The second session of the workshop focused on approaches for assessing long-term performance of site remedies. Presentations on this subject were provided by a stakeholder and a government regulator:
- Willie Preacher, from the Tribal Department of Energy (DOE) Program and a member of the Shoshone-Bannock Tribes, and
- William Reckley, branch chief in the Japan Lessons Learned Project Directorate, Office of New Reactors, U.S. Nuclear Regulatory Commission respectively.
Federal and state case studies were provided by:
- Richard G. Mach, Jr., director of environmental compliance and restoration policy in the Office of the Deputy Assistant Secretary of the Navy, and
- Roger Petrie, Federal Facility Agreement (FFA) projects manager for the DOE Oversight Division of the Tennessee Department of Environment and Conservation.
The session was moderated by Patricia Culligan, professor, civil engineering and engineering mechanics, Columbia University and planning committee member.
This chapter provides summaries of the key points made by each of these individuals and by participants in the subsequent discussion sessions. These statements reflect the viewpoints of the individual speakers, not the
consensus views of the workshop participants or of the National Academy of Sciences.
3.1 LONG-TERM LAND USE AND OPPORTUNITIES TO AFFECT DECISIONS
Mr. Preacher worked at what is now the Idaho National Laboratory starting in 1973. He has been the interface between the tribes and DOE for many years. In 2002, he became director and spokesman for seven tribes. In government-to-government discussions (tribes have sovereign nation status), Mr. Preacher cannot speak for the tribes but he can provide a communication link between the federal government and the tribes. He is involved with state groups, DOE’s Environmental Management Advisory Board (EMAB), National Transportation Stakeholders Forum, and the Citizen Advisory Board (CAB) at Idaho. The main priorities for the tribes on environmental management issues can be seen in Figure 3-1.
FIGURE 3-1 Priorities of the tribes on environmental remediation and long-term stewardship (LTS) of contaminated sites.
SOURCE: Preacher 2014.
Order 144.1, Principle IV: Policy Principle for Cultural and Natural Resource Protection
Department-wide compliance with applicable cultural resource protection and other laws and Executive Orders will assist in the preservation and protection of historic and cultural sites including traditional religious practices, and traditional cultural properties and places (DOE 2009).
The American Indian Tribal Policy, Order 144.1 (DOE 2009), defines how DOE headquarters and field offices should interact with the tribes. Early and frequent communication is important—particularly when funding issues arise. All tribes understand the funding issue, but early communication of the issues will help with future decision making.
The protection of cultural and natural resources is a priority for the tribes (Principle IV within Order 144.1, see Box 3.1). There can be misunderstandings between the tribes and DOE about what is considered a cultural resource. For example, Idaho National Laboratory (INL) was initially a Naval Proving Ground (see Chapter 1 in this volume). Tribes had used the lands as a cultural resource for hunting and gathering but have been denied access since the proving grounds were established. Mr. Preacher represents a set of tribes that has a good working relationship with the DOE INL field office. The field office keeps the tribes informed of DOE Headquarters’ upcoming changes.
Long-term performance of cleanup remedies is an important issue for the tribes. Independent monitoring and surveillance programs have been useful in establishing trust between the tribes and DOE. Almost every tribe has some form of monitoring and surveillance program of the sites. These programs collect samples in parallel to DOE monitoring programs. Analysis is performed at independent laboratories, and the tribes share the results with DOE. This verification of environmental sampling is also important to establishing trust between the tribes and DOE. Unfortunately, federal budget sequestration eliminated funding of the INL tribe monitoring program, even though DOE had agreed to continued-funding of this effort.
Another issue relates to the “trust responsibility” of the sites. Trust responsibility requires the U.S. government to uphold treaty rights, lands,
assets, and resources of tribal nations.1 The tribes would like resolution of the issues of trust responsibility and land use for cultural purposes.
The tribes would also like verification that different types of contamination from new reactor designs under development (e.g., advanced reactor designs) are being considered and studied. Questions exist about the types of contaminants that these new reactors will produce after they are built, operated, and eventually decommissioned. For example, areas across INL contain buried, entombed, and decommissioned reactors.2 Reactors have been encased in concrete and buried at the Idaho CERCLA Disposal Facility (ICDF). The tribes question the soundness of this practice, especially for future generations.
