UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA

I, G. Scot Carlson, hereby declare as follows this 20th day of March, 1997:

1.  I received my doctorate in Social Sciences from the University of California at Irvine in 1980, where I specialized in statistical and textual analysis of cultural aspects of technology.  I am currently self-employed as a research analyst.

2.  In my capacity as consultant to the Los Alamos Study Group, Santa Fe, NM, I have read the portions of the Final Programmatic Environmental Impact Statement for Stockpile Stewardship and Management (hereafter, the PEIS) that refer to the production of plutonium `pits' (ignition triggers for nuclear weapons) at Los Alamos National Laboratory.  I have examined the text and the data for accuracy and consistency.  As the PEIS is a large and complex document, I limit my discussion to selected examples of inaccuracy and inconsistency within and between the text and the tabulated data. 

3.  This paragraph summarizes the findings to be presented in paragraphs 4 through 9 of this affidavit.  Analysis of six passages in the PEIS demonstrates that the PEIS states that LANL is currently able to produce 50 pits per year (paragraph 5), at the same time that the document states that they would be unable to produce 80 pits per year without modification of their current facilities (paragraph 6).  The passages are confusing (e. g., compare paragraphs 6 and 7), but conclude that the `No Action' alternative allows LANL the production of 50 pits per year using existing equipment (pp. 3-4, 3-111), while the "base case" considers either 20 (paragraph 6) or 80 (paragraph 7) pits per year.  Many of the tables analyze only the "surge" case of 80 pits per year (e. g., A-122).  The action proposed by the PEIS is not the introduction of pit production at LANL(p. A-120).  Rather, the proposal is to modify the existing production facilities to allow the production of two types of pits on a single shift (p. 3-57).  This modification would enable the production of 80 pits per year with three-shift ("surge") operation (p. A-120).

4.  "Analysis is provided for facilities that would be sized to support estimated average annual manufacturing requirements resulting from the base case stockpile size [>3,500 weapons] assuming single shift operation, 5 days per week.  This PEIS analyzes environmental impacts of the base case quantitatively, including an evaluation of three-shift operation, 5 days per week (surge operation), to provide a bounding analysis... Existing facilities that would be downsized or modified have inherent differences in capacities when operated in the base case three shift surge mode.  For a given stockpile management mission, the downsize alternatives generally have a greater inherent capacity than other alternatives... [A] portion of the environmental impacts are due to the higher output associated with the three-shift surge mode of operation." (p. 3-3)

5.  The footnote to Table 3.1.1.2-1 explains, "Capability based capacity - the facility capacity (up to 50 per year) inherent with the facilities and equipment required to manufacture one component for any stockpile system." (p. 3-4)

6.  Later, the PEIS introduces the description of pit fabrication alternatives with this account, "For the base case analysis, workload requirements are assumed to be at a level necessary to maintain competence and to replace components destroyed during surveillance testing.  This base case production rate is approximately 20 pits per year.  In order to ensure that DOE is able to support the national security mission, equipment would be installed to provide the capability to fabricate one each of every pit type in the post-2005 stockpile.  This concept is called capability-based capacity. Operating this array of equipment 5 days a week, on a single shift, provides an annual capacity of approximately 50 pits of, at most, 2 different types. (p. 3-57).

7.  "Under the No Action alternative, ...LANL maintains a limited capability to fabricate plutonium components ... and performs surveillance operations on plutonium components returned from the stockpile... Under No Action, DOE would not have the capability to perform pit fabrication to meet ... the base case [*80 pits per year]." (pp. 3-57&8).

8.  On page 3-111, there appears Table 3.7.1-3, "Summary Comparison of Impacts for the Pit Fabrication Mission."  This table compares the 'No Action' alternative to "Reestablish at LANL."  Consulting footnote `a': "All data for operations are based on three shift except for No Action, which is based on one shift."

