Vol. 21 No. 48 December 6, 2001 Page 481
Documents Are a Mess; Key Issues Unresolved: Former OCRWM Head
ROCKVILLE, MD. The Yucca Mountain, Nev., radioactive waste repository program is a mess, concluded the former head of DOE´s Office of Civilian Radioactive Waste Management (OCRWM) in the first Bush administration. He has spent almost the last year reviewing DOE´s pre-liminary site suitability evaluation and supporting documentation.
A comprehensive, independent peer technical review of all aspects of DOE´s Total Systems Per-formance Assessment (TSPA) analyses is needed for there to be any confidence in the program, John Bartlett, now a private consultant with S. Cohen & Associates, McLean, Va., told the Nuclear Regula-tory Commission´s Advisory Committee on Nuclear Waste (ACNW) Nov. 29.
However, this review will be a daunting task because of the poor state of the documentation, which contains many gaps, inconsistencies among documents, contradictory models and analyses, generalities and lack of traceable technical sup-porting information.
Bartlett reviewed the program for Clark County, Nev., one of the counties adjacent to the planned repository. A county planning official, En-gelbrecht van Tiesenhausen, told ACNW that county officials had neither seen nor asked to see Bartett´s presentation before the ACNW meeting.
Key Documents Reviewed
The principle documents Bartlett reviewed were:
The documents were very difficult to work with and in some cases it was almost impossible to trace the relationships among models, assumptions and data, Bartlett said.
Overall, technical traceability and continuity´ was lacking, and seems to have been interpreted by DOE as information accounting,´ Bartlett said, adding that this makes review of the documents ex-tremely difficult.
Bartlett pointed out that the Nuclear Waste Technical Review Board also has been stressing the need for traceability for several years.
Bartlett explained in detail some of the prob-lems encountered just in trying to review the site recommendation documentation.
Substantive technical information concerned with TSPA efforts was limited and very difficult to trace within the documents. Most of the substantive information was in the AMRs and the PMRs. No-where did the documents clearly spell out the rela-tionships between models, assumptions and data.
Information on any given technical topic was not compete in any single document. The example Bartlett used was cladding degradation information, which is spread out among various documents without any cross-referencing.
Substantive Issues Also Show Problems
Along with the documentation inconsistencies, Bartlett identified a number of substantive issues in the technical analyses.
Many of the assumptions in the TSPA analyses are extreme and are not related to data or realism, he said. Also, many of the department´s assump-tions appear to be highly conservative or non-conservative, but DOE never describes the basis for its assertions that a particular assumption is conser-vative or not conservative.
Bartlett found major differences between TSPA-SR and S-TSPA methods and results, but the basis for these differences is never made clear in the documentation.
As an example, he said, S-TSPA assumed early canister failures. The TSPA-SR assumed such failures would not occur until 10,000 to 100,000 years after closure. These conflicting assumptions make it very difficult to follow the projections of dose rate over time.
TSPA Lacks Sound Basis
The documentation does not provide a sound foundation for S-TSPA methods and results, which are the basis for the PSSE findings.
Overall, the TSPA-SR and S-TSPA methods and results, individually and in comparison, give the impression that projects of performance are more an artifact of models and assumptions than they are a realistic assessment of expected repository perform-ance, he said.
Bartlett, like other analysts, was troubled by DOE´s decision not to use a specific repository de-sign as the basis for its analyses. Variations on the high-temperature design and peformance as the ba-sis for low-temperature repository performance as-sessments do not provide an appropriate or suffi-cient suite of TSPA results, he told ACNW.
Alloy 22 was another trouble point, he said. The S-TSPA results for the regulatory compliance period depend solely on Alloy 22 performance, for which the key data base is small and fragile, and the long-term performance is unknowable.
The use of Alloy 22 essentially makes DOE´s repository an underground engineered storage fa-cility, Bartlett said, referring to the need to be able to retrieve the waste if Alloy 22 does not perform as expected under repository conditions.
Weakest Link Is Forged of Alloy 22
Bartlett specifically criticized DOE´s use of one-off analyses of Alloy 22 sub-system per-formance factors, such as localized corrosion and stress-corrosion cracking.
Such analyses for Alloy 22 as a whole have not been performed, despite the fact that Alloy 22 controls overall system performance.
Although most repository performance factors are temperature dependent, DOE´s performance assessments find, without adequate documented substantiation, that overall system performance is virtually independent of temperature.
DOE´s TSPA results for high- and low-temperature repositories, which show no difference in system performance after about 10,000 years, imply that coupled effects during the period up to 10,000 years have no significant or persistent ef-fects, Bartlett said.
Neither Hot Nor Cold Site Fully Studied
DOE also has not evaluated in detail either hot or cold repositories, despite the fact that they pose different and unresolved performance issues, Bartlett added.
As a result of coupled effects, a hot repository could alter natural features in uncertain ways, he said, adding that this, in turn, could have a negative impact on Alloy 22 performance.
A cold repository may require a significantly increased repository footprint, which in turn may require additional site characterization and revision of the expected contribution of natural features to repository performance.
Furthermore, Bartlett pointed out that DOE could only expand the repository to the north, the direction of the highest gradient.
Given that the program technologies are on the cutting edge of knowledge and understanding, Bartlett stressed that independent peer review is the only way for DOE to have confidence in the scien-tific basis for the site suitability evaluation. Such peer review is particularly needed for the supple-mental TSPA models and results.
Contact: John Bartlett, SC&A Inc., 6858 Old Dominion Dr., Suite 301, McLean, VA 22101.
Reprinted with permission of Nuclear Waste News.
©2001 Business Publishers, Inc. All rights reserved.