Global analysis of yield benefits and risks from integrating trees with rice and implications for agroforestry research in Africa

While agroforestry is a well-established approach for agroecological intensification, rice is less often integrated with trees than other annual staple crops. The benefits and risks from rice agroforestry practices have not been systematically explored. Considering the need for strategies that may address low fertility and high degradation of arable soils and contribute to smallholder farm productivity, livelihoods and climate resilience, such exploration would both be timely and relevant. This study, therefore, reviews the published literature on integrating trees in rice production worldwide and provides perspectives for future research, with special attention to Africa, where the potential for sustainable productivity enhancement is deemed highest. Worldwide, six improved rice agroforestry practices are distinguished: hedgerow alley-cropping, short-term (0.5–4 years) improved fallows, pre-rice green manuring, biomass transfer, systematically arranged rice – tree intercropping and irregularly dispersed trees in fields. The rice agroforestry practices in the 87 publications reviewed were associated with 204 woody perennial species world-wide. Rice agroforestry practices provide a range of products and services to farmers but rice yield is the only quantitative performance indicator reported widely enough to enable meta-analysis. Frequently reported comparative or additional effects of fertilizer application, made it possible to include this aspect in the analyses. Across all types of agroforestry practices enumerated, the average effect of adding trees compared to a no-fertilizer and no-tree control is + 38%. The most beneficial practices in terms of enhancing rice yield were biomass transfer, pre-rice green manuring (100% of data points showing positive responses for both practices) and hedgerow alley-cropping (21% positive cases overall but 64% where fertilizer was not applied). Yield reductions occurred with fertilized intercropping compared to a fertilized mono-crop (in 95% of cases) and with the unfertilized short fallow practice (50% of data points showed yield reduction due to competition in the relay intercropping stage). Tree species that combined rice yield enhancements (alongside other products and services) with wide environmental adaptability across the African continent, include Sesbania rostrata, Aeschynomene afraspera, Acacia auriculiformis, Gliricidia sepium and Gmelia arborea. Yield benefits and risks from integrating trees with smallholder rice cropping depend on the type of agroforestry practice used and how each practice interacts with fertilizer application. Further research is needed to investigate the impact of different ways of integrating trees with rice cropping on wider environmental, social and economic sustainability aspects, that are driving increasing interest in rice agroforestry