9.  "The facilities that are currently used by stockpile surveillance activities would be shared with the pit fabrication group until dedicated facilities become available.  The current stockpile Pit Rebuild Program at LANL would be absorbed within the pit fabrication effort as the activity is the same; only the number of pits produced would change.  The number of pits fabricated annually is projected to be from 20 to 50 (depending on equipment availability), but could be about 80 if surge mode (multiple shifts, personnel overtime, and use of equipment to full capacity) were exercised." (p. A-120)

10.  It is instructive to return to Table 3.7.1-3, "Summary Comparison of Impacts for the Pit Fabrication Mission" (p. 3-111) to demonstrate that the information conveyed by the table is internally inconsistent and cannot be relied on for analytical purposes.  This table compares the "No Action" alternative (single-shift operation) to the "Reestablish at LANL" alternative (three shifts), as noted in footnote `a' to the table.  Under No Action, the entire LANL apparently would disturb no land, use no water, employ no one, suffer no accidents, and produce no waste.  Clearly, the zeroes in the column could only refer to the change of value from the current capability based capacity of 50 pits per year.  However, the figures listed in the column under "Reestablish at Los Alamos" represent absolute values, not the change from No Action (except in those cases specifically noted in the table). 

11.  The information reported in the tables may or may not agree with the figures in the text or with other tables.  Continuing consideration of Table 3.7.1-3 of the preceding paragraph, water use is claimed to be 0 under No Action and 5,790 million liters per year (MLY) under "Reestablish at Los Alamos."  On page 4-287 is presented the examination of groundwater usage.  "During operation, an additional 30.2 MLY of water would be required to support pit fabrication activities, which is a 0.52-percent increase over the No Action groundwater withdrawal of 5,760 MLY."  However, these figures are correctly conveyed in Table 4.6.3.4-1 (p. 4-284).  This is but a single example among many occurrences of disagreements between and within the text and the tables.

12.  Another example of differences between texts and tables concerns the scope of facilities that are, or will be, used in pit fabrication at LANL.  The facilities are discussed twice in the text (pp. 3-58, A-117, A-120) and once in a table (p. A-120).  Table A.3.3.1-1 (p. A-120) purports to list "Los Alamos Pit Fabrication Facility Data."  It mentions 6 buildings in two Technical Areas (TA-55 and TA-3).  Omitted from the table, but included in the text on pages A-117 and A-120, are these Technical Areas: 8, 35, 50, and 54.  Included in the table, but omitted from the text, is the Nuclear Material Storage Facility, PF-41, which is misidentified as PF-4.  The discussion of construction plans (p. 3-58) mentions Technical Areas 3, 8, 50, 54, and 55, but omits TA-35 and PF-41. 

13. The discussion of construction plans states, "Existing equipment would be retained as much as possible, but some equipment would be upgraded to production quality." (p. 3-58)  Table 3.4.3.2-1 (p. 3-61) indicates that "Pit Fabrication Facility Construction Requirements" would be "Minimal" in all resource categories.  The impression given is one of modest upgrade, easily accomplished.  Table 3.4.3.2-3 (p. 3-62) lists the construction waste volumes generated each year of the three- year construction period, measured in cubic meters.  Footnotes `a' and `b' to the table mention a conversion factor of 1500 kg/m3, and state the actual amount of waste to be produced.  The expected radioactive waste totals 90 tons of gloveboxes, ventilation piping and ductwork, of which 30 tons would be transuranic waste, and 60 tons would be low-level waste.  Also, there would be 153 m3 of hazardous waste.

14.  Whether or not the text and the tables described in paragraph 14 agree with each other is a matter of judgement.  In my opinion, 90 tons is more than "some" equipment, and a three-year construction project requires more than "minimal" resources.  The point here is to show that tables and text in the PEIS can disagree qualitatively such that the reader might reasonably question the representation in the textual discussion of the tabulated data.

I declare under penalty of perjury that the foregoing is true and correct.  Executed this 21st day of March, 1997 at Santa Fe, New Mexico.
______________________

G. Scot Carlson, Ph.D.