The Transformer – May 24, 2007

Volume 4, Number 3

Replacing the power

from the Lower Snake dams


On Nov. 15, 2006, salmon recovery advocates released Revenue Stream, a compilation of earlier reports on the costs and benefits of removing the four Lower Snake River dams in southeastern Washington to restore Columbia Basin salmon species now on threatened or endangered lists due to the establishment and operation of the Northwest’s federal hydropower system.

Revenue Stream graphically demonstrates that dam removal will reduce costs to the region … to the tune of billions of dollars.

Central to the dam-removal bottom line was calculating the cost of replacing the electricity generated by the four dams, and doing so without compromising grid reliability. NW Energy Coalition policy staff provided this piece of the puzzle.

This edition of The Transformer considers the Coalition’s assessment of the costs of replacing the power from the four Lower Snake River dams and addresses the primary objections raised by Revenue Stream opponents to those calculations.

Replacing dams’ power relatively cheap

To estimate replacement costs for the electricity produced by the four Lower Snake River dams, NW Energy Coalition staff drew on data from the federal government, Northwest Power and Conservation Council, Bonneville Power Administration and other sources, including the Coalition- and Save Our wild Salmon-sponsored Tellus Institute assessment of clean energy resources in the Northwest.

Only clean energy options were considered. After analyzing the numbers, the Coalition determined that replacing the four dams’ electricity with energy efficiency and renewables would cost between $79 million and $179 million per year for 20 years. The lower figure represents a mix of 90 percent efficiency (“conservation”) and 10 percent new wind power; the top figure represents a greater reliance on renewables and/or higher-than-expected costs for acquiring conservation.

Figures such as $79 million (later revised to $86 million due to a calculation error) or $179 million might seem large. But even if ratepayers were to carry the entire replacement-cost burden, residential power bills would be expected to rise no more than 65 cents to $2 per month. Even the highest cost estimates are dwarfed by the billions of dollars in new income expected from additional tourism, commercial and recreational fishing, and other business income engendered by salmon restoration. Revenue Stream shows that investment in clean energy replacement would more than pay for itself.

Vested interests opposed to dam removal quickly reacted to supposed errors in the energy-replacement calculations.

  • Some bemoaned the potential loss of more than 3,000 megawatts of generating capacity.
  • Some wondered why surplus power sales income wasn’t figured in. (It was.)
  • Others doubted the availability and questioned the estimated costs of clean energy alternatives, or worried that the new renewable power would be difficult to integrate into the grid and would require expensive backup (“shaping”) resources.

Let’s consider these and other objections.

1. The output of the four Lower Snake River dams

The four Lower Snake River dams (Ice Harbor, Lower Monumental, Little Goose and Lower Granite) have a collective nameplate generating capacity of 3,033 megawatts. That number means little, however, since these are run-of-the-river dams with virtually no storage capacity; their generation is almost totally dependent on the amount of the snowpack and rate of runoff.

Thus, nameplate generation levels are possible on only a few spring days of maximum water flow. In actuality, the dams’ combined average yearly output is around a third of their nominal capacity — 1,075 average megawatts (about the load of Seattle City Light).

Even that number overstates the dams’ importance. Together, they produce in the course of a year only 790 average megawatts of firm power — the amount electric utilities can count on in a drought. And most of it comes during spring snowmelt when the region doesn’t need the power. When the grid is stretched thin in winter and late summer, these dams are good for only 425-525 megawatts — about 2 percent of the region’s load at those times. If Seattle were counting on the dams, it would have blackouts much of the year.

These kinds of dams, on this kind of river, are fairly unreliable sources of power. Because their power generation capability is so uncertain, the region already keeps ample generating capacity in reserve for droughts and cold snaps. From an energy need standpoint, the Lower Snake dams will hardly be missed.

2. Cost and availability of clean energy alternatives

Northwest Power and Conservation Council studies lend credence to the Tellus Institute’s conclusion that the Northwest has plenty of achievable energy efficiency and renewable energy to replace the four dams’ relatively small amount of power at the costs cited in Revenue Stream.

