Multi-criteria evaluation of organic agriculture in sub-Saharan Africa using probabilistic elicitation of expert knowledge

Many African agricultural systems can potentially be described as organic. However, the capacity of organic agriculture to address food security issues in Africa is poorly known, as only few experimental studies are available in this region. Probabilistic elicitation is used here to represent knowledge of experts on one or more quantities of interest in a quantitative manner, and to describe the associated levels of uncertainty. The objective of this study is to explore how probabilistic elicitation of expert knowledge could be used to overcome the lack of quantitative experimental data on the performance of organic agriculture in Africa. This approach was applied in Senegal, Burkina Faso and Cameroon where 17 experts were interviewed on the relative performance of organic and conventional agriculture on different aspects of the local food security in sub-Saharan Africa. Our results show that the yields of organic systems are about 41% lower than the yields of conventional systems, while the prices of organic products are 34% higher than prices of products from conventional agriculture. According to the experts, the cost of labor, measured in man-days per unit of cultivated land area, is higher in organic systems. This study is the first to demonstrate the interest of the probabilistic elicitation of expert knowledge to evaluate cropping systems in situations where the experimental data are scarce or unreliable.