Groundwater contamination is a difficult issue for the tribes to understand. In the past DOE has focused on cleaning up the groundwater, but now it appears as though cleanup may never be complete. The tribes recognize that cleanup costs a great deal. High costs and hazards to populations are the usual reasons given for incomplete cleanup of groundwater. Groundwater plumes migrate and may encroach onto tribal lands. Wells have been drilled to monitor plume migration and, as was described earlier, tribal monitoring programs have been set up to independently measure contamination.
The tribes would like early communication and involvement in land transfer of the legacy sites. They would like first consideration for any lands to be transferred from the U.S. government, as supported by treaty rights. The tribes would like all above- and belowground contamination to be addressed prior to land transfer. This could mean, for example, cleanup to background levels for remediation of radioactive contamination. Radioactive contamination and radioactivity are also difficult concepts for the tribes to understand. However, the continuation of monitoring programs could help to educate the tribes about these concepts and could potentially help with communication.
Concerns exist about sensitive, archeological sites that were never recorded by the tribes and have not yet been discovered. The tribes expect DOE to respect the American Indian Tribal Policy (DOE Order 144.1, DOE 2009) when these sites are discovered. For example, before DOE Order 144.1 was enacted (circa 1973), INL employees would collect arrow heads
1 Federal Indian trust responsibility is explained on the Bureau of Indian Affairs site: “The federal Indian trust responsibility is a legally enforceable fiduciary obligation on the part of the United States to protect tribal treaty rights, lands, assets, and resources, as well as a duty to carry out the mandates of federal law with respect to American Indian and Alaska Native tribes and villages. In several cases discussing the trust responsibility, the Supreme Court has used language suggesting that it entails legal duties, moral obligations, and the fulfillment of understandings and expectations that have arisen over the entire course of the relationship between the United States and the federally recognized tribes” (http://www.bia.gov/FAQs/).
2 For example, the Boiling Water Reactor Experiment (BORAX)-1 is buried.
and other artifacts found on the site. Through Order-mandated training, DOE employees and contractors were directed to not pick up artifacts. The training has resulted in a noticeable shift in practices and an impact on employee actions.
Figure 3-1 summarizes the main priorities of the tribes. Early communication and involvement on decisions and land transfers are key to a good relationship between DOE and the tribes, which was reaffirmed by the recent Blue Ribbon Commission report on America’s Nuclear Future.3 The tribes are concerned about long-term stewardship and the need to transfer information and history about sites to future generations (e.g., information about buried waste and reactors). Trust responsibility and the recognition of treaty rights are also important. Finally, with every Administration change, the tribes have found that re-education of the new set of DOE federal employees is needed (e.g., an “Indian 101” overview). The tribes meet yearly to discuss these and other issues.
3.2 SUMMARY OF DISCUSSION SESSION
Monitoring Program. Session moderator and planning committee member Patricia Culligan (Columbia University) asked whether the tribes have monitoring and surveillance programs outside of the ones that are conducted with DOE. Mr. Preacher responded that the Shoshone-Bannock tribe collects and measures water samples, soil, plants, and animals (live and road kill) at the borders of the site and throughout the reservation. Recently, the tribe found trace amounts of radiation from the Fukushima accident on stagnant pond samples. The sampling program has temporarily stopped but is expected to restart soon. Idaho’s Department of Environmental Quality also has a sampling program. The tribes and state monitoring programs share their independent reports with DOE.
Kevin Crowley (National Academy of Sciences staff) noted the importance that the tribes place on sites being contamination-free above and below ground. He wondered whether there have been any discussions with EM about the tribes acting as long-term stewards of sites for DOE, which could be a valuable arrangement. Tribes are multi-generational and have a strong interest in protecting the lands. Mr. Preacher responded that, when former DOE Secretary Bill Richardson visited INL, the tribes asked if an area north of the site, the Sagebrush Steppe Reserve, could be given back to the tribes. Secretary Richardson’s response that the land would be trans-
ferred to the Bureau of Land Management (BLM),4 left the tribes with a “sour note” for future DOE land transfers. Tribes live close to the borders of many sites and could potentially benefit from land transfers. For the Shoshone, INL is approximately 35 miles away. The tribe’s main objective is to gain access to the site to hunt and fish, but there are concerns about contamination, which leads back to the sampling programs. Animals near and on the site have been tested. Rock chicks sampled near and on the onsite disposal facilities contain radioactivity, but road kill has not been contaminated.
