During my MS in environmental engineering at Stanford, I spent a year as project manager on a Department of Energy funded FERC initiative looking at water data in river basins. The question we were trying to answer:
If stakeholders had better access to river basin data, could they make better decisions about how to manage it?
Hydropower relicensing decisions, dam removals, and basin management plans all depend on water data. But that data is scattered across dozens of agencies, tribes, utilities, and research institutions, often in formats that don't talk to each other. The Potter Valley Project, a hydroelectric facility connecting the Eel and Russian River basins in Northern California, became our case study for a bigger national conversation: if we improved access to and integration of that data, could stakeholders make better trade-off decisions in basin management?
I managed the project from start to finish, which mostly involved:
The work fed into a broader Stanford Woods Institute effort called the Uncommon Dialogue on Hydropower, River Restoration, and Public Safety, which brought together hydropower companies, tribes, NGOs, and government agencies to find common ground on decarbonizing hydropower while restoring river health. That coalition's work later helped shape over $3 billion in dam and hydropower funding in the 2022 Infrastructure Investment and Jobs Act, and the same tribes-industry alignment was covered by the Wall Street Journal.
The project wrapped with a peer reviewed technical report through Pacific Northwest National Laboratory, published on OSTI: Improving the Discovery, Access, and Use of Water Data for Basin-Scale River Management: Phase 2 Workshop Report. It documents the May 2021 workshop outcomes alongside interactive data stories for five river basins across the country, including Potter Valley's Eel and Russian rivers.
Looking back, I learned as much about managing stakeholders as I did about water data. Tribes, hydropower operators, and conservation groups often want very different things from the same river, so getting everyone to agree on what "good data" looks like took most of the effort. The technical side was actually the easier part.
Links: