The farming systems described and characterized in Box 3.2 evolve and develop continuously, aiming at higher productivity, continuity and greater efficiency. Figure 3.2 illustrates changes in productivity per hectare over a 40-year period. The outcomes of science and technology may be reflected as changes in agricultural productivity, where productivity is expressed as returns from employing the factors of production - land, labour and external inputs such as nitrogen and phosphate. Productivity can be evaluated at different levels of aggregation, from crop, farm and agro-eco system level up to global level. For the purposes of this study the land productivity at the farming systems level is examined, but with sparse data for labour and input use in some cases only by way of illustration.

Agricultural production statistics databases for Africa (e.g., the agricultural statistics database, FAOSTAT) usually provide data only at the national level; therefore it is necessary to disaggregate the country data to the farming systems level. The farming systems distribution maps of Dixon and colleagues (2001) were used to estimate the proportion of the land area of each African country that falls within a farming system. Next, countries were selected that had 50 percent or more of the land area within a farming system. It is expected that the higher the proportion of a country that falls within a given farming system the more representative its national production statistics would be of trends within that farming system. One disadvantage is the omission of data for Nigeria from this analysis. In this case, all farming systems occupy less than 50 percent of the country's total crop area, thus the increased yields of maize grown in the savannah (cereal/root crop mixed system) during the 1980s and 1990s are not reflected in the system's trends (Smith et al., 1994). The irrigated system also presents a special case - a limited proportion of agricultural land area is irrigated in all countries except in Egypt, where all agricultural land is irrigated. Thus only Egypt is represented in the analysis of productivity trends in the irrigated farming system. Also, productivity trends for some crops in the sparse (arid) farming systems are strongly influenced by the fact that they are cultivated in irrigated systems. All farming system averages are weighted by crop area.

Changes in yields (land productivity) for several major farming systems using five-year averages from 1961 show the following general patterns:

• Root crop yields have grown modestly in farming systems in which they are the principal commodities (e.g., cassava in the tree crop-based, forest- based and maize mixed systems, yam in the cereal/root crop mixed system, and cocoyam in the forest based system). There has been hardly any growth in yields where these commodities are grown as secondary crops in the farming system (e.g., in the agro-pastoral or highland perennial systems).
• Cereal crop yields (maize, millet, sorghum, rice and wheat) have grown significantly in the irrigated and commercial farming systems.
• Rice is the only cereal whose yields have increased consistently in other farming systems, especially since the mid-1980s. But the increases have been modest in the farming systems in the humid zones (tree crop based, cereal/root crop mixed) where most of the rice is grown under rainfed conditions. The growth in the sparse (arid) and agro-pastoral systems reflects the fact that rice is grown mainly under irrigation in those systems.
• The trends in cereal crop yield generally show a slight drop in the second half of the 1980s and 1990s, especially for maize.
• The effect of civil conflict on agricultural productivity is illustrated in the dramatic decline in crop yields since the 1980s in the highland perennial farming system (Rwanda and Burundi), especially for the food security root crop, cassava.
• The steady increase in yields over the last decades has not kept pace with the population growth in all regions of Africa. Since the expansion of agricultural area was also limited, per capita food productivity declined, with a consequent decrease in food security.

Major discontinuities in the increase of agricultural productivity per hectare occurring in the Western world in the 1950s and in Asia in the 1970s - Green Revolutions - did not occur in Africa. These Green Revolutions occurred in farming systems dominated by rice, wheat or maize. In Africa such dominating systems are minimal, as demonstrated earlier.

A range of factors underlies the productivity trends described above. In this chapter factors that impact yield across the major systems are studied. Chapter 4 describes more closely the specific technical constraints that limit productivity of the dominant crops in the priority systems and that research must address over the next 10-15 years to contribute to the achievement of the un Millennium Development Goals.

<< Previous - [Page 39 of 171] - Next >>