Climate change and diseases are threatening global crop production. Agroforestry systems, which are characterized by complex  multispecies interactions, are considered to provide nature-based solutions for climate change mitigation and pest and disease regulation. Understanding the role of the abiotic environment and species interactions in shaping diseases and yield in agroforestry systems would enable us to develop effective ecologically-informed pest and disease management under a changing climate, support sustainable agricultural practices, and maximize the benefits gained from agroforestry systems. To gain such a comprehensive understanding of what shapes pest and disease levels and yield in agroforestry systems, we need to investigate how the interactions between agroforestry system components, such as trees, crops and
their associated organisms, vary in space and time, and how they are influenced by abiotic factors in terms of pests and diseases and yield.

In this thesis, my overarching goal wasto understand how microclimate and management impact major coffee pests and diseases, their natural enemies, and coffee yield, as well as farmers’ perceptions of climate change and climatemediated changes in disease dynamics and yield, with the aim of using these insights to optimize management decisions for smallholder farmers in southwestern Ethiopia. With this aim, I selected 58 sites along a gradient of management intensity, ranging from minimal management in the natural forest to moderate management in smallholder farms and intensive management in commercial plantations. As an approach, I combined observational and interview studies to examine i) the impact of shade tree species identity and canopy cover on coffee pests and diseases, ii) the effect of climate and management
on coffee berry disease and yield, iii) the impact of climate on a host-hyperparasite interaction, and iv) farmers’ perceptions of climate change and climate-mediated changes in disease dynamics and yield.

I found that tree identity affected the incidence and severity of coffee diseases, whereas insect pests were strongly affected by canopy cover, but in a species-specific way (I). Both climate and management affected coffee berry disease and yield. Importantly, the effect of climatic variables on disease and yield differed strongly between the developmental stages from flowering to ripening (II). In chapter (III), I found that the climatic niches of coffee leaf rust and its hyperparasite differed, with coffee leaf rust severity preferring high maximum temperatures, whereas the hyperparasite preferred cold
nights. The interviews revealed that the majority of farmers perceived long-term changes in one or more aspects of the climate, and the majority of farmers perceived an increase in coffee leaf rust and a decrease in coffee berry disease. Climate data also supported farmers’ knowledge on climate-disease-yield relationships (IV). 

Taken together, my thesis advances our understanding of the relationship between climate and management of coffee pests, diseases and yield, and this may contribute to the development of ecologically-informed pest and disease management strategies for coffee production and other agroforestry crops.