Large-scale versus small-scale agriculture: Disentangling the relative effects of the farming system and semi-natural habitats on birds' habitat preferences in the Ethiopian highlands

While the Western World is facing an inexorable decline of its farmland biodiversity following continuous intensification of production modes, agricultural landscapes in Africa are still largely dominated by small-scale subsistence farming operated by smallholders, mostly harbouring high biodiversity. However, as most African countries are confronted to an unprecedented population growth and a rapid economic development, efforts to intensify food production are widespread, with concomitant potentially negative effects on biodiversity. We conducted a study in a highly contrasted agricultural landscape of the Ethiopian highlands comprising two distinct farming systems: large-scale farming relying on modern, combine machinery and technology (e.g. enhanced crop varieties, application of herbicides, pesticides and synthetic fertilizers) vs small-scale traditional farming. Our objective was to disentangle the effects upon avian biodiversity of the operating farming system and the extent of semi-natural habitat features in the wider landscape. We performed a model selection approach to assess habitat selection by the overall bird community as well as the wintering, endemic and open habitat species, respectively. Our results show that habitat preferences of birds in the Ethiopian highlands were mainly driven by the amount of semi-natural habitats within the landscape, with varying effects depending on the farming system itself. While large-scale farming had overall more negative effects on birds, some typical open habitat species were mostly restricted to these wide-open landscapes. Our findings thereby suggest that both farming systems could coexist as long as semi-natural habitats are preserved and agricultural management maintained in its current practices. We emphasize the urgent need to conduct further studies integrating the socio-economic aspects in order to better predict future impacts of agricultural intensification processes on African farmland biodiversity.