|
For immediate release: September 3, 2025
Below:
- Multiple studies underway
- Current legislative deliberations
- Cost and schedule overruns
- Is the two-site production plan feasible?
- Study Group comments
Albuquerque, NM -- The Department of Energy (DOE), National Nuclear Security Administration (NNSA), and Government Accountability Office (GAO), are conducting seven reviews of NNSA's progress in acquiring the capability to produce plutonium warhead cores ("pits"):
- DOE Office of Enterprise Assessments (OEA) "special study" of pit production "leadership and management." The completed study is due by Dec. 8, 2025. See this Aug. 19 press release.
- GAO, biennial assessment of NNSA major projects, similar to GAO-23-104402. The 2023 review covered NNSA projects costing $100 million or more, including ten major line-item projects required for pit production, including the Savannah River Plutonium Processing Facility (SRPPF) and nine major projects at the Los Alamos National Laboratory (LANL). This review is likely to be complete in the fall of 2025.
- GAO, assessment of NNSA pit production acquisition, likely complete in the second half of 2026, per congressional mandate.
- DOE, NNSA, and Office of Management and Budget (OMB), preparation of NNSA's FY2027 annual congressional funding request, including for pit production. Required in early February 2026 but often late. This request is likely to incorporate the Administration's conclusions and near-term ambitions for pit production as of approximately the end of the year.
There is also likely to be a FY2026 Stockpile Stewardship and Management Plan (SSMP) in preparation.
Like GAO's review of pit production, GAO's review of NNSA major projects (the second item above) was requested by the armed services committees. GAO has continuously found, over the last 35 years, that DOE's (and NNSA's) inadequate management and oversight of contractors left these agencies vulnerable to fraud, waste, and abuse. In 2009, this "High Risk" finding was narrowed to NNSA acquisition and program management plus DOE Environment Management, a status these programs "enjoy" today.
2. There are also routine legislative deliberations underway.
These will result in:
- Passage of the FY2026 National Defense Authorization Act (NDAA). There is no deadline for this but it is expected in 2025. The two armed services committees have completed their markups (House Rules Committee print, 8/21/25, with accompanying report); Senate Armed Services as reported, 7/15/25, with accompanying report). We have already discussed some of the implications of the Senate bill, which would increase pit production requirements. From the Senate report (p. 348):
The committee recognizes that dramatic shifts in the international security environment have compelled the Department of Defense to reassess anticipated nuclear weapons stockpile requirements, which are likely to levy additional production needs on the nuclear security enterprise beyond the original targets established in 2015. As such, the committee acknowledges that the 80 pits per year production target is likely insufficient to facilitate timely modernization of the U.S. nuclear weapons stockpile in a manner that preserves deterrence against growing threats from China, Russia, and North Korea.
As noted on August 15, Congress likely to set unrealistic schedules again. They believe they can legislate capability -- and do so by an early date certain. The increased production requirements being proposed by the Senate for LANL -- specifically, ≥50 ppy by FY2029, and ≥80 ppy by FY2032 -- were deemed inadvisable by the Institute for Defense Analyses (IDA) in a 2019 study requested by Congress. As the IDA put it 6 years ago (pp. vi, vii),
"...attempting to surge at LANL offers the only possibility for producing significantly more than 30 ppy by 2030. IDA's asessment is that producing more than 30 ppy using a two-shift "surge" at LANL appears technically possible, but would be very challenging to execute and could jeopardize executing the PSP [Plutonium Sustainment Program, 30 ppy with moderate (50%) confidence] as well as other LANL programs. Producing 80 ppy using this strategy is unlikely.
...Trying to increase production at PF-4 by installing additional equipment and operating a second shift is very high risk.
This inadvisable "second shift" operation is exactly what LANL is doing. LANL has been operating its plutonium facility (PF-4) around the clock during weekdays for more than two years. This April, LANL began 24/7 operations in PF-4, extending night-shift work through the weekend. At PF-4. 24/7 work in support of pit production is expected to continue at least until the completion of the "30R" subproject, currently slated for the end of FY2032, which when complete would allow LANL to make 30 ppy in a single shift (p. 173).
In effect, the Senate NDAA would permanently mandate a second shift at LANL, at least until SRPPF achieves full rate production, 10-15 years from now.
In 2017 NNSA assessed, in its only detailed study of pit production alternatives to date, that LANL could never be a reliable, enduring pit production site absent an entirely new, dedicated pit production facility. The Senate NDAA embodies the hope that LANL can be precisely that -- a reliable, enduring pit production site capable of producing at least 80 ppy indefinitely, from FY32 onward.
