Coral reefs are diverse ecosystems that provide services to millions of people around the world. However, these ecosystems are rapidly transforming in the age of the Anthropocene. In this thesis, I explore how a trait-based approach can be used as a methodological tool to understand and conserve coral reef ecosystems through prioritising ecosystem functioning. In Chapter 1, I systematically review the literature on coral reef fish traits through a response-and-effect framework. I identify where there is evidence linking traits to disturbances, management actions, and ecosystem processes, and where there are gaps in the literature. In Chapter 2, I use traits identified through my review of the literature to look at functional changes over time in no-take marine reserves in Kenya. I demonstrate that over 44 years of protection, both abundance and biomass increase, and the functional trait space of the fish communities within the reserves is novel and not stabilising. In Chapter 3, I expand on my work from Chapter 2, where I used surveys conducted by the same individual over multiple decades, to explore more participatory ways to monitor fish communities and biomass trends across a proposed Transboundary Conservation Area between Kenya and Tanzania. I find that fish traps can be used as a participatory monitoring tool to detect trends across a gradient of protection from fishing, and video transects can provide relatively precise biomass estimates and serve as an opportunity for collective learning. Finally, in Chapter 4, I shift to looking at traits on a global scale and apply phylogenetic models to predict the fecundity of 831 coral reef fish species. I then look at global drivers of site-level fecundity and demonstrate the positive impact of marine protection on the fecundity of important fisheries species. Overall, this thesis adds to a necessary diversity of tools needed to conserve coral reef functioning in the Anthropocene.