The USA is out of step with the rest of the world in seeking to cut off its repository compliance period before peak dose and release rates are reached, British environmental scientist Mike Thorne told the National Academy of Sciences (NAS) Board on Radioactive Waste Management (BRWM) on 20 September 2004 in Washington, DC.
Thorne has served as a consultant to AEA Technologies, RWE Nukem, Sweden’s SKB, and the UK’s BNFL and Nirex among others, and now has been retained by the US state of Nevada to compare proposed standards for a possible repository at Yucca Mountain, Nevada, to standards used around the world.
In late July, a three-judge panel of the Federal Appeals Court for the District of Columbia found that the 10,000-year compliance period that forms the basis for the Yucca Mountain licensing standard violates the 1992 Energy Policy Act because it was not “based upon and consistent with” the findings contained in the 1995 NAS report, Technical Bases for Yucca Mountain Standards. On 8 October, the court denied a motion from the Nuclear Energy Institute (NEI), the US nuclear industry trade association, to the keep the standard in place until the US Supreme Court decides whether to hear NEI’s appeal of the federal court ruling.
Given the court ruling and high-probability that the US Environmental Protection Agency (EPA) will have to go back to the drawing board and rewrite a standard for Yucca Mountain, BRWM is again reviewing the US repository programme and the proposed repository at Yucca Mountain. In its 20 September information-gathering meeting, the board heard widely diverging views on how the programme arrived at its present impasse, and possible future courses of action – including abandoning Yucca Mountain and rethinking the programme. A number of key players – most notably the Department of Energy (DoE) which is responsible for the repository programme – declined invitations to appear at the meeting.
Some industry representatives suggested Congress should step in and override the court decision by passing legislation reinstating the EPA standard. However, other presenters, including a long-time staff member from a key congressional committee, warned that this would further erode the credibility of the repository programme, arguing that it would be seen as a congressionally forced ‘dumbing down’ of the programme because Yucca Mountain could not meet a scientifically rigorous standard.
SUFFICIENTLY LONG TIMESCALE
Countries that plan to use geological disposal aim to isolate radioactive waste from the accessible environment over timescales that are “sufficiently long to ensure very substantial reductions in their radioactivity,” Thorne said. However, because it is generally accepted that, at some future time, radionuclides may be released from the wastes and transported to the biosphere, these countries have set limits or targets on individual dose or risk as the criteria for evaluating the safety of proposed facilities.
Although specific regulatory requirements differ among countries, the Nuclear Energy Agency (NEA) of the Organisation for Economic Cooperation and Development (OECD) summarised the international consensus on these standards in its report, The Handling of Timescales in Assessing Post-closure Safety of Deep Geological Repositories, Thorne said.
NEA’s conclusions were similar to those reached by NAS in 1995: no ethical arguments justify imposing a definite time limit to the period addressed by these safety standards.
The NEA Timescales report also provided details on regulations in a number of countries that require dose or risk calculations to be carried out at least until the time that safety indicators attain their maximum values, regardless of when this occurs. Swiss regulations, for example, state that doses and risks shall at no time exceed specified values, Thorne said. Although Yucca Mountain is a younger geologic formation than those being considered in Europe and Canada, geologists see no reason why its general geological stability cannot be accessed over the next million years or more, he added.
In fact, Thorne pointed out that since the 1995 NAS report, DoE has conducted simulations of radionuclide release and groundwater transport out to one million years after closure of the repository. These simulations have covered a sufficiently long time period to demonstrate that “calculations through to the time of peak risk can be reasonably undertaken.”
Thorne contrasted the US programme with Sweden’s. Like the USA, Sweden is looking at a highly engineered near-field environment. However, Sweden proposes to site its repository in the saturated zone, in geology that can provide a constant groundwater environment over the very long term. Sweden also has deliberately kept both the engineered and geological environments as simple as possible, he said. In contrast, Yucca Mountain has both a much more complex engineered barrier system and a much more complex geology, in that it is situated in the unsaturated zone. Characterising flow and transport is an order of magnitude more complex in the unsaturated zone than in the saturated zone, he said.
Massive trucks simulate seismic events during the study of Yucca Mountain
WHAT TYPE OF STANDARD?
Back in 1995, an NAS panel chaired by Robert Fri from the Washington DC-based think tank Resources for the Future, responded to a request from Congress to looked at possible health-based standards for Yucca Mountain, as well as at the possibility of predicting and/or preventing human intrusion into a potential repository.
Fri told the BRWM that the earlier panel recommended that the standard:
- Be risk-based, rather than dose-based, because the dose-response relationship changes over time.
- Focus on the critical group, the small group of people that contains the individual at highest risk, while avoiding overly conservative assumptions.
- Protect the highest risk individual at the time of greatest risk, which would be sometime after the canisters begin to fail in large numbers around a million years after closure.
- Did not need to include a population standard because the individual standard would protect the population as a whole.
On the question of human intrusion, the panel concluded that there was no way to prevent intrusion, and no scientific basis for predicting intrusion. Since the possibility of intrusion could neither be ruled out nor predicted, the panel recommended that the standard assume a single intrusion into the repository.
The 1995 panel also determined that an adequate scientific basis exists to analyse radionuclide plume transport over the very long term, “certainly well beyond 10,000 years,” Fri said. The geology can be assumed to be stable for at least a million years.
Exposure scenarios are what creates the problem with assessing compliance because these deal with human behaviour and how people will come in contact with the radionuclides.
The panel, therefore, recommended that EPA use long-term probabilistic models for both radionuclide transport and the critical group, relying on present-day knowledge and reasonable assumptions. It called for a long assessment period, at least through to the period of likely peak dose.