Dr. Crowley posed the same question to Bill Levitan (EM). Mr. Levitan responded that DOE has considered this approach. He could best speak about the Hanford site. Currently DOE and the tribes are discussing access rights for traditional uses by the tribes that recognize the challenges of remaining contamination. One of the challenges to granting rights is that nearly half of the Hanford site has been designated as a national monument and is under management by the U.S. Fish and Wildlife Service (FWS). Therefore, permits are required through the FWS, which highlights another detail that should be considered during the land-transfer and land-access discussions.
Diversity of Tribal Perspectives. Mr. Preacher noted that each tribe will have different values and perspectives, so one should resist grouping tribes together. For example, although 50 tribes are located along the subject transportation routes, only three tribal representatives speak at meetings of the National Transportation Stakeholder Forum. They cannot speak for all of the tribes.
3.3 A RISK-MANAGEMENT FRAMEWORK FOR DECISION MAKING
In early 2012, prior to the Fukushima accident, an activity headed by Commissioner Apostolakis5 of the USNRC resulted in a report on a regulatory framework that incorporated risk. Although the framework does not focus exclusively on long-term stewardship decisions, it could be used for such decisions. The framework recognizes that federal agencies are trying to
4 In 1999, the Sagebrush Steppe Reserve was established to preserve Idaho’s high-desert sagebrush. See http://www.bnl.gov/bnlweb/pubaf/pr/2000/bnldoepr110900.html for more details. Part 1 of 2: http://ar.inel.gov/images/pdf/200409/2004090700578GSJ.pdf. Part 2 of 2: http://ar.inel.gov/images/pdf/200409/2004090700582TLR.pdf.
5 NUREG-2150 defines the new framework (see http://www.nrc.gov/reading-rm/doc-collections/nuregs/staff/sr2150/, accessed March 12, 2014).
establish processes that can be followed when making many different types of decisions and that produce common-sense answers, a theme throughout this workshop.
A task force to develop the strategic visions and implementation for the new framework has the following charter (Reckley 2014, p. 2):
[D]evelop a strategic vision and options for adopting a more comprehensive, holistic, risk-informed, performance-based regulatory approach for reactors, materials, waste, fuel cycle, and transportation that would continue to ensure the safe and secure use of nuclear material.
The task force added objectives and goals to the USNRC’s mission statement and outlined a decision-making process (see Figure 3-2). In addition, it force reviewed other federal agencies’ decision-making processes, including those of EPA and the Department of Homeland Security (DHS).
The mission includes the “protection of human health and safety,” the latter being defined as the absence of danger or risk. This framework refers to controlling risk or danger to acceptable levels, not removing them altogether. Acceptable levels of risk will differ depending on stakeholder perspectives.
FIGURE 3-2 Proposed risk management regulatory framework from the USNRC.
SOURCE: Reckley 2014.
Risk is introduced in this plan at the objective level (see Figure 3-2). Risk management requires a goal that highlights the importance of defense-in-depth (a large number of barriers would have to fail for an unwanted event to occur). The task force found similar aspects in other agencies’ decision-making processes: all had a feedback or circular process, and many of the elements (identify the issue, identify the options, analyze, deliberate, implement, and monitor) were the same.
Monitoring as part of the decision-making process is relevant to the topic of this session and the workshop. Monitoring ensures that the targeted outcome is preserved, and it also allows for addressing unforeseen problems. New problems are addressed by initiating a new cycle of the process. Although not included in Figure 3-2, communication at each of the decision-making process stages is important. All stakeholders need to have the same understanding of the issues.
How does this new framework translate into regulatory change? Although the task force does not foresee a dramatic change to the current regulatory approach, some parts of the USNRC, including some of the commissioners, are concerned about implementation. To temper this concern, one should recognize that the USNRC has been moving toward a risk-informed approach for the past 3 decades.
