Race to Finish $6.5 Billion Nuclear Cleanup by 2035 On Track At Reduced Cost

Why it Matters

The federal government is sitting on more than $6.5 billion in environmental liabilities tied to nuclear propulsion research sites scattered across Idaho, New York, and Pennsylvania. They represent decades of contaminated soil, aging radioactive structures, and facilities that once powered America's nuclear-powered Navy but now require careful, costly dismantlement.

A cleanup effort that could have dragged on for more than a century at a cost of nearly $6 billion is now on track to be finished by 2035, a fraction of that price, thanks to an interagency partnership that the Government Accountability Office (GAO) says is working.

GAO describes how the Department of Energy's Office of Naval Reactors (ONR) teamed with Department of Energy's Office of Environmental Management (OEM) to execute large-scale decontamination and decommissioning work at four sites. The savings projected so far are striking. It's an estimated $4.8 billion compared to what ONR would have spent doing the work itself.

The Scale of the Problem

The Department of Energy environmental liabilities tied to the Naval Nuclear Propulsion Program are spread across four sites:

• Naval Reactors Facility (NRF), near Idaho Falls, Idaho — $2.0 billion in fiscal year 2025 liabilities
• Kenneth A. Kesselring Site, West Milton, New York — $1.8 billion
• Knolls Atomic Power Laboratory, Niskayuna, New York — $1.4 billion
• Bettis Atomic Power Laboratory, West Mifflin, Pennsylvania — $1.3 billion

Together, they represent $6.5 billion in environmental obligations, costs that ONR officials describe as conservative, representing the upper limit of what cleanup is expected to cost. The federal government's environmental liability has been on GAO's High Risk List since 2017, flagged as an area vulnerable to waste, mismanagement, and in need of structural reform.

At the pace of pre-2018 funding, ONR estimated it would have taken more than 100 years to complete the decontamination and decommissioning work on its own. The new partnership changes that math dramatically.

How the Office of Naval Reactors Partnership Came Together

In May 2019, ONR signed a Memorandum of Agreement with DOE's Office of Environmental Management (OEM), giving it the authority to manage large-scale decommissioning work on ONR's behalf. The logic behind the decision was clear. OEM has spent decades building a nationwide contracting network specifically designed for nuclear facility decommissioning. ONR, by contrast, had neither the workforce nor the contracting infrastructure to execute cleanup at this scale efficiently.

ONR estimated the work now covered by the partnership would cost approximately $5.8 billion in 2025 dollars. OEM estimates it can complete the same inventory of work for roughly $1 billion, a potential $4.8 billion in DOE environmental cleanup cost savings if all planned work is completed.

The original target completion date was 2050. It has since been accelerated to 2035, in part to avoid the costly inefficiency of demobilizing and then remobilizing OEM's workforce over a longer timeline.

The work is being executed under two contracts. The first covers the ONR facility in Idaho with three major decommissioning projects, all on track for completion by 2031. The second contract covers Bettis, Kesselring, and Knolls, with 16 major projects planned across the three sites, all scheduled for completion by 2035.

ONR's Cleanup

The most concrete evidence that the partnership is delivering comes from Idaho. The demolition of the S1W, the Navy's first nuclear propulsion prototype, a submarine reactor built in the Idaho desert to train the sailors who would crew the USS Nautilus, is complete.

ONR's 2025 inflation-adjusted estimate for that project was $870 million. OEM completed it for approximately $117 million, a reduction of 87 percent from the baseline estimate, and a savings of roughly $753 million on a single project.

GAO attributes the savings to three factors: OEM's established contracting network; a graded approach to implementing protective controls during decommissioning (that is, calibrating safety measures to the actual contamination level of a given activity rather than applying a blanket standard); and innovative on-the-ground problem solving.

Among the innovations cited in the report is a sponge-blasting technique to remove PCB-containing paint from metal debris, allowing the metal to be recycled rather than disposed of as hazardous waste. By September 2025, OEM had recycled more than 2,100 tons of non-contaminated metal debris from the Idaho site, resulting in nearly $2 million in avoided disposal costs. Workers also restored an inoperable 125-ton overhead bridge crane at the facility rather than renting one, a decision GAO says could yield tens to hundreds of millions of dollars in additional savings.

In another case, rather than demolishing the S5G facility, a building that housed a fifth-generation nuclear propulsion prototype, crews preserved it for use as a warehouse. Building a comparable structure from scratch would have cost more than the entire decommissioning project.

The Risks Still Ahead

The GAO report is notably positive, but it does not ignore the pressure points. ONR's own planning documents identify potential funding shortfalls of between $64 million and $82 million per year from 2031 through 2035 for OEM work at Bettis, Kesselring, and Knolls. Officials acknowledged those shortfalls could grow, depending on annual budget decisions.

The highest-priority project under the second contract is the Materials Evaluation Laboratory at Bettis, in West Mifflin, Pennsylvania. It is both the agency's costliest remaining liability, and the site of soil contamination identified as a potential risk to nearby water sources. D&D work there is expected to begin in October 2026.

At the other end of the priority scale is the Hortonsphere at Kesselring, a steel-plated spherical containment structure 225 feet in diameter that once housed two nuclear propulsion prototypes. The contamination risk to the public and environment is assessed as low, and ONR officials acknowledged it is one project that may not be completed under the partnership if funding falls short.

In June 2025, ONR updated its sequencing plans to prioritize the highest-risk liabilities first, a decision that reflects the reality that not every project on the list is guaranteed to get done.

Who Ordered the Review

The report was mandated by Senate Report 118-188, which directed GAO to evaluate Naval Reactors' cleanup plans. It was addressed to four congressional committees, specifically the Senate and House Armed Services Committees and the Energy and Water Development subcommittees of both chambers' Appropriations Committees.

The Senate Armed Services Committee is chaired by Sen. Roger F. Wicker, with Sen. Jack Reed as ranking member. The Senate Energy and Water subcommittee is chaired by Sen. John Kennedy, with Sen. Patty Murray as ranking member. On the House side, the Armed Services Committee is led by Chairman Mike Rogers, with Rep. Adam Smith as ranking member, and the Energy and Water subcommittee is chaired by Rep. Chuck Fleischmann, with Rep. Marcy Kaptur as ranking member.

The report was provided to DOE and the Department of Defense for technical review, and both agencies' comments were incorporated. The findings stand as an assessment of a naval reactor site remediation effort that, at least for now, is tracking ahead of schedule and well under its projected cost.

Access the Legis1 platform for comprehensive political news, data, and insights.