In its final standard, EPA did neither. It used a very short assessment period (10,000 years) and a deterministic, rather than probabilistic, exposure scenario, Fri said. This disregard for the earlier NAS recommendations was the crux of the court decision overturning the standard and sending EPA back to the drawing board.
Discussions among BRWM members also highlighted the difficulties in dealing with population change issues. One pointed out the variability of population over time, and the lack of a good historical database for the development of modern population dynamics. A thousand years ago, for example, Manhattan Island was largely unpopulated. A thousand years from now, Yucca Mountain could be surrounded by dense urban development. However, since the 1995 panel had no scientifically valid method of predicting population changes over time, it simply agreed to assume – for the purpose of the standard – that the population would remain the same as it is today. “Yucca Mountain was picked because there are not a lot of people around,” and we assumed it would stay that way, Fri said.
NEW OPTIONS OR NEW START
In the wake of this year’s court decision, EPA is looking at its options, Jeffrey Holmstead from EPA’s Office of Air and Radiation Programs told BRWM. He would not discuss what options were on the table, but told the board he was at the meeting mainly to learn what BRWM sees as EPA’s choices if the agency has to rewrite the standard. “We are dealing with a timeframe that is very different from anything else EPA has ever handled,” he said.
Holmstead also offered little in the way of schedules. Since the court did not set a specific date for EPA to produce a new standard, “we are responding as quickly as we can,” he said, but declined to say whether the effort would take months or years. He would only joke that it would take “less than 10,000 years.”
A congressional staff member voiced concern that EPA will just try to substitute one million years for 10,000 years in the standard and expect that to solve the problem. Like Fri, Sam Fowler, of the Senate Energy and Natural Resources Committee, pointed out that NAS had called for a risk-based standard. “Just changing the time could lead to an unworkable standard.” Fowler’s committee will have jurisdiction over any bill to override the court decision.
Now may be the time to start looking at what happens if there is no repository at Yucca Mountain, Fowler suggested, pointing to a number of ‘potentially fatal problems’ with the current repository programme.
As well as uncertainty over the licensing standard, the repository programme also faces major uncertainties in federal funding, Fowler said. As it stands now, the repository will get $131 million in fiscal year 2005. The DoE, however, says it needs $880 million to keep the programme on track.
Furthermore, congressional analysts and some staff members at the Office of Management and Budget are taking a hard look at the billions of dollars in potential damages that utilities are seeking through breach-of-contract lawsuits over the DoE’s failure to accept spent fuel by the specified time in 1998, Fowler said. As long as the programme exists, damages will continue to mount for every year that a repository is not in place. But if the programme were to be terminated, Congress could use the Nuclear Waste Fund to settle the lawsuits and pay damages, which the utilities could then use to pay for on-site storage.
BRWM ALSO TACKLES WASTE RECLASSIFICATION
Fowler also tied the uncertainties at Yucca Mountain to another issue BRWM is considering: congressional consideration of a DoE proposal to allow the department to reclassify some of the high-level waste in tanks at the Savannah River Site in South Carolina as ‘waste incidental to reprocessing’. This change in nomenclature would allow the DoE to leave some of this waste in place, and grout or otherwise stabilise the tanks rather than remove all the waste for vitrification or other treatment.
The DoE initially sought to use waste reclassification at three sites: the Hanford Site in the state of Washington, Idaho National Engineering and Environmental Laboratory, and the Savannah River Site in South Carolina. However, in July 2003, the Idaho Federal District Court ruled that the DoE’s reclassification effort violated the express directions of the Nuclear Waste Policy Act for the disposal of high-level radioactive waste (NRDC versus Abraham). With plans stopped by court order, the DoE simultaneously appealed the ruling to the US Court of Appeals for the Ninth Circuit and asked Congress for a legislative mandate to allow the reclassification.
On 7 September, Congress gave the DoE a partial victory. South Carolina Senator Lindsey Graham succeeded in attaching an amendment to the 995-page, $447 billion defence authorisation bill that would allow the DoE to reclassify sludges in 49 tanks at Savannah River. The provisions do not affect tank wastes at the Hanford or Idaho sites.
Graham argued that his amendment will cut almost 20 years off the cleanup schedule for Savannah River and will save taxpayers more than $16 billion.
Reclassification of DoE waste is a ‘slippery slope’, Fowler said, explaining that some members of Congress already are asking why, if high-level waste can be left in place indefinitely, cannot the same be done with civilian spent nuclear fuel?
Washington and Idaho want no part of such legislation. “Careless legislation could allow the DoE to abandon in place significant quantities of waste,” Suzanne Dahl from the Washington State Department of Ecology told BRWM. Dahl pointed out that the Hanford tanks are located 7-11 miles from the Columbia River, and groundwater flows past the tanks to the river. Two percent of the tank waste (1 million gallons) already has leaked into the soil, and 10 of the 12 single-shell tank farms already have impacted groundwater to the extent that groundwater concentrations of some radionuclides are 70 times drinking water standards at one farm.
Washington State’s position is that the Nuclear Waste Policy Act limits the DoE’s discretion to reclassify high-level waste and that the district court ruling in NRDC versus Abraham upheld this position, Washington assistant attorney general Joseph Shorin told BRWM. Washington State participated as a friend of the court in NRDC versus Abraham, he said. Nothing in the ruling halts the DoE’s cleanup efforts, it only prevents the DoE from unilaterally deciding to abandon waste in place, he concluded.
Dahl said that there are some tanks at Hanford that the DoE claims never contained high-level waste. If the DoE can demonstrate this, the state will work with the department to allow the waste to be removed from the tank, processed and shipped to the Waste Isolation Pilot Plant (WIPP) in New Mexico. However, Washington State wants a consensus agreement on this in place between DoE headquarters and New Mexico. “What we don’t want is 27,000 orphan drums sitting at Hanford,” that New Mexico refuses to accept, she said.