The USNRC regulates facilities and activities widely ranging from nuclear material storage to nuclear power facilities. The level of detail in a risk analysis should fit the scale of the problem. Risk analysis to support the storage of a small amount of radioactive material should not entail the same level of detail and complexity as an analysis to support a nuclear power facility (see Figure 3-3). Traditional engineering analyses, such as deterministic approaches (e.g., the impact of a large pipe breaking at a power facility), are ways to estimate risk and evaluate the impacts. Risk analysis is one of several considerations that feed into the deliberation of regulatory decision making.
NUREG-2150 defines the new framework. It was published in April 2012, and the public comment period, originally ending in November 2013, was extended to February 2014.6 Already, recent USNRC decisions regarding response to the Fukushima accident have incorporated aspects of risk analysis and the decision-making process. The USNRC is considering the inclusion of economic impacts (e.g., evacuation and its long-term consequences) into the analysis.
6 See http://www.gpo.gov/fdsys/pkg/FR-2013-11-25/html/2013-28065.htm, accessed March 12, 2014.
FIGURE 3-3 Decision-making process and the role of analysis, including risk assessment. Simpler facilities will have less complex analyses, while larger facilities such as nuclear power plants will have substantially more complex analyses. PRA = probabilistic risk analysis, PA = performance assessment, ISA = integrated safety analysis.
SOURCE: Reckley 2014.
3.4 SUMMARY OF DISCUSSION SESSION
Performance-based versus Compliance-based Approaches. Session moderator and planning committee member Patricia Culligan (Columbia University) asked if Mr. Reckley had seen a preference for performance-based or compliance-based monitoring. Mr. Reckley responded that the nuclear power utilities prefer prescriptive monitoring such as those designed to ensure compliance. Because of costs and liabilities, the utilities prefer to clearly understand the USNRC requirements. Although performance-based standards and monitoring allow for more flexibility, they are not preferred by everyone. The ultimate performance-based program is ALARA (as low as reasonably achievable), which provides the licensee with freedom on how to achieve the lowest dose allowable.
Bill Levitan (EM) provided the DOE perspective by introducing examples of two remedies for a low-level waste (LLW) containment system
that have different long-term environmental risks; one remedy was selected using performance-based criteria, the other with a prescriptive approach. For an LLW facility permitted under DOE authority, a performance-based standard of 25 mrem measured at 100 m from the boundary is used. The resulting design of the LLW facility will often utilize monitored natural attenuation and will have no liner (depending on the site for which it was designed). However, for a LLW facility permitted under RCRA or CERCLA, the remedy will often include liners because of prescriptive regulations. Ultimately, this remedy may result in increased environmental risk in future years when the liners fail. David Maloney (CH2M HILL) agreed and noted that lingering risks at some sites (e.g., Rocky Flats or Oak Ridge) are due to liners that were selected during cleanup following prescribed regulations or guidance.
Planning committee chair Paul Gilman (Covanta Energy) asked if EM has consent decrees or a tri-party agreement that has a prescriptive standard but implement a less-prescriptive, more performance-based standard. Mr. Levitan identified the Lower Watts Bar Reservoir presentation (see Petrie 2014) as one such example. Hanford is site for which a less-prescriptive approach has been introduced in a waste disposal facility. The encapsulation of contaminated material at the RCRA facility is not allowed under the prescriptive standard (regulation), which requires that the material be removed from the disposal facility, encapsulated, and then returned to the disposal facility. At Hanford, they are demonstrating that encapsulation can take place at the disposal facility to reduce time, exposure, and costs of transporting the contaminated material to another location.
Paul Gilman provided an example scenario in which stakeholders demand standards that exceed USNRC regulatory limits. The USNRC is overseeing the decommissioning of a uranium facility using ALARA, but the local residents want exposures that are lower than those determined by the facility and the USNRC. How would the USNRC—which is the regulatory authority and makes the final determination—resolve the dispute? The USNRC will engage with the stakeholders to understand and their concerns and then factor them into the decision-making process. Dr. Abu-Eid (USNRC) interjected that the decommissioning rule does have flexibility.7 Probabilistic risk analysis could be used to evaluate decisions and options. Also, understanding of the planned land use is important when making a final decision. For example, if the land use will be restricted, then the residents and local stakeholders should have a full understanding of the restrictions. Comments and concerns are collected through public meetings and feedback.