- Passage of the Energy and Water Development and Related Agencies Appropriations Act or its functional equivalent(s) -- one or successive continuing resolutions (CRs) ("clean" or with "anomalies," etc.), and/or omnibus appropriations bill(s). Something, at least a CR, must be passed by Sept. 30, 2025. In the House, the appropriations bill is ready for floor debate (H.R. 4553, House Rules Committee print, with report). From the Senate Appropriations Committee, only a summary is so far available.
Of note, the House Energy and Water bill does not include monies available to NNSA from the recently-passed reconciliation bill, which are considerable. Most of the increase proposed for NNSA as a whole and for pit production in particular comes from the reconciliation bill, as we noted in June. This table (p. 2) shows the eight categories of additional funds available to NNSA in the reconciliation bill, the largest three of which are applicable to pit production.
If the House Energy and Water bill became law in its present form, $2.833 billion (B) in discretionary funds would be available to pit production, of which $1.505 B would be provided to LANL, $1.205 B to the Savannah River Site (SRS), and $0.122 B to other sites. Reconciliation funds would augment these totals along the lines of NNSA's budget request, which said that in FY2026, NNSA requires a total of $3.795 B for the pit production program as a whole, $1.716 B at LANL, $1.933 at SRS, and $0.145 B at other sites.
In addition to these two funding sources (discretionary, and reconciliation, the latter a type of mandatory spending), there was also an emergency appropriation of $1.884 B for Weapons Activities earlier this fiscal year in the "Disaster Relief Supplemental Appropriations Act (Public Law 118-158, p.23), which provided this funding to cover damages to NNSA sites from hurricanes Helene and Milton. This emergency appropriation would "remain available until expended."
In addition to these three funding sources, there are also unobligated balances from prior fiscal years. In May, we guesstimated that the total of these unspent funds was roughly $13 B in Weapons Activities, using USAspending.gov.
3. These studies and these proposed congressional actions are in part responses to egregious pit production schedule and cost overruns.
Schedule overruns
The story of broken pit production commitments at LANL is three decades long ( DOE/NNSA War Reserve (WR) pit production commitments and results at LANL). At present, "reliable" production of 30 pits per year (ppy) is, according to NNSA, 7 years away. In 1996, reliable production of 100 ppy (!) was also said to be 7 years away, and this is how it has more or less been in all the years in between, despite what is now a total investment (just since 2005) of $13 B ( details; graph). (If constant dollars are used, the total investment in pit production at LANL over the 2005-2025 period swells to $15 B, or $17 B over the 2000-2025 period.)
NNSA and its LANL contractor currently have a statutory obligation to produce at least 20 War Reserve (WR) pits by the end of this month, 29 days from now. No new WR pit production has been announced so far this fiscal year.
As mentioned on Aug. 15, the anticipated 4-year delay announced in January 2023 for completion of the basic equipment needed to produce 30 ppy at "moderate" confidence at LANL was a particularly major blow to the LANL schedule, affecting the entire pit program.
Schedule delays are costly. At LANL, each year's delay will cost at least a billion dollars in operating ("program") expenses. The opposite is also true. Higher estimated costs are often indicative of expected schedule delays.
At SRPPF, the current schedule has not dilated greatly -- yet -- from the ballpark estimates of 2017 and 2018. In 2017, NNSA said it could likely construct what is now SRPPF by 2031, and be producing 80 ppy there by no later than 2036 ( slide 9). The 2018 Pit Production Engineering Assessment (EA) said SRPPF could be completed by 2030 at the latest, and would require 5-10 years to ramp up to 50 ppy, i.e. until 2035-2040 ( pp. 8-9).
Today's estimates for completion of construction run as early as 2032 and as late as 2035 ( p. 203). In April 2024, then-Administrator Dr. Jill Hruby was "focused" on completing SRPPF by 2032 and achieving rate production by 2035 -- the near ends of the EA windows. At the same time however, NNSA's upper-end cost estimate for SRPPF increased from $11.1 B to $25 B, more than doubling in the worst case. It is hard to reconcile the increased scope of work now required at SRPPF ( p. 200) and the predicted additional $7 B - $13 B in costs, with Dr. Hruby's schedule optimism. In the Study Group's judgment, NNSA is unlikely to achieve CD-4 (completion of construction) by FY2032.
This work was necessary because NNSA has not published anything approaching a valid cost estimate for the acquisition cost of pit production capability. Part of this would be a high-quality cost estimate ("baseline" cost) for its main pit production projects (LAP4 and SRPPF), which we also do not have. In fact the decision to build each of these projects is being " reaffirmed" in two of the studies listed above, a necessary prior step to producing these baselines.
