (1995).
a Global annual deposition (primarily in rain) amounts to about 60 million tons of NH3-N per year.

Emissions from deforestation and biomass burning have not been considered in this study. Estimates of deforestation for the developing countries were made in FAO (1995) and (FAO, in prep.). Biomass burning is a topic currently in study in the framework of the Global Emission Inventory Activity (GEIA).

Environmental research generally involves interrelated elements of the society-biosphere-atmosphere system. The wide range of issues is reflected in, for example, UNEP (1989), IGBP (1990) and Houghton et al. (1990; 1992; 1995). Scientific and policy oriented assessments of the environment also include reconnaissances of the future on the basis of scenario analysis using different models. For example, the current integrated assessment models include detailed descriptions of society-economy-agriculture-environment processes and interactions and more simple descriptions of the ocean and atmosphere subsystems. Integrated models are appropriate for scientific questions, but since emission scenarios can be adjusted to study the effect of, for example, certain abatement strategies, they can also be used to address policy-oriented problems. Other models to study the interaction between society, biosphere and climate. Global Circulation Models, can also include models of atmospheric chemistry.

The world population is expected to continue to grow rapidly during at least the coming 5-6 decades. Therefore, reconnaissances must cover a period of 50-100 years to study the environmental impact of population growth and changing societies. As forecasts are not realistic or even possible for such a long term, reconnaissance studies must be done in the form of scenario analyses. Comparison of the various published scenarios have made clear that further work in this field is required, in particular regarding agriculture and land use scenarios (Alcamo et al., 1995). The aim of this study is to develop a methodology to calculate regional emissions of atmospheric pollutants derived from agriculture on the basis of scenarios of the linkages between crop and livestock production and land use. The methodology can be used as an input for such global integrated models as IMAGE (Alcamo et al., 1994) to test the various scenario assumptions in a georeferenced model. The IMAGE model includes a version of the FAO AEZ approach, allowing to test the feasibility of the crop production scenarios.

The purpose of this study is not to present new estimates or forecasts of the magnitude of environmental effects caused by agricultural production and land use changes. In the study comparisons will be made between livestock production scenarios for feed requirements and animal waste production. Crop production scenarios will be compared on aerial extent and land use practices. The indicators for environmental effects resulting from agricultural production include NH3, N2O and CH4 emissions. The procedures used to estimate these emissions are based on the Intergovernmental Panel on Climate Change (IPCC) guidelines for National Greenhouse Gas Inventories (IPCC, 1995), the FAO Livestock Environment Study, and procedures proposed by the Global Emissions Inventory Activity (GEIA), a project of the International Global Atmospheric Chemistry Programme of IGBP (IGBP, 1990).

The scale of the study does not allow quantitative assessment of land degradation and other environmental pollution effects, such as contamination of soil and groundwater with nutrients and agricultural chemicals. Where relevant, these issues will be addressed qualitatively.

The forecasts up to the year 2010 made by FAO in the study "World Agriculture: Towards 2010" (FAO, 1993; Alexandratos, 1995) form the basis and starting point for the projection period of 1990-2100. The results will be presented as scenarios, not as forecasts. In the scenario development the general criteria formulated by Alcamo et al. (1995) were used as guidelines. These criteria include:

· Comprehensiveness. Scenarios must take into account all factors and processes that are related to agricultural production and land use.

· Reproducibility. Scenarios must be documented to ensure reproducibility and traceability of results.

· Uncertainty. Assumptions must reflect a wide range of views to express uncertainties.

· Consistency. Scenarios must be internally consistent.

To comply with these criteria, a range of uncertainty will be used for each entity around the 2010 forecast of FAO AT2010. Specific constraints must be taken into account to meet the consistency criterion. For example, the irrigated area must not exceed the irrigation potential, and the arable area may not exceed the area of productive land.

Chapter 2 of this report will discuss general aspects of the data and methods used. In Chapters 3 to 6 the scenarios for agricultural production are discussed in more detail, starting with animal production scenarios (Chapter 3). Scenarios for animal feed requirements will be developed in Chapter 4. Crop production is discussed in Chapter 5 and fertilizer use in Chapter 6. The implications of the above scenarios for land use, including those for animal feed use, will be presented in Chapter 7. In Chapter 8 the atmospheric emissions associated with the above scenario assumptions will be discussed.

Previous Page Top of Page Next Page