Interactions between forests, pollinators, livelihoods, and sustainability in rural Africa

The importance of pollination services to commercially traded crops that support human nutrition is well-recognised, and the impacts of pollinator decline driven by habitat loss in the developed world are now well-researched. These aspects are poorly understood in the developing world, however, despite the risks associated with pollinator decline being greatest in the Global South. Although there is growing public cognisance of pollinators and their importance, some key knowledge gaps remain, which may have consequences for pollinator conservation. Forests support livelihoods, especially in undeveloped regions, where forest-based livelihoods programs, including beekeeping, are frequently used as mechanisms of forest conservation. Little is known, however, about the role of pollinators for ensuring the provision of forest products to communities, which prevents accurate estimations of pollinator dependence. Unsustainable forest product harvesting contributes to forest loss with consequences for biodiversity and associated ecosystem goods and services including pollination. Despite this, the sustainability of forest use is rarely quantified, limiting the assessment of progress or regression with respect to sustainable forest management. This thesis investigated the interactions between forests, pollinators, livelihoods, and forest use sustainability. Smallholder farming households were surveyed in North-Western Zambia revealing substantial reliance on pollinator dependent crops and forest products. While farming was the dominant livelihood, forests were also important for most households and provided resources in between crop harvests. Despite the reliance on pollinators, there was little understanding of pollinator diversity and their role in forest pollination. Pollinator exclusion experiments on key forest product trees revealed variable reliance on biotic pollination. Symptoms of pollen limitation in one tree species indicated that pollinators could be a limiting resource. Pollen limitation could be exacerbated by forest loss which was found to have mixed effects on pollinators. Pollinator abundance and taxonomic richness were greater at deforested sites, possibly indicating sufficiently high floral diversity in agricultural areas to support pollinators. However, there were negative effects of forest loss on pollinator evenness, and evidence that forest loss has altered pollinator communities. Similarly, the effects of forest loss on honeybee foraging were mixed but showed some evidence of the possible importance of trees for honeybee foraging. In some cases, honeybee forage diversity was high in deforested sites, indicating that floral diversity in smallholder agricultural landscapes may be sufficient to support honeybees. Finally, a composite indicator was developed to quantify the sustainability of forest use by rural communities, enabling a comparison between current forest use and a future projection of more sustainable use, showing that forests could be used more sustainably by addressing harvesting practices. A possible attitude-behaviour gap was identified by comparing rural community attitudes to forests with forest use indicators, suggesting that there may be barriers preventing pro-forest attitudes being acted on. The results emphasize the importance of protecting forests and floral diversity in rural smallholder agriculture landscapes. Floral diversity in deforested agricultural areas will likely not satisfactorily substitute for forest habitats, however, since forests provide important resources for pollinators and people, including during seasons when agricultural landscapes are unproductive. Addressing key knowledge gaps around pollinators, and potential attitude-behavior gaps, may help to garner support for forest and pollinator conservation. Quantifying the use of forests using an indicator approach as developed in this thesis will be useful for tracking progress towards such forest conservation goals.