We appreciate the opportunity to meet with you today to hear the views of the Commission Staff and others, as well as present you with our summary view of portions of the Department of Energy's Viability Assessment of a Repository at Yucca Mountain (VA). As you will note, our presentation is intended to be cast both in the context of the VA, and the Commission's role in the pre-licensing period and as the repository regulator.
In his February 8, 1999 presentation to you, Lake Barrett, Acting Director of the Office of Civilian Radioactive Waste Management, pointed out that "While the Viability Assessment is not one of the decision points defined in the Nuclear Waste Policy Act, its completion is significant because it gives policy makers key information regarding the prospects for geologic disposal at Yucca Mountain." Since the Commission has decided to review technical aspects of the VA, it too can contribute key information to policy makers regarding the prospects for geologic disposal at Yucca Mountain. This objective is somewhat different from the Staff's early intent in suggesting such a review, which was to seek to assure that there will be "no surprises" from the Commission for DOE at the time of site recommendation, if the Secretary finds the site suitable. The Nuclear Waste Policy Act requires that at site recommendation, the Commission provide its "preliminary comments...concerning the extent to which the at-depth site characterization analysis and the waste form proposal for such site seem to be sufficient for inclusion in any application to be submitted by the Secretary for licensing of such site as a repository."
There are already significantly contrasting views about the message policy makers can draw from the VA regarding the prospects for geologic disposal at Yucca Mountain. On one hand, the VA states (and Lake Barrett told you on February 8) that "based on the viability assessment, DOE believes that Yucca Mountain remains a promising site for a geologic repository and that work should proceed to support a decision in 2001 on whether to recommend the site to the President for development as a repository... Uncertainties remain about key natural processes, the preliminary design, and how the site and design would interact." Mr. Barrett also informed you that while the VA reveals no "show stoppers", it does identify areas where additional work is necessary before site suitability can be determined.
On the other hand, the Total System Performance Assessment - Viability Assessment (TSPA-VA) Peer Review Panel, formed by the waste management system Management and Operating Contractor to evaluate and critique the TSPA-VA and make recommendations for its improvement for possible use in a license application, has taken a much less optimistic view. The Panel, in its February 11, 1999 Final Report, points out that Congress defined the objective of the TSPA-VA to be the assessment of the "probable behavior of the repository." The Panel's conclusion is that "it is unlikely that the TSPA-VA, taken as a whole, describes the long-term probable behavior of the proposed repository." The Panel goes on to say that "at the present time, an assessment of the future probable behavior of the proposed repository may be beyond the analytical capabilities of any scientific and engineering team. This is due to the complexity of the system and the nature of the data that now exist or that could be obtained within a reasonable time and cost."
The repository system's postclosure performance, as analyzed in the VA, relies on four key attributes identified in the DOE's Repository Safety Strategy:
According to the VA, these key attributes constitute a continuum that describes the Yucca Mountain repository concept: First, only a small fraction of the precipitation that infiltrates Yucca Mountain will appear in the waste emplacement drifts as seeps, and only a portion of the waste containers will be contacted by drips that can lead to corrosion failure.
Second, in the repository environment, the waste packages will fail at varying rates, based on the availability of moisture contact, but on the whole, the packages will be resistant to corrosion failure for a very long period of time, with only a few corrosion failures predicted in the first 10,000 years after emplacement.
Third, once a waste package fails, radionuclide release will be limited by the resistance of most of the fuel cladding to corrosion, and the low solubility of most radionuclides in the contacting water.
And fourth, after radionuclides are released from the waste package, there may be some sorption and matrix diffusion that reduces their concentrations in the groundwater, as well as dilution during transport down gradient that would serve to limit calculated doses to individuals at the accessible environment boundary.
Projected repository performance, i.e., individual dose at the accessible environment boundary, relies on each of these attributes contributing its expected share to the combined natural and engineered barrier system. The failure of any one component to function as well as predicted will have an adverse effect on total system performance. This is confirmed by an analysis reported by the Yucca Mountain Project to the Technical Review Board in a January 25, 1999 meeting. The analysis was designed to illustrate the relative contribution of the repository system barriers by neutralizing one barrier at a time in successive runs of the total system performance model for the initial 10,000 postclosure years. The result indicated that during this period, the waste package is responsible for over 99% of the expected repository performance, and if it were eliminated from the system, the expected individual dose rate at the accessible environment would be about 1 rem per year within about 2,000 years after closure. In contrast, if the sum of all natural barriers' contributions to performance during the same period were neutralized, and the waste packages were the only barrier, the expected dose rate would be only about 1 millirem per year. Of course, the relative importance of barriers changes through time as the waste packages fail, but it is important to note that the expected peak dose in the base case calculation, with all barriers functioning as expected, is on the order of about 1 rem per year, occurring at about 200,000 years after closure when most of the waste packages have failed.
The result of this analysis is significant for a few reasons. First, it indicates that the proposed repository system does not exhibit "defense in depth", as stated by the VA to be "the property of a system of multiple barriers that are diverse, independent, and redundant such that failure by any single barrier will not result in failure of the entire system." While the engineered barrier may be planned to illustrate defense in depth through dual waste package layers and possible drip shields and backfill, the repository system as a whole does not meet the VA's description of defense in depth. The engineered barrier does not function as an independent means of limiting individual dose. If it functions as expected, the waste package only serves to delay the time of peak dose that the natural barriers would permit with or without the engineered barrier. And, we do not believe that under any regulatory circumstance, an expected individual dose rate of 1 rem per year to members of the public should be considered acceptable. In this belief, we are mindful of the Commission's often stated principle that future generations should not be subjected to radiation doses from a repository any greater than those considered acceptable to the current generation from other sources.
