Coordinating Mid‑Columbia Dams Could Raise Revenue 3.3%

Estimated reading time: 5 minutes · Last updated: 28 August 2026

Modeling by nlr.gov for Grant Public Utility District (PUD) shows that operating the seven dams in Washington’s Mid‑Columbia cascade as a centrally coordinated resource could raise annual revenue across those dams by about 3.3%, with individual dams seeing increases between 2.8% and 4.3%. The researchers ran a fully coordinated dispatch against an uncoordinated, owner‑segmented case and found that coordinating storage and release allowed operators to shift generation into higher‑price hours and reduce spill. The U.S. Department of Energy Hydropower and Hydrokinetic Office provided funding for the project, which the research team describes in a paper in Energy (doi: 10.1016/j.energy.2026.142148).

“In a cascaded hydropower system like the Mid-Columbia, water flowing out of one dam greatly influences the operation of the one right below it,”

Wesley Cole, NLR researcher and lead author

Why a cascaded hydropower system like the Mid‑Columbia resists simple dispatch

In a cascaded hydropower system like the Mid‑Columbia, an upstream release directly sets the flow available to the plant immediately downstream; that physical coupling means one owner’s decision constrains the next. The Mid‑Columbia cascade contains seven dams whose flows and storage interact, and ownership is split among multiple parties: two federal agencies, five nonprofit municipal corporations and Grant Public Utility District, which operates the last two—Wanapum and Priest Rapids.

That mix of owners removes the natural single-owner incentive to optimise the chain. Grant PUD is particularly exposed because it runs the last two dams and therefore receives whatever water the upstream owners release; at the same time Priest Rapids lies above the Columbia’s longest free‑flowing reach and a salmon spawning ground, imposing ecological and operational limits on timing and quantity of releases. Those constraints are central to evaluating whether coordination can be implemented without harming other water uses.

How NLR tested coordination: combining river and grid models

The research team ran two parallel simulations that combined river hydraulics with production‑cost grid modelling. In the coordinated scenario the seven dams were treated as if centrally dispatched to minimise production costs across the Western Interconnection; in the uncoordinated case the dams were divided into four groups by owner or power marketer and operated independently.

By explicitly combining river flow and production cost models, the team could trade off storage decisions against hourly electricity prices across the Western Interconnection. That let the model show when holding water upstream for later high‑price hours reduces total system value, and conversely when upstream releases timed to market signals create more opportunity for downstream plants to operate during peak-price hours.

Who benefits and how the gains materialise

The coordinated dispatch increased annual generation at all seven dams and produced an average revenue uplift of 3.3% across the cascade; individual plants posted revenue gains between 2.8% and 4.3%. The mechanism is straightforward: when the full resource is optimised, storage levels and release timing are aligned so more water is available to generate during higher‑priced hours rather than being spilled or used at low‑value times.

Because the modeled owners are public utilities, NLR frames the gains as likely to flow to customers rather than private shareholders, although the study emphasises that capturing those gains requires changes in operational coordination, information sharing and possibly market participation such as joining day‑ahead markets being developed in the West.

Barriers: ecology, governance and the cost of coordination

The modelling assumes coordination is operationally feasible, but real‑world limits matter. Priest Rapids’s proximity to a salmon spawning ground means releases are constrained by fisheries requirements, and those ecological limits reduce the degrees of freedom available to shift generation into peak hours. The paper and NLR’s work deliberate these constraints rather than ignore them.

Governance is the other material barrier: the cascade is split among multiple owners, and bringing together a pair of federal agencies with five nonprofit municipal corporations would create transaction costs, require sharing operational information and could necessitate legal or contractual changes. The study’s stated aim was to assess whether the projected revenue and efficiency gains justify the expense and effort of establishing that institutional framework.

Why coordination might spread — and what could stop it

The bottom line

The team's paired river‑and‑grid modelling for Grant PUD demonstrates that coordinated dispatch of the seven Mid‑Columbia dams can raise generation and produce a measurable revenue uplift—about 3.3% systemwide and 2.8%–4.3% per dam. Those gains occur when storage and release choices are timed to market signals so water is used during higher‑value hours. Putting the modelled improvements into operation will require addressing ecological constraints at Priest Rapids, negotiating governance arrangements that involve a pair of federal agencies and five nonprofit municipal corporations, and deciding whether to participate in Western day‑ahead markets. The study provides a quantitative basis for those governance and market decisions.

What to watch

• watch whether Mid‑Columbia owners decide to pursue formal participation in Western day‑ahead markets; no date has been set.
• watch for publication of follow‑on modelling that expands the coordinated dispatch framework to other cascaded systems; no date has been set.