DOE names the fix before weighing options. Its own rule says don't
Summary
GAO reviewed 21 mission need statements for the nuclear cleanup office's biggest projects; 11 pre-name a solution, which DOE Order 413.3B says should not happen — precisely so cheaper options aren't excluded at the start. The documented price of answers-first: a $1 billion overrun in Idaho on a technology reviewers warned against in 2001, $3 billion sunk at Hanford before a stop-work, and a $500 million disposal option sidelined while the $3 billion route remains the operative default.
The documents
Five primary documents, read against each other: GAO-26-108193⧉ (July 2026), the mission-need review; GAO-26-107820⧉ (March 2026), the cost-and-schedule ledger; GAO-24-106989⧉ (2024), the Hanford alternatives study; the 2025 High-Risk report⧉, where acquisition has sat for over three decades and the government's environmental liability — grown from $212 billion in FY1997 to $645 billion in FY2023 — remains listed, with recent progress credits on some criteria; and DOE's FY2025 Agency Financial Report⧉ — the audited self-portrait. The rule under examination is Order 413.3B, quoted throughout the July report: mission need defines the capability gap, never the answer.
View data as table
| Identify a particular solution | 11 | contrary to DOE Order 413.3B |
|---|---|---|
| Do not (aligned with the order) | 10 |
The money
The scale: 33 active capital projects estimated at $44 billion⧉, growing by more than $2 billion since 2022 — while the operations side grew by about $75 billion⧉ over the same window. Above it all sits the tank waste: roughly 90 million gallons at Hanford, Savannah River, and Idaho, booked at approximately $545 billion of liability. What answers-first costs is documented case by case: Hanford's WTP Pretreatment Facility absorbed about $3 billion before work stopped in 2012⧉, and EM has been paying for bypass projects since. Idaho's calcine offers the cleanest arithmetic: EM's own 2021 analysis put direct disposal at about $500 million against at least $3 billion for vitrification — and direct disposal was not considered feasible in early planning because of the regulatory frame, not the engineering.
View data as table
| Tank-waste financial liability | 545 | ~90M gallons at Hanford, Savannah River, Idaho |
|---|---|---|
| Operations cost growth since 2022 | 75 | most expensive operations activities |
| Active capital projects (33) | 44 | +$2B since 2022 |
| Hanford HLW facilities remaining | 20 | completing the halted Pretreatment Facility: +$9B |
The people and the sites
This story pins to four sites and a headquarters. At Idaho, the Integrated Waste Treatment Unit⧉ ran over 12 years late and almost $1 billion over by the time it operated in 2023 — on a steam-reforming technology a technical review team recommended against in March 2001, which entered anyway through a contractor's 2004 bid whose claimed maturity experts say was misrepresented. At Savannah River, in-tank precipitation was pursued past its safety problems until suspension in 1998. At Oak Ridge, the mercury-treatment mission need simply names 'a new mercury treatment facility.' And at Hanford, 's 2024 work found that managing portions of the tank waste as low-level rather than high-level waste could save billions⧉ — the remaining high-level-waste facilities alone are estimated at $20 billion, with completion of the halted Pretreatment Facility once priced at another $9 billion.
View data as table
| WTP Pretreatment sunk before stop-work | 3 | Hanford, invested by 2012; bypass projects since |
|---|---|---|
| Calcine vitrification vs direct disposal | 2.5 | ≥$3B vs ~$500M — the gap regulators' framing preserves |
| IWTU cost overrun | 1 | plus a 12-year delay; tech warned against in 2001 |
The disagreement between documents
The first disagreement is between 's rule and 's practice: Order 413.3B says the mission need 'should not propose a particular solution'; the majority of EM's large-project mission needs do, and EM officials told they did not see an issue⧉ because alternatives are explored later. 's case files answer that directly: it found instances where cost-saving options were never considered or pursued because the preferred solution was set at mission need. The second disagreement is between the ledger and the audit: 's financial report carries the $545 billion tank-waste liability, and the same report's independent audit flags a material weakness over EM's environmental liability estimate⧉ — the largest number in the story is the one 's own auditors trust least.
What happens next
's two recommendations target the entry gate: revise any mission need that pre-names a solution, and put independent experts — people outside EM's legal and regulatory agreements — into the mission-need review before a solution is agreed with regulators. concurred with both. The tests arrive with the next mission needs: at least five EM sites anticipate future projects over $100 million each, and Idaho's calcine deadline — waste ready for shipment by 2035 under the binding agreement with the state — keeps the $500 million-vs-$3 billion question live. 's ongoing Hanford work continues on the reclassification question worth billions.