No doubt more clarity about schedules and costs will be available for both sites as a result of the above seven studies.
4. Is the two-site production plan even feasible?
When the two-site decision was made in May 2018, NNSA was working with construction cost estimates of $2 B at LANL for a 30 ppy capability ( slide 2) and up to $4.6 B ( slide 8) to $5.4 B ( slide 9) at SRS, or $6.6 B to $7.4 B in all for construction. Six years later in April 2024, NNSA estimated the total construction-only cost of its two-site strategy at $28 B - $37 B, 4-5 times as much as was thought in 2018.
Now -- and this time including program costs -- the current total estimated cost rises much higher, to $49 B - $58 B as we explained in July. This is roughly 7 times as much as the cost NNSA was working with when it made the two-site decision.
So far, that resiliency has not manifested. What we see instead is competition for scarce resources -- gloveboxes, process design engineers, and so on.
In this regard we recall the observation of the 2017 NNSA AoA, that "split production" (as it was called then) would be very costly ( pp. 45-46), which was part of why NNSA rejected using PF-4 on a permanent basis for pit production ( pp. 2, 46-48). In 2017, NNSA had no idea of just how "costly" split production would prove to be, not just in dollars but also in time lost.
For its part the IDA said in 2019, diplomatically, "...eventual success of the [two-site] strategy to reconstitute plutonium pit production is far from certain" ( p. v).
5. Comments
Study Group director Greg Mello:
"We do not believe NNSA can manage the successful construction and operation of two pit "gigaprojects," under any circumstances. Management reforms will not be enough. NNSA's two-site plan is fundamentally flawed. Both of NNSA's major pit production projects, the one in New Mexico and the one in South Carolina, are in trouble and are being reevaluated. A new 'Analysis of Alternatives' is needed, as logic and DOE project management rules require.
"NNSA's two-site pit plan was predicated on cost estimates that are now 4-7 times lower than today's more realistic estimates. It ought to be clear to all parties that NNSA's 2018 attempt to "split the baby" between Los Alamos and the Savannah River Site was a terrible decision.
"NNSA has no valid estimates of pit acquisition costs at either main site. Published estimates omit program costs, at both sites. NNSA grossly underestimates the cost of acquiring pit production at LANL in particular by omitting the majority of the costs, which are not construction line-items. In short, NNSA is hiding LANL's pit costs, which are rising each year.
"Not only is NNSA (still) low-balling pit costs, but these costs are carefully segregated from its estimated costs for designing and producing warheads -- specifically, the cost of the W87-1 warhead. Given LANL's low production capacity and the limited life of main production facility, the W87-1 pit is likely to be the only pit LANL ever produces in quantity, assuming it produces pits at all. Why aren't LANL's pit costs included in the estimated cost of the W87-1? If pit costs were included, the W87-1 unit cost would be at least $100 million. The W87-1 warhead is entirely unnecessary, provocative, and wasteful.
"For almost thirty years, DOE and then NNSA have said LANL has been about 7 years from producing 30 or more pits per year. Carefully adding up each year's appropriation, we can see that NNSA has invested at least $15 billion in LANL pit production over that period. Using NNSA's current projections, LANL needs at least another $15 billion to get to the 30-pit-per-year goal. Is there no limit to this waste?
"There are now 27 days remaining in this fiscal year. LANL is required by law to have made 20 "War Reserve" pits by then. Where are these pits? Do they exist? I guess we'll find out soon.
"If LANL hasn't made 20 war reserve pits by September 30, after 3 decades of trying, NNSA needs to put the proposed LANL pit factory out of its misery and focus on the only viable long-term option it has, namely the SRPPF in South Carolina.
"The forward-looking costs -- the money yet to be spent -- at LANL to acquire pit production are increasing, rather than decreasing, with each passing year. Why? Investments at LANL appear to have negative "earned value." This is a hole that keeps getting deeper. Stop digging, please.
"Does the U.S. need new pits right now, in this decade? I think all experts will agree it does not. Will the U.S. need new pits in the 2030s? Again the answer is no. The reasons given for the current rush are always ideological, not scientific. But the real reasons are primarily political. Spending money in key places is a goal, not something to be avoided.
"Right now, it is not the Russians or the Chinese or our allies which we 'need' to impress with our pit production. It's ourselves, our self-image as the leader of the 'free world.' The reality that the U.S. cannot make all-new nuclear weapons is too jarring to our delicate political elites, even though we have thousands of warheads and bombs that will last decades to come unless we can muster the wisdom to disarm.
