Factors affecting variation in farm yields of irrigated lowland rice in southern-central Benin

For increasing rice production in West Africa, both expansion of rice harvested area and raising rice yield are required. Development of small-scale irrigation schemes is given high priority in national rice development plans. For realizing potential of the newly developed schemes, it is essential to understand yield level, farmers’ crop management practices and production constraints. A series of field surveys were conducted in six small-scale irrigation schemes in Zou department, Benin during the dry season in 2010–2011 to assess variation in rice yields and identify factors affecting the variation. The schemes were established between 1969 and 2009. Rice yields ranged from 1.3 to 7.8 t ha−1 with an average yield of 4.8 t ha−1. The average yield was only 2.9 t ha−1 for newer irrigation schemes developed in 2002 and 2009. Multiple regression analysis using farmers’ crop management practices as well as abiotic and biotic stresses as independent variables revealed that 75% of the variation in yields could be explained by five agronomic factors (fallow residue management, ploughing method, water stress, rat damage and N application rate) and two edaphic factors (sloped surfaces and sand content in the soil). Removing fallow residue from the fields for land preparation reduced yields. Yields were lower in plots ploughed by hand than by machine. Sloped surface, water stress and rat damage reduced yields. Yield increase due to N application ranged from 0.8 to 1.6 t ha−1. Higher sand content was associated with lowered yields. The low yields in new irrigation schemes caused by sub-optimal crop management practices suggest that farmer-to-farmer learning and extension of good agricultural principles and practices can increase yields. Organizational capacity is also important to ensure the use of common resources such as irrigation water and tractors for land preparation.