- 11 of 21 mission need statements for EM's large cleanup projects pre-name the solution — 's own Order 413.3B says none should, because that is where cheaper options get excluded.
- The documented price of answers-first: IWTU's ~$1B overrun on a technology warned against in 2001; $3B sunk in Hanford's Pretreatment Facility before stop-work; a ~$500M calcine option sidelined against a ≥$3B one.
- The tank waste behind it all is booked at ~$545B of liability — and 's own audit flags a material weakness in that estimate.
- concurred with both fixes: no pre-named solutions, and independent experts in the mission-need review before deals with regulators.
Method notes. The 11-of-21 count and all case histories are -26-108193's; the $2B/$75B growth figures are -26-107820's comparisons against its 2022 baseline (different cost types: capital projects vs lifecycle operations). Calcine figures are EM's own 2021 analysis of alternatives as cited by ; direct disposal carries regulatory uncertainty that and experts note, and this piece does not treat it as risk-free. The $545B is 's FY2025 AFR figure for tank-waste-related liability specifically, within the government-wide environmental liability tracks as high-risk; the same AFR's audit reports a material weakness over EM's environmental liability estimate, so the figure's uncertainty runs in both directions. This story pins at Hanford, Idaho National Laboratory, Savannah River, Oak Ridge, and headquarters (multi-pin doctrine).
Sources(5) ▾
- U.S. Government Accountability Office, Nuclear Waste Cleanup: Changes Needed to Ensure DOE Is Not Prematurely Excluding Less Expensive Options for Large Projects (GAO-26-108193) (2026-07-02) — The new report: 11 of 21 mission need statements for EM's large projects identify a particular solution, contrary to Order 413.3B; the case histories (IWTU's 12-year delay and ~$1 billion overrun; the $3 billion invested in Hanford's WTP Pretreatment Facility before work stopped in 2012; calcine direct disposal at ~$500 million vs at least $3 billion for vitrification); the finding that mission-need reviews include no independent experts; both recommendations concurred. gao.gov · original document
- U.S. Government Accountability Office, Nuclear Waste Cleanup: Cost and Schedule Information for Major Projects and Activities (GAO-26-107820) (2026-03-02) — The trajectory ledger: combined costs for EM's most expensive capital asset projects grew by more than $2 billion, and combined costs for its most expensive operations activities increased by about $75 billion, since 's 2022 report. gao.gov · original document
- U.S. Government Accountability Office, Hanford Cleanup: Alternatives for Treating and Disposing of High-Level Waste Could Save Billions of Dollars and Reduce Certain Risks (GAO-24-106989) (2024-09-26) — The 2024 companion on the biggest prize: managing portions of Hanford's tank waste as low-level or transuranic waste instead of high-level waste could save billions; the remaining high-level-waste facilities are estimated at about $20 billion, and estimated completing the halted WTP Pretreatment Facility would cost an additional $9 billion. gao.gov · original document
- U.S. Government Accountability Office, High-Risk Series: Heightened Attention Could Save Billions More and Improve Government Efficiency and Effectiveness (GAO-25-107743) (2025-02-25) — The standing context: acquisition management has been high-risk for over three decades; the U.S. government's environmental liability remains on the High-Risk List — grown from $212 billion (FY1997) to $645 billion (FY2023) — with crediting recent progress on some criteria. gao.gov · original document
- U.S. Department of Energy, U.S. Department of Energy Agency Financial Report, Fiscal Year 2025 (2025-11-30) — 's own audited books: roughly 90 million gallons of tank waste at Hanford, Savannah River, and Idaho represent a financial liability of approximately $545 billion — and the independent audit identified a material weakness over EM's environmental liability estimate itself. energy.gov · original document
Comments
Always open. Logged-in readers can annotate paragraphs in place.
's project rulebook is explicit: when the nuclear cleanup office starts a big project, the mission need statement must not propose a particular solution⧉ — that is the stage where cheaper approaches live or die. reviewed 21 of the Office of Environmental Management's mission need statements for projects of $100 million and up: 11 name the solution anyway, and EM told it sees no issue. The receipts of answers-first run through every major site — Idaho, Hanford, Savannah River, Oak Ridge — while the tank waste behind these projects sits on DOE's own audited books at roughly $545 billion⧉ in liability. The audit adds a warning label: the liability estimate itself carries a material weakness.