Accelerating biodiversity change is a defining characteristic of the Anthropocene, and evidence accumulated from almost 30 years of research is often invoked to suggest that these changes will have catastrophic effects on ecosystems and the services they provide to humanity. In this thesis I use theory, empirical analysis and their combination to address key remaining issues surrounding the relationship between biodiversity, ecosystem function and ecosystem services. First, while the asymptotic relationship between biodiversity and ecosystem function is substantiated from experiments that randomly assemble plant communities, the response of ecosystems to directional biodiversity loss is highly variable. In the first two chapters I investigate how species level attributes (vulnerabilities, functional contributions) and community dynamics (compensation, non-random extinction) scale to affect individual and multiple ecosystem functions simultaneously. Second, a narrow set of plant-based ecosystem functions have come to dominate the field, and with few exceptions, linking these to the ecosystem services that directly affect human well-being has been challenging.

Inland fisheries provide millions of people with their primary source of essential nutrients (e.g., protein, omega-3 fatty acids, iron, zinc), but are threatened by hydropower development, overfishing and climate change. The last three chapters explore how fish biodiversity both responds to these threats, influences fisheries production and affects nutritional security in the Amazon, Earth’s largest and most productive river basin. Additionally, I evaluate how alternative strategies to ameliorate fisheries declines that hinge on substituting wild fish with farmed animals can meet nutritional goals. By combining theoretical and empirical approaches and integrating concepts from ecology, fisheries, nutrition and economics, this body of work illuminates key drivers surrounding the variation observed in how ecosystems respond to biodiversity change, and the implications of these changes for the sustainability of aquatic food systems. Compensation can maintain biomass production, but incur strong changes in community composition.

Differences in species vulnerabilities as well as in their functional contributions can predict the degree to which these compositional changes affect ecosystem functions. When considering critical ecosystems services, such as the contribution of inland fisheries to human nutrition, declining biodiversity always comes at the cost of increasing nutritional risk. This risk cannot be minimized by substituting wild fish with poultry or aquaculture species. Thus, investment in managing biomass production together with biodiversity, such as through protecting key habitats, maintaining riverine connectivity and enacting temporary closures, will maximize the long-term contribution of wild fisheries to human nutrition. Additionally, diversifying farmed animal production by interspersing high valued species with highly nutritious species can deliver both economic and food security benefits. More broadly, by illuminating how biodiversity contributes to the sustainability of food systems, this thesis provides new basic and applied dimensions to the field of biodiversity and ecosystem function.

Further, the findings presented here demonstrate how an interdisciplinary approach can shed light on the intertwined relationships between biodiversity, ecological dynamics and Earth’s ongoing sustainability.