7 The decommissioning rule can be found at http://www.gpo.gov/fdsys/pkg/FR-1996-07-29/pdf/96-19031.pdf.
Kevin Crowley (National Academy of Sciences staff) expressed interest in the pushback within the USNRC to the new framework. He asked for details on the type of pushback (i.e., is it fundamental to the process or is it centered on implementation?) and, if possible, on general stakeholder feedback received thus far. Mr. Reckley responded that the pushback from the USNRC primarily centers on implementation and the impact on regulations; there is not a fundamental concern about the approach. In terms of public comments, the utilities have generally been lukewarm to the new framework, because it represents change and change costs money. Even if the change is in language only (such as the requirement to include defense-in-depth in analysis), costs are associated with it. There is also some concern that formal risk assessment methodologies, such as probabilistic risk assessment (PRA), are costly to carry out. The cost-benefit is difficult for the utilities to understand. Paul Black (Neptune) commented that if deterministic risk analyses were performed thoroughly, then they would be similar in complexity and costs to PRAs. Mr. Reckley agreed and pointed out that deterministic models originally did cost a great deal—through the development of reactor cores that were melted and other experiments. But those costs have been expended and absorbed by the utilities. Dr. Black made a further point that PRA accounts for consequences better than do deterministic models.
3.5 BETHPAGE/NORTHROP GRUMMAN AIRFIELD REMEDIATION AND STEWARDSHIP ON LONG ISLAND
Richard G. Mach, Jr.
In April 2012, the U.S. Navy issued updated guidance on optimizing its environmental remediation and removal actions at U.S. Department of Navy sites. The updates represent two decades of evolution on the Navy’s environmental decision making.8 Initially, the Navy approached environmental cleanup by focusing on identifying and installing remedies as quickly as possible. This resulted in deploying many pump-and-treat systems (P&T) that operated for a long time. It is not clear that the P&T systems were effective, but they were certainly costly. The Navy has reconsidered its initial approach, recognizing that initial remedies may not be effective or optimal. The updated guidance recognizes that optimization begins with thorough site investigation. which inherently includes risk-based approaches and assessments.
The Navy’s approach depends on good conceptual site models. Each
site is different, but risk and feasibility assessments based on the models guide the remediation decisions. The most recent changes to the Navy’s approach incorporate green and sustainable remediation practices. There is a strong belief that all three sustainability pillars fit easily into the CERCLA nine criteria—economic and societal values are included in feasibility assessments.
Approaches for managing contaminated groundwater plumes are site specific, especially for large complex sites. Factors include whether the site is active or closed, large or small, and whether the contamination is contained within or outside of the site. The DOD’s BRAC process resulted in many lessons learned on how to include the public in these assessments and decisions. Not only is plume management site specific, but also one treatment for the entire plume may not be appropriate for large plumes. In these cases, the plume is segmented into discrete areas. Treatment trains (combined or sequential remediation actions applied to sites with numerous or complex contaminants) are developed (e.g., P&T) through holistic thought and consideration of how to move from active to passive remediation systems (e.g., whether to contain the source or provide wellhead treatment).
Understanding the remediation goals and developing a remediation exit strategy are critical considerations early in the cleanup process. An “exit strategy” may not be defined as leaving the site entirely but it should include a mechanism for considering when to stop, modify, or change the current remedy if it is becoming less effective.9 For example, when the amount of money spent to reduce risk increases dramatically after a large fraction of the risk has been addressed, it may be a good time to reconsider the current remedy.
A good case study demonstrating the Navy’s current approach to remediation is the Bethpage Airfield. Established in the 1940s, the Bethpage Airfield is currently shut down (see Figure 3-4). The airfield was originally a Navy-owned property surrounded by a non-governmental entity (Northrop Grumman). Most of the government-owned property has been transferred and developed. The property is located in a heavily populated area with multiple water districts and multiple potentially responsible parties (PRPs) beyond the Navy and Northrup Grumman.10 However, the Navy and Northrup Grumman remain the responsible parties for the Bethpage groundwater plume, which resulted from more than 50 years of building Naval aircraft. The main contaminants are chlorinated solvents, primarily
9 When effectiveness continually decreases by year, perform an optimization study to increase performance, reduce costs, or define metrics for shutdown and convert to a passive remedy.