The nearly complete reliance of the Yucca Mountain TSPA-VA on the waste package (and other possible engineered barrier contributors) is a contradiction of the geologic disposal concept as described in the DOE's 1980 Final Environmental Impact Statement (EIS) Management of Commercially Generated Radioactive Waste. The EIS states, "Geologic barriers are expected to provide isolation of the waste for at least 10,000 years after the waste is emplaced in a repository and probably will provide isolation for millennia thereafter. Engineered barriers are those designed to assure total containment of the waste within the disposal package during the initial period during which most of the intermediate-lived fission products decay. This time period might be as long as 1,000 years..."
Each of the key attributes of the Repository Safety Strategy is subject to broad uncertainty, as exhibited in the VA. The uncertainty in the waste package lifetime is said to be about three orders of magnitude. And, the TSPA-VA shows an uncertainty range in dose rate projections in the 10,000 year calculation of about 4 orders of magnitude, with the 1 million year period at about 6 orders of magnitude. The question is: Can these uncertainties be reduced significantly? The TSPA-VA Peer Review Panel appears to think that the answer is No, at least in the near future and at a reasonable cost relative to the DOE's schedule and resources, and the answer may be Never. Primary contributors to this conclusion, with which we agree, are the complexity of the Yucca Mountain site and geologic setting, and the long time period involved in performance projection.
An interesting example of irreducible uncertainty involves an assumption about climate change in the TSPA-VA model. A relatively small shift in the projected periodicity of the short-term superpluvial climate condition can result in the calculated individual peak dose rate being not 1 rem per year, but 5 rems per year.
Another even simpler example involves the model assumption of one juvenile failure of a waste package 1,000 years after emplacement. Any assumption that is in excess of this one juvenile failure will result in a proportional increase in the expected individual dose rate. There is no scientific basis for the TSPA-VA model assumption of only one such failure, but within reasonable bounds, there is no basis for this or any other assumption, therefore the uncertainty remains irreducible.
While the DOE has said that the VA has been written independent of regulatory considerations, it must be recognized that the results of the TSPA-VA are being evaluated within a context of regulatory and safety standards whether specific standards for a Yucca Mountain repository exist or not. We have said earlier that an expected individual annual dose rate from a Yucca Mountain repository of 1 rem per year is unacceptable. And, since the primary release path from the repository is into currently potable groundwater, it also is unacceptable that expected doses to the public resulting from the repository contaminating this drinking water supply would exceed existing national standards. This level of dose rate from groundwater could be reached rapidly if the VA performance assessment model assumed 10 to 20 juvenile failures out of the thousands of waste packages rather than only one, and if the waste package failures occur earlier than the arbitrarily set 1,000 years after emplacement in the repository.
Much of the technical presentation in the Viability Assessment is based on data, analyses, and codes that do not meet the Commission's Quality Assurance requirements for licensing. The DOE is now engaged in an intensive program to repair these shortcomings that have been observable in the program since its beginning in 1983. This repair effort cannot be completely successful. It is clear that some of the information in the VA and its sources will not be properly qualified for use in a license application, although it may be needed to meet a completeness standard and to not further expand the already broad ranges of uncertainty in the performance assessment. The ultimate consequence of this shortcoming is that it detracts from the credibility of any claims of safety that may be made about a Yucca Mountain repository system.
The TSPA-VA reveals the expectation of very rapid groundwater flow from the repository location to the boundary of the accessible environment, assumed in the model to be 20 km (12 miles) from the edge of the repository. It is clear from the model realizations published in the VA that highly soluble radionuclides released from the repository can arrive at the 20 km boundary in as little as 500 years after release. This indicates that groundwater travel time from the undisturbed Yucca Mountain site to the accessible environment is thought to be as rapid as 500 years by the DOE. The median and mean values for the model realizations are slightly below and above 1,000 years respectively.
The matter of rapid groundwater travel time from the repository location to an assumed distant boundary of the accessible environment, as shown by the TSPA-VA, raises two regulatory issues, one for the Commission and one for DOE. For the Commission, the groundwater travel time that can be inferred from the TSPA-VA model realizations is in conflict with the Commission's 10 CFR Part 60.113(a)(2), which states:
There has been no groundwater travel time other than the 1,000 year criterion approved or specified by the Commission. Therefore, notwithstanding the fact that the Commission currently is seeking to eliminate this requirement through its proposed new Yucca Mountain specific licensing rule, 10 CFR Part 63, the Commission should inform DOE immediately that the Yucca Mountain site does not conform to its existing repository licensing requirements. The Commission should state that this information is based upon its review of the Viability Assessment.
Likewise, the Secretary of Energy should disqualify the Yucca Mountain site from consideration for development of a repository because it meets the groundwater travel time disqualifying condition, 10 CFR Part 960.4-2-1(d) of its site recommendation guidelines, in which the Commission has concurred as a statutory requirement. The disqualifying condition states:
Because the travel time has been inferred from realizations of the DOE's Yucca Mountain performance model, with numerous realizations indicating travel times less than 1,000 years, the tests of the regulatory language "fastest", "likely", and "significant" all have been met.
The Commission should inform the Secretary of its finding immediately because the Secretary's duty under the law is to act in a timely manner. The Nuclear Waste Policy Act as amended requires "If the Secretary at any time determines the Yucca Mountain site to be unsuitable for development as a repository, the Secretary shall - (A) terminate all site characterization activities at such site..." (emphasis added)(Sec. 113(c)(3).
The Viability Assessment reveals a number of important factors regarding the potential safety of a Yucca Mountain high-level nuclear waste repository system. These include:
The Viability Assessment suggests a number of issues for the Commission's consideration during its review. These include:
Thank you for the opportunity to present our views of the DOE' Viability Assessment and associated regulatory issues.