The conservation community is increasingly focused on managing nature explicitly for ecosystem services that provide benefits to humans, rather than for its intrinsic value. Though often debated, the consequences of conserving ecosystems for their services rather than for the explicit goal of biodiversity protection are not clear. While biodiversity can contribute to ecosystem services, this scientific link is fraught with uncertainty. How many and which species to protect is thus an important challenge for applied and theoretical environmental science, as well as new policy initiatives, such as the intergovernmental platform on biodiversity and ecosystem services (IPBES).

My first chapter presents new theory to address the extent to which a goal of ecosystem service provision creates a significant economic incentive for biodiversity protection when facing uncertainty over how biodiversity produces services. I derive the level of biodiversity protection that maximizes ecosystem service provision under uncertainty and define a criterion that can be used to determine when managing for ecosystem services economically justifies broad-scale biodiversity protection. I illustrate the utility of this criterion, applying it to several ecosystem services across locations (pollination, wave attenuation, and carbon storage). Next my second chapter considers factors other than uncertainty that may increase alignment between management to obtain ecosystem services versus explicitly to conserve biodiversity.

I find the objectives of conserving biodiversity and delivering ecosystem services align more frequently than would be predicted by considering only the known and direct contributions of abundant species to current ecosystem service provisioning. Specifically, I review existing literature on four factors that strengthen alignment, finding the following. First, the number of species that contribute significantly to services is much larger than many current estimates from observational studies. Second, coarse management actions to enhance ecosystem services, such as protecting a location or restoring a habitat type, can provide "incidental" conservation benefits at no additional cost. Third, managing for services should provide broader benefits to biodiversity by disproportionately protecting at-risk species if the species most at risk of local extirpation also contribute significantly to one or more services (i.e., rare or threatened species).

Fourth, managing for even a single, relatively low value ecosystem service can have indirect benefits to biodiversity if the species that are significant to service provision have functional ecosystem roles that promote biodiversity (e.g., if they are habitat forming species or keystone predators). However, significant uncertainty remains, especially surrounding the service provisioning roles of the vast number of rare species in nearly every ecosystem.

Finally, in my third chapter, I aim to reduce uncertainty about the role of biodiversity in the provision of a specific ecosystem service (fisheries yields) in variable climates. I develop theory to predict the consequences of within-year temperature variability for yields and explore the role diversity might play in offsetting potential impacts. I hypothesize that higher functional diversity (FD), measured with traits related to species' responses to temperature, can mitigate impacts from temperature variability on yields. Using a global marine fisheries dataset, I find that within-year temperature variability reduces yields but current FD of targeted species largely offsets this effect, avoiding annual losses of 8% on average globally relative to if FD were degraded to the lowest level observed in the data.

All together, this dissertation contributes our understanding of when management strategies targeting ecosystem services versus biodiversity conservation align or when they do not. This knowledge is critical to new policies initiatives like IPBES with the dual of objective of protecting biodiversity and benefits nature provides people.