10 A potentially responsible party is a company or individual who had knowledge and was responsible for the contamination. More information can be found at http://www2.epa.gov/enforcement/superfund-enforcement.
FIGURE 3-4 The Bethpage Airfield property and contamination plume. The large map shows the relative distance of the Atlantic Ocean to the Bethpage site and its contamination plumes. The insert provides details on the location of the Navy-owned property with respect to the Northrop Grumman property. Residential properties surround the site.
SOURCE: Mach 2014.
perchloroethylene (PCE) and trichloroethylene (TCE). The contaminant plume is traveling south, has separated into “fingers,” and is expected to eventually reach the Atlantic Ocean (see Figure 3-4). Unlike the DOE legacy sites, radioactive contaminants are not a consideration.11 Bethpage is not an EPA site; the state (New York Department of Conservation [NYDEC]) is the lead regulatory agency.
The plume is split into operable units (OUs): OU1 is the onsite unit
11 Radon has been detected during cleanup, but it is not a significant contaminant.
and includes the source(s) of contamination, OU2 is a joint offsite plume shared by the Navy and Northrup Grumman, OU3 is Northrup Grumman’s offsite plume (see Figure 3-5). The original Record of Decision (ROD) for OU2 was signed by the Navy, Northrop Grumman, and NYDEC in 2003. OU2 remedies include onsite containment with P&T at the site boundaries, a large monitoring network, wellhead treatment for affected water districts, and sentry wells situated ahead of the plume’s migration.
Both onsite and offsite conceptual models were developed to better understand the plume’s migration. Onsite conceptual models benefitted from two factors: a concentrated and well-defined plume due to its proximity to the source and the ability to place wells in optimal locations to characterize the plume and validate the model. Offsite conceptual models were more challenging to develop and validate: the shape of the plume becomes complicated the farther it moves from the source, with many fingers and discontinuities. Monitoring for characterization and model validation is complicated by the inability to place wells in specific locations because of residential and industrial development. The rate at which the water districts pump water also affects the offsite plume. Each district can pump up to 106 gallons/day, but pumping varies by season (e.g., in winter months, less water is pumped). This affects the dynamics of the plume in ways that were not predicted by the original offsite conceptual models.
The current remedy for OU2 of the plume includes hot spot treatment, P&T with an exit strategy, and wellhead treatment. Wellhead treatment for affected water districts is in place and operational. Sentry wells for districts not yet affected by the plume but within its projected path are also part of the remedy. Sentry wells were established to detect contamination before it reached public water supplies. However, they are not always effective.
In one example, a private water company (Aqua New York) detected PCE in its water supply but the sentry wells did not. Wellhead treatment was installed at Aqua New York’s wells to treat the PCE, and the issue was resolved. Following this incident, many water districts not yet affected by the plumes became concerned, and the U.S. Congress became involved. Water districts close to the Atlantic Ocean wanted the plume to be contained to prevent its reaching their wells.
To address the concerns, the Navy worked with congress, the districts, EPA, the U.S. Geological Survey (USGS), and the U.S. Army Corps of Engineers to perform a full plume analysis and remedy optimization. The final report issued by the Navy in June 2011 considered several alternatives:
- Searching for and treating additional hot spots
- Developing improved conceptual site models
- Improving performance of sentry wells
- Fully containing the plume
The study found, with consensus among the participants, that the original ROD presented a responsible, smart, and protective remedy. Better characterization of the plume was needed, and additional hot spots should be identified. Full containment of the plume was not an option. This study resulted in changes to the current remedy to OU2, including paying existing wellhead treated facilities to pump at a constant rate year round to keep flows more predictable and to stabilize the plume.
Treating contaminated groundwater at the public water systems—i.e., at the point of human use—is a creative solution to a very complex problem. Most offsite cleanups are technically challenging, but asking the questions “When do you treat the plume?” and “Where do you treat the plume?” may prompt interesting answers that can help to define creative solutions. In this case, do you return all of the groundwater to drinking-water standards, or do you treat it at the public water districts when you are ready to drink the water?