"Absent a 'need' to MIRV Sentinel, LANL pit production just lucrative busywork for the warhead complex, especially Livermore. To what extent is the W93 warhead -- the first pit type to be made at SRPPF -- just busywork as well, and a favor to the UK? Apparently, NNSA will now be using recycled pits for portions of both the W87-1 and W93 programs. [See Multi-year delays in plutonium "pit" production at Los Alamos now require the use of recycled pits for some new warheads The purpose of the Los Alamos facility, expected to cost more than $22 billion when all costs are tallied, was to avoid recycling pits, Jul 2, 2025.] Both these warhead programs could be canceled without diminishing the 'nuclear deterrent' in any way. If they do go forward, why not use recycled pits for both of them? This argument applies to LANL a fortiori, since the LANL factory is inherently temporary.
"The raison d'etre for quantity pit production at LANL, the factor which outweighed all the negatives, was the speed with which it could be started there. That's gone. Reliable production at LANL still 7 years away, according to NNSA. As a result, it's long past time for NNSA to look for an exit ramp at LANL.
"Why are pit costs rising, and pit schedules dilating? There are many reasons:
- Some of the original estimates were low-balled. The Pit Production Engineering Assessment (EA) assumed that 'nominal' pit production at LANL, which was to be established in 8 years from 2018, was somehow equivalent to, or could be turned into, reliable, enduring production at zero additional cost. The EA also assumed that the floor area needed for production could be cut in half from NNSA's best estimates of the year before. There was no explanation for either of these assumptions.
- 'Scope creep.' The production capacity sought at each of the two main sites has increased to at least twice the capacity commonly portrayed. NNSA is pushing for at least 80 pits per year at LANL, and the design capacity of SRS is highly likely to be in the range of 125 pits per year, on average. The four-year delay at LANL that was announced in early 2023 appears to have sent shock waves through the nuclear weapons community, reinforcing the need for greater capacity at SRS. Scope creep has been negatively synergistic with other delay factors.
- The degree of competition for scarce design talent and scarce equipment that NNSA has experienced was not foreseen. This creates upward cost pressure as well as schedule delays.
-
Covid was not foreseen. Various construction issues at both sites were not adequately foreseen. The scarcity of labor at LANL in particular was not foreseen. Construction costs have escalated faster than inflation generally.
-
Management at both sites and at NNSA headquarters has not always been strong. Across the world, most very large construction projects encounter correspondingly-large problems.
-
In short, NNSA bit off much more than it thought in 2018 -- and then decided to bite off even more. To this day, NNSA is not being realistic.
-
This happened because Congress and the White House have not made NNSA accountable, under any president. Up to now, increasing costs are simply not seen as a problem by any significant stakeholder in the policy process, whether it be congresspersons, committees, contractors, NNSA, DoD, the military, or the White House. Many of these stakeholders see increased costs as beneficial. Spending large amounts of federal money is what they want.
"We believe the scope of SRPPF now includes the sum of the original minimum single-shift production capacities of both sites, i.e. ≥30 ppy (~41 ppy average) at LANL and ≥50 ppy (~84 ppy average) at SRS (p. 13 in the AoA), or ~ 125-140 ppy average, single-shift. (For 140, see testimony of Dr. Charles Verdon, May 19, 2021) See also SRS EIS, p. S-27: SRS has NEPA coverage for 125 ppy. That is, SRPPF is now the Modern Pit Facility (MPF) reborn.
"Of note, this increased cost and scope happened during the Biden Administration, sometime before April of 2024.
"Trump needs Lindsay Graham and Tim Scott 'on-side', as well as Joe Wilson. So without question SRPPF will go forward for this reason alone until 2029. In addition, construction and operation of SRPPF is about to mandated by statute, this year.
"All this is the 'icing on the cake' for SRPPF, because even though SRPPF is expensive, from the perspective of the Nuclear Weapons Council it is necessary. For them, it is the best and indeed the only way to acquire enduring pit production capability both in terms of schedule and cost -- even if it costs $25 B. For them, that's cheap. It's not just the 'least dirty shirt in the laundry basket,' it's the only shirt. NNSA will never in a million years depend solely on the aging, overtaxed PF-4 for pits in the 2040s and afterward.
"NNSA's pit program needs a hard, realistic, non-partisan review, by journalists and by analysts inside government and out. We can only hope that the reviews underway within government will shed useful light and be supplemented by independent expert review, and that the Administration and Congress will listen."
|
