June 24, 2021 SASC/SF hearing re NNSA, notable error and correction in Sen. Reed statement re pit production cost

June 28, 2021

Dear _____________, others --

We here finally had a chance to watch the June 24 SASC/SF hearing re NNSA programs and budget, which had several interesting bits.

One of those was the exchange between SASC Chairman Reed and Acting NNSA Administrator Verdon, at ~19:00 in the posted video.

Chairman Reed referred to a 2018 engineering study, presumably the 2018 Engineering Analysis, and stated that the cost of modular facility next door to LANL's main plutonium facility was about $5 billion, for "25" square feet.

Unsure of his meaning, I checked and sure enough, two of the three LANL options in the EA (Options 2a and 2c) are close to 25 thousand square feet in size and close to $5 billion (in what Verdon called a "top-down" estimated cost). Cost estimates in the EA were rough order-of-magnitude (ROM) estimates ("Class 5" estimates in the DOE classification), to be used for comparisons only, and not to be used for budgeting purposes. The actual cost of a project could be twice as much as a Class 5 estimate as we see happening in the case of EA Option 1 (SRPPF). (It should also be noted that the scope of the Option 1 project has increased since the EA, as Dr. Verdon noted. And as most of you have heard me say in years past, the EA made various questionable optimistic assumptions for all its options.)

Chairman Reed said that LANL modular option would cost $20,000 to $25,000 per square foot. That was very wrong and it seems important to point this out.

In round numbers, a $5 billion project that provides 25,000 sq. ft. will cost $200,000 per sq. ft. The three LANL options in the EA were estimated to cost $201,000/sq. ft., $238,000/sq. ft., and $248,000/sq. ft., for options 2a, 2b, and 2c respectively.

So Chairman Reed cited LANL costs that were a factor of ten less than the EA actually said in their rough estimate.

Option 2c (underground modules) was highly disfavored in the EA analysis, and rightly so. As you know the GAO savaged this option.

Regarding the other two LANL options, my professional view is similar to NNSA's 2017 Analysis of Alternatives (AoA), namely that none of these LANL options are actually feasible from the engineering perspective at that location relative to an operating plutonium facility and its security perimeter, on a narrow, congested mesa, not even considering the other construction disadvantages of the LANL site. None of them solve the problem of dependence on LANL's aging plutonium facility. The AoA concluded that if LANL was to be the sole pit production site, an unspecified greenfield location at LANL would be necessary. Where that location might be found at LANL is a question the AoA did not address. The AoA, I should add, omitted any mention of the geotechnical issues at TA-55 which played such havoc with CMRR-NF design and costs.

It is odd, isn't it, for so many senators and representatives to be concerned about aging facilities, on the one hand, and yet some of them appear ready to rely on the 43-year old LANL plutonium facility, located just 3,000 feet from residences -- the same separation as in the case of Livermore's plutonium facility -- designed for R&D purposes in another safety era with essentially no knowledge of local seismic conditions?

I should add that the estimated cost of space in the CMRR-NF was $152,000/sq. ft. in 2011 dollars (p. 2), using the publicly-available CMRR-NF cost of $5.86 billion. By that time OMB -- and, we heard, NNSA -- was estimating that the cost of CMRR-NF would eventually rise to more than $10 billion. So a Class 5 estimate for a similar building in the same place, at $200,000 to $250,000 per square foot in 2018 dollars does not appear unreasonable.

As noted here,

SRPPF’s 400,000 sq. ft. of useable space (p. 79) is much greater than the 110,000 sq. ft. of useable space needed for the ≥80 ppy mission, or 130,000 sq. ft. including essential building services (p. 76), leaving plenty of space in SRPPF for analytical chemistry and other supporting functions. If SRPPF is completed there would be no need for quantity production at PF-4, releasing that facility for R&D missions, as NNSA proposed in 2017 (p.2).

Applying an estimated cost of $11.1 billion including several supporting facilities, security perimeter, etc., SRPPF’s 400,000 sq. ft. of useable space comes to $27,750 per square foot. As Dr. Verdon said, it's complicated, but this is a factor of 8 less than the comparable capital cost at LANL.

On a per pit basis the cost differential will also favor SRS, as far as we can tell. But by how much? We don't know.

Unfortunately redacted CD-1 documents for LANL (and soon SRS) are not available for independent analysis, as yet. In the interest of good government, the administration could fix that.

Please feel free to contact me if you have any questions.

Best wishes,

Greg