3.6 SUMMARY OF DISCUSSION SESSION
Vapor Intrusion. Rula Deeb (Geosyntec) asked whether there were any vapor intrusion issues with any of the residences surrounding the site. Mr. Mach responded that few homes have been affected because the plume is not close to the surface. However, approximately 10-16 homes have soil affected by vapor intrusion. The Navy has worked with NYDEC and Department of Health to establish remedies and monitoring.
Litigation. Planning committee member Michael Kavanaugh (Geosyntec) asked whether litigation and equipment failure were part of the risk calculation. Mr. Mach responded that he did not know whether or how litigation risk was accounted for in the risk calculation but noted that litigation is part of the current solution. New York public water systems require their personnel to acquire, install, and operate any wellhead treatment system attached to a public water system. The Navy cannot use cleanup money to pay the public water districts to purchase or operate wellhead treatment equipment. The current solution is for water districts to place a claim against the U.S. government. A settlement agreement is worked out in which the Navy pays capital and operation and maintenance (O&M) costs to the public water systems for the wellhead treatment equipment. The exception is Aqua New York; the Navy is building and operating the wellhead treatment system for that location because it is a private company.
Technical Impracticability (TI) Waivers. Dr. Kavanaugh noted that an interesting aspect of this case study is the apparent public acceptance of “the plume is too big to contain” argument—that containing it would not
be technically practicable. He also noted that the Navy is not a strong supporter of TI waivers. He asked whether Bethpage is a clever way around this issue. How might others, including EM, deal with this situation? Mr. Mach responded that TI waivers were not part of the approach for the Bethpage remedy. Rather, the Navy presented a solution that made sense for a massive plume. Although the public water districts affected by the plume did not initially support the wellhead treatment remedy, the wells were already being treated for other contaminants. The proposed solution was understood and eventually accepted by all affected districts.
The Navy has a saying: “TI/RI”-technical impracticability/regulatory impossibility. The TI process is difficult and not worth pursuing from the Navy’s perspective. Another approach is to provide program managers with tools to consider other options when defining remedies. For example, the use of MCL in the aquifer implies that the aquifer is the receptor. Another option is to apply the standard to groundwater at the drinking point. In the Bethpage remedy, wellhead treatment is used to eliminate the exposure at the point of use, and water as a resource is still respected.
Robin Anderson (EPA) pointed out that pressure is important to improving technologies and remedies for sites without a clear path to full remediation, such as in the Bethpage example. What mechanisms are in place to push for new technologies or alternate solutions?Mr. Mach noted that the Navy’s Optimization Policy requires continual assessment and reviews with third-party evaluations. The Navy is always looking for better solutions and pathways for new technologies.
Dr. Deeb recalled an example from Workshop 1. Steve Cobb’s presentation on the Anniston Army Depot remediation provided a similar problem with a similar solution as the Bethpage case study. If a zone of the plume cannot be treated, as has been shown in the Bethpage remedy, then draw a circle around it and revisit it every 5 years to see if a solution exists, but consider using a TI waiver. This would have alleviated the requirement for the Army to find the source of the contamination—an activity for which it had already spent tens of millions of dollars.
Dr. Deeb asked Ms. Anderson how TI waivers could be a viable option for EPA in such cases. Ms. Anderson responded that TI waivers have an inherent perception problem. The implication is that, unless there is constant pressure to overturn a TI waiver, it will not be overturned. Mr. Mach agreed with this assessment, noting the Navy’s decision to not pursue TI waivers is partially due to the current stigma attached to them. Creative solutions to address the contamination at large, complex sites are possible without TI waivers and include plans that are responsive to stakeholder concerns.
Intrinsic Value of the Environment. Dan Goode (USGS) commented that, in the Bethpage example, the environment is left out of the solution.
Was someone representing the environment in the Navy study or the negotiations for the final remedy? Mr. Mach noted that, although ecologic receptors were considered in the remedy, few were identified (the plume is deep, at 800 feet below the surface). Studies have shown that once the plume reaches the ocean, the environmental risk minimal for discharge of chlorinated solvents into surface water is minimal. Bill Levitan (EM) noted that the Bethpage case study is another example of a performance-based solution: the intrinsic value of groundwater is included in the assessment. In this case, the value of groundwater on Long Island is derived from the drinking.
