This dissertation explores a core tension in contemporary hydropower governance between two competing views: hydropower as a machine and as a geography. Policymakers rely on economic and engineering optimization models to plan energy system adaptations to climate change. The developers of those models tend to narrowly view the hydropower system as a combination of infrastructure and operations constraints that are knowable and replicable. They rely on quantifications of values to produce “optimal” outcomes given constraints. This is hydropower the machine. But more holistically, hydropower is a geography. Hydropower transforms waterways, implicating competing interests and values in limited resources: surface water and the landscapes it traverses. The first paper of this dissertation, published in Natural Resources Journal, takes an institutional economics lens to introduce the tension between the two views. Through an examination of two optimization models and an overview of hydropower law in California, that article argues that otherwise well-regarded and well-funded models of the hydropower system nonetheless fall short of informing policy because those models rely on an incomplete concept of governance. A partial resolution of the tension may be possible. The history of the development of one hydropower optimization model confirms the bulk of literature on co-production and collaborative science research. The participation of hydropower practitioners into the research and modelling process influenced the parameters and inputs to the model. And in doing so, the collaborative research produced a product that was credible, legitimate, and salient. As such, despite the mechanistic view of hydropower prevalent in the optimization model, the end result fit better within the immediate political context as communicated by the oversight committee. Additionally, the combination of formal rules for collaboration and the simultaneous creation of informal knowledge networks helped to bolster acceptability of the product. Even after formal processes and funding ended, the knowledge networks continued and were able to facilitate the eventual implementation of the model as a decision support tool. Yet, a closer look at just one aspect of hydropower governance—litigation in federal courts—reinforces the divide between hydropower the machine and the geography. A review of the past century of federal hydropower caselaw offers a window in to the multiple and evolving ways that water and energy law influence one another. Ultimately that history embraces variation as a core characteristic of hydropower governance, a condition which is at odds with the practical needs of modelling.