Lower Snake River Dams Replacement Power Study by E3

Replacing the four lower Snake River dams while meeting clean energy goals and system reliability is possible, but comes at a substantial cost, according to a study by the San Francisco environmental consulting firm Energy + Environmental Economics (E3). The study says that the cost of replacing the output would range from $415 million to $860 million per year through 2045, and that the reliability of the regional power system could depend on technologies such as hydrogen-fueled combustion turbines that are not yet readily available. The study assumes the dams would be breached in 2032 except for one of several scenarios that used 2024 as the breaching date.

Arne Olsen and Aaron Burdick of E3 discussed the study at the July Council meeting. Bonneville commissioned the BPA Lower Snake River Dams Power Replacement Study. The Council does not have a position on breaching.

The E3 presentation to the Council occurred on the same day the Biden administration released two reports on the Snake River dams. The White House’s Council on Environmental Quality announced the release of the E3 report and also a draft report on “Rebuilding Interior Columbia Basin Salmon and Steelhead,” prepared by the National Oceanic and Atmospheric Administration (NOAA), with input from scientists and managers at the U.S. Fish & Wildlife Service, Nez Perce Tribe and the State of Oregon. The draft report identifies actions that NOAA and the others consider to have the highest potential to significantly increase salmon abundance, including that “[f]or Snake River stocks, it is essential that the lower Snake River be restored via dam breaching” and “[f]or upper Columbia River stocks, it is essential to provide passage into blocked areas.”

According to E3’s presentation, the four federal dams – Lower Granite, Little Goose, Lower Monumental, and Ice Harbor – are capable of generating a total of approximately 3,500 megawatts of electricity (a megawatt is 1 million watts). During periods of high demand, such as the late June heat dome event in 2021 or a prolonged winter cold snap, the dams can provide sustained generation of 2,300 megawatts, helping to keep the Northwest power system reliable during emergencies. The report states that the dams generate an amount of electricity every year that is approximately equal to the annual electricity demand of Portland and provide essential services to the regional power grid such as smoothing the integration of renewable energy like solar and wind power. If the dams are breached, the electrical output of the dams will need to be replaced to ensure the Northwest power system remains reliable.

Earlier this year, Bonneville contracted E3 to build on the analysis performed in the U.S. Army Corps of Engineers’ Columbia River System Operations (CRSO) Environmental Impact Statement (EIS) regarding replacement resources and costs associated with a scenario where the four lower Snake River dams may be breached in the future. The CRSO EIS analysis examined a series of power replacement portfolios using the Council’s resource cost estimates in its 2021 Northwest Power Plan (February 2022) to reflect reasonable power replacement alternatives and associated costs. For its study, E3 relied on its own data sets, criteria, and objectives as inputs to its resource portfolio optimizer and dispatch model, called RESOLVE to create least-cost replacement portfolios.

Here are the key findings of the study:

• While wind, solar, utility-scale batteries, and combustion with natural gas and/or hydrogen are the main sources of replacement power, other resources also were considered, including small modular nuclear reactors, pumped-storage hydropower, geothermal, energy efficiency, demand response, floating offshore wind, and dual-fuel combustion plants that use both natural gas and hydrogen
• The power replacement costs would be substantial, even assuming emerging technologies are available. Replacement would require 2,300 – 4,300 megawatts of resources and would have an annual cost of $415 million to $860 million by 2045, for a total cost of $11.2 billion to $19.6 billion (in 2022 dollars) assuming 3-percent discounting over a 50-year time horizon following the date of breaching; the study assumes these costs would be paid by customers of public utilities with no federal cost share, an unlikely outcome
• Even in a best-case scenario, replacement power would cost several times as much as energy from the lower Snake River dams costs, which Bonneville estimates as $13 to $17 per megawatt-hour. Replacement resources would cost between $77 and $139 per megawatt-hour, according to the study.
• Electricity rates for public power customers of Bonneville would increase $100 to $230 per household per year (an 8-percent to 18-percent increase) by 2045
• The biggest cost drivers for replacement resources are 1) the need to replace the lost firm capacity for regional resource adequacy and 2) the need to replace the lost zero-carbon hydropower. Replacement becomes more costly over time due to increasingly stringent clean-energy standards and electrification-driven load growth
• Emerging technologies such as hydrogen, advanced nuclear, and carbon capture can limit the cost of replacement resources to achieve a zero-emissions electric system, but the pace of their commercialization is highly uncertain
• In the study’s economy-wide deep decarbonization scenarios, replacement without any emerging technologies requires very large renewable resource additions at a very high cost; for example, 12,000 megawatts of wind and solar at a (net present value) cost of $42 billion to $77 billion

Council members asked questions following the presentation, including whether the additional load for green hydrogen is factored into the study (it is not, Olsen said), whether the amount of demand response as one of the replacement resources is consistent across the various replacement scenarios (it is, mostly), and whether new transmission lines would have to be built for some of the replacement resources (some new lines likely would be built, but better estimates are needed of the amount and the potential cost).

Council members also were curious about relying on new resources that are under development but not yet available, such as advanced small nuclear reactors. “We assume breaching is far out in the future, and that the technologies will be available when we need them,” Olsen said. “That may be a heroic assumption, but there is a lot of money going into emerging technologies right now. At some point in the 2030s we expect them to be available.”