The strategy the Coalition studied was replacing the dams’ 1,075 average megawatts with 895 average megawatts of energy efficiency and 107 average megawatts of wind power. That adds up to 1,002 average megawatts rather than 1,075 because energy efficiency doesn’t require transmission and distribution and thus does not incur the 7.5-percent line losses experienced when getting generated power to end users. So fewer megawatts of efficiency are needed to replace the dams’ output.

BPA now spends about $1.5 million for a megawatt of energy efficiency (investments in efficient motors, appliances, insulation, windows, lighting, etc.). Clean renewable resources such as wind, geothermal and biomass are more expensive, about $3.5 million per average megawatt. At those costs, replacing the dams’ power with 90 percent conservation and 10 percent renewables costs $1.7 billion. Amortized over 20 years that’s about $86 million per year (in today’s dollars). The upper-range estimate is based on doubled energy efficiency costs and/or a higher proportion of renewables.

While critics are correct in putting the market value of the power now generated by the dams at around $600 million per year, the point here is to calculate the cost for the energy efficiency programs and renewable resources needed to replace the power.

3. Shaping and reserve capacity

While the four Lower Snake River dams have no long-term storage, they can provide some 1,000-1,500 megawatts of reserve “capacity” to the system for short (4- to 24-hour) periods. This capacity allows operators to vary or shape the dams’ output to generate more during peak hours of the day or during emergencies. This flexibility makes the dams’ more valuable than if they were purely run-of-the-river.

On the other hand, replacing the dams’ power primarily with energy efficiency provides much more value than the dams do. Efficiency is always available (firm) in any emergency or in times of high prices and doesn’t depend on good rainfall.

Efficiency tends to follow load, with savings generally occurring during peak periods. Recent analysis by Portland General Electric, for example, found that 100 average megawatts of efficiency reduce peaks by 140 megawatts. Using this rule of thumb, the 895 average megawatts (actually 968 when adjusted for line losses) used in Revenue Stream to replace the dams’ power provide 1,355 megawatts of peaking value — very similar to that provided by the dams. Even this value is conservative. Because efficiency reduces load, system reserves can be reduced by about 48 megawatts.

This added value (which, for simplicity, was not included in the Revenue Stream analysis) counterbalances the dams’ capacity benefit and so was not included as either a cost or benefit in the comparison.

4. Hydropower and wind integration

Some have suggested that removing the four Lower Snake dams would make it much harder, or more expensive, to integrate intermittent renewable resources — especially wind – into the grid.

In reality, shedding 968 megawatts (again, adjusted for line losses) of load through efficiency is a boon to wind integration. First, it frees up more than 1,300 peak megawatts of valuable transmission space on the grid. Second, it reduces reserve requirements — lower loads mean lower reserves. Similarly, it reduces the shaping and regulation needs for existing generation, since it is serving a smaller load, allowing those resources to shape and regulate new wind farms.

By the way, Texas, which has virtually no hydropower, is now integrating more than 5,000 megawatts of wind power.

Hydropower can be used to integrate wind, but other means exist. (And remember, the Lower Snake dams are hardly the only or biggest dams in the region.) Gas-fired turbines can easily ramp up and down to accommodate wind, adding not that much to the cost. As noted, due to the unreliability of hydropower, the region already has lots of underutilized gas-fired generation.

Wind power production varies much as loads do, and all power systems are designed to deal with changing loads. Integrating variable loads is hardly a new problem, and removing these dams will not be a large impediment to wind development. If their power is replaced mostly with energy efficiency, dam removal may prove a benefit to renewables development and integration.


The Coalition stands by the rather conservative energy-replacement cost analysis presented in Revenue Stream and encourages honest scrutiny and discussion of our methods, calculations and conclusions.

Revenue Stream has already altered the terms of the salmon restoration debate, and clean-energy advocates will have ample opportunities at the regional and national levels to defend and explain its findings. We welcome a comprehensive, adequately financed federal study of ALL the costs and benefits of dam removal and we are supporting the Salmon Economic Analysis and Planning Act introduced in the U.S. House of Representatives to do just that. We’re confident an objective appraisal will confirm most if not all of our analysis and set the Northwest firmly on the road to a clean, affordable and fish-friendly energy future.

Additional Resources:

Analysis by Portland General Electric (p.17 has relevant info)

What do you think?

We are interested in your reactions to these articles. We will print as many responses as possible in future editions of The Transformer. Please email comments to