Rainfed agriculture

The term Rainfed agriculture is used to describe farming practises that rely on rainfall for water. It provides much of the food consumed by poor communities in developing countries. For example, rainfed agriculture accounts for more than 95% of farmed land in sub-Saharan Africa, 90% in Latin America, 75% in the Near East and North Africa; 65% in East Asia and 60% in South Asia.^[1]

Levels of productivity, particularly in parts of sub-Saharan Africa and South Asia, are low due to degraded soils, high levels of evaporation, droughts, floods and a general lack of effective water management. A major study into water use by agriculture, known as the Comprehensive Assessment of Water Management in Agriculture, coordinated by the International Water Management Institute, noted a close correlation between hunger, poverty and water. However, it concluded that there was much opportunity to raise productivity from rainfed farming.

The authors considered that managing rainwater and soil moisture more effectively, and using supplemental and small-scale irrigation, held the key to helping the greatest number of poor people. It called for a new era of water investments and policies for upgrading rainfed agriculture that would go beyond controlling field-level soil and water to bring new freshwater sources through better local management of rainfall and runoff.^[2]

The importance of rainfed agriculture varies regionally but produces most food for poor communities in developing countries. In subSaharan Africa more than 95% of the farmed land is rainfed, while the corresponding figure for Latin America is almost 90%, for South Asia about 60%, for East Asia 65% and for the Near East and North Africa 75% (FAOSTAT, 2005). Most countries in the world depend primarily on rainfed agriculture for their grain food. Despite large strides made in improving productivity and environmental conditions in many developing countries, a great number of poor families in Africa and Asia still face poverty, hunger, food insecurity and malnutrition where rainfed agriculture is the main agricultural activity. These problems are exacerbated by adverse biophysical growing conditions and the poor socioeconomic infrastructure in many areas in the semi-arid tropics (SAT). The SAT is the home to 38% of the developing countries’ poor, 75% of whom live in rural areas. Over 45% of the world’s hungry and more than 70% of its malnourished children live in the SAT

There is a correlation between poverty, hunger and water stress (Falkenmark, 1986). The UN Millennium Development Project has identified the ‘hot spot’ countries in the world suffering from the largest prevalence of malnourishment. These countries coincide closely with those located in the semi-arid and dry subhumid hydroclimates in the world (Fig. 1.1), i.e. savannahs and steppe ecosystems, where rainfed agriculture is the dominating source of food and where water constitutes a key limiting factor to crop growth (SEI, 2005). Of the 850 million undernourished people in the world, essentially all live in poor, developing countries, which predominantly are located in tropical regions (UNSTAT, 2005).

Since the late 1960s, agricultural land use has expanded by 20–25%, which has contributed to approximately 30% of the overall grain production growth during the period (FAO, 2002; Ramankutty et al., 2002). The remaining yield outputs originated from intensification through yield increases per unit land area. However, the regional variation is large, as is the difference between irrigated and rainfed agriculture. In developing countries rainfed grain yields are on average 1.5 t/ha, compared with 3.1 t/ha for irrigated yields (Rosegrant et al., 2002), and increase in production from rainfed agriculture has mainly originated from land expansion. Trends are clearly different for different regions. With 99% rainfed production of main cereals such as maize, millet and sorghum, the cultivated cereal area in sub-Saharan Africa has doubled since 1960 while the yield per unit of land has been nearly stagnant for these staple crops (FAOSTAT, 2005). In South Asia, there has been a major shift away from more drought-tolerant, low-yielding crops such as sorghum and millet, while wheat and maize hasapproximately doubled in area since 1961 (FAOSTAT, 2005). During the same period, the yield per unit of land for maize and wheat has more than doubled (Fig. 1.2). For predominantly rainfed systems, maize crops per unit of land have nearly tripled and wheat more than doubled during the same time period. Rainfed maize yield differs substantially between regions (Fig. 1.2a). In Latin America (including the Caribbean) it exceeds 3 t/ha, while in South Asia it is around 2 t/ha and in subSaharan Africa it only just exceeds 1 t/ha. This can be compared with maize yields in the USA or southern Europe, which normally amount to approximately 7–10 t/ha (most maize in these regions is irrigated). The average regional yield per unit of land for wheat in Latin America (including the Caribbean) and South Asia is similar to the average yield output of 2.5–2.7 t/ha in North America (Fig. 1.2b). In comparison, wheat yield in Western Europe is approximately twice as large (5 t/ha), while in sub-Saharan Africa it remains below 2 t/ha. In view of the historic regional difference in development of yields, there appears to exist a significant potential for raised yields in rainfed agriculture, particularly in sub-Saharan Africa and South Asia.

Rural development through sustainable management of land and water resources gives a plausible solution for alleviating rural poverty and improving the livelihoods of the rural poor. In an effective convergence mode for improving the rural livelihoods in the target districts, with watersheds as the operational units, a holistic integrated systems approach by drawing attention to the past experiences, existing opportunities and skills, and supported partnerships can enable change and improve the livelihoods of the rural poor. The well-being of the rural poor depends on fostering their fair and equitable access to productive resources. The rationale behind convergence through watersheds has been that these watersheds help in ‘cross-learning’ and drawing on a wide range of experiences from different sectors. A significant conclusion is that there should be a balance between attending to needs and priorities of rural livelihoods and enhancing positive directions of change by building effective and sustainable partnerships. Based on the experience and performance of the existing integrated community watersheds in different socioeconomic environments, appropriate exit strategies, which include proper sequencing of interventions, building up of financial, technical and organizational capacity of local communities to internalize and sustain interventions, and the requirement for any minimal external technical and organizational support needs to be identified.

While absolute grain yield variations exist between different global locations as cited in this article, the potential for improved rain fed grain yields may be less than is suggested by a comparison between sub-Saharan and European locations for example, this applies particularly in areas where grain yield is primarily determined by the growing season rainfall. A more accurate formula for measuring yield potential is y x a = X, where X is grain yield in Kg/hectare, y is millimetres of growing season rainfall and a is the variable yield factor, a number that may vary somewhere between 5 and 20. Thus assuming a variable yield factor of 15 and a growing season rainfall of 220mm the formula would express as 220mm x 15 = 3300kg/ha yield potential. This formula, while not taking account of either the carryover benefits of stored rainfall in the soil profile resulting from out of season rainfall or the impact of temperature or soil fertility, still gives a more accurate picture of the degree to which actual grain yields are matching the region's potential yields and is a better basis for comparison between very different regions such as Europe and the sub-Sahara. A European yield of 5000kg/ha from a rainfall of 500mm results in a variable yield factor of 10 while a yield of 2500kg/ha in a 200mm rainfall area has a variable yield factor of 12.5, in such an example the lower yielding crop has actually been 25% more efficient in its rainfall utilisation than the higher yielding crop. Agronomy needs to be based on attaining the highest possible variable yield factor rather than the highest absolute yield, factor numbers as high as 17+ are achievable.

See also

• Water, Land and Ecosystems

References