Conceptual Models. Robin Anderson (EPA) asked about the effectiveness of the OU2 remedy and whether data have shown that the plume has detached from the source. Mr. Mach responded that models and data show that the plume is starting to detach but not enough data exist to confirm separation. The ROD for OU3 recently signed by Northrup Grumman and NYDEC has the same remedy as OU2, suggesting that NYDEC is comfortable with the remedy’s effectiveness.
Ming Zhu (EM) emphasized the importance of modeling in decision making. In this presentation, the original offsite conceptual models predicted that the public water systems would not be affected for 40 years, but the contamination showed up in 10 years. He asked, “What lessons did this experience teach you?” Mr. Mach responded that the optimization team determined the initial models to be wrong for several reasons, most importantly that the model was designed for a particular piece of the plume but was extrapolated beyond that point. The optimization team recommended more accurate and localized models be developed.
3.7 INSTITUTIONAL CONTROLS ON LOWER WATTS BAR RESERVOIR
The Lower Watts Bar Reservoir was constructed by the Tennessee Valley Authority (TVA) between 1939 and 1942 for the purposes of navigation, power generation, flood control, recreation, and irrigation. It has also been a source of drinking water for the local communities. Most of the land along the Lower Watts Bar Reservoir is privately owned. Numerous rivers, creeks, and streams feed into the reservoir. Dams located along the rivers affect their flow rates and flow directions. Upstream of Lower Watts Bar are several potential sources of contamination to its water, soil, and sediments including the Oak Ridge Reservation (ORR, Figure 3-6).
In the late 1980s, there was rising concern that the sediments in Lower
Watts Bar Reservoir may have been contaminated because of prior releases from DOE nuclear weapons development operations at ORR. By 1989, ORR was placed on the National Priorities List as a CERCLA site requiring investigation by EPA. In an effort to protect the public, an Interagency Working Group (IAWG) consisting of DOE, EPA, the State of Tennessee (the State), TVA, and the Army Corps of Engineers (the Corps) was formed in 1991. The IAWG put in place institutional controls on sediment in Lower Watts Bar Reservoir to protect the public from potential human health risks. In 1992, the Federal Facility Agreement (FFA) for the Oak Ridge Reservation was signed by DOE, EPA, and the State. The FFA governs the CERCLA cleanup of the Oak Ridge Reservation and any offsite areas impacted by DOE operations in Oak Ridge.
One of the first offsite areas addressed under the agreement was Lower Watts Bar Reservoir. As with the Bethpage example (see Mach 2014), at the time there was no formalized framework for addressing remediation of
FIGURE 3-6 Locations of the Lower Watts Bar Reservoir, Clinch River, Poplar Creek, and the Oak Ridge Reservation (ORR) (site boundary shown in dotted line). ORR’s major facilities are shown: East Tennessee Technology Park ([ETTP] previously K25), the Oak Ridge National Lab ([ORNL] previously X10), and Y12. SOURCE: DOE, http://energy.gov/sites/prod/files/2014/01/f6/LowerWattsBarRes_0.pdf.
contaminants from offsite sources. The selected remedy for Lower Watts Bar Reservoir was to control human exposure to sediments through institutional controls. Because the IAWG was already in place, it became part of the selected remedy. The Record of Decision for Lower Watts Bar Reservoir was signed in 1995. In 2001, a unified 5-year CERCLA review schedule for all sites and programs was established, focusing all programs into a single review cycle. There have been two additional reviews (in 2006 and 2011).
To better understand the constituents of the contamination of the sediment, one must understand the history of DOE activities at Oak Ridge. ORR had three main facilities: Y12, Oak Ridge National Laboratory (ORNL, previously X10), and ETTP (previously K25, see Chapter 1 in this volume, Oak Ridge Reservation). A myriad of contaminants were released from these facilities during past nuclear weapons production operations. However, cleanup efforts have focused on a subset of the contaminant list: mercury, cesium (for radioactive by-products), and PCBs. This subset of representative contaminants is used to gauge the presence of other contaminants.
Y12 was originally used to enrich uranium. Significant quantities of mercury were used in the lithium separation process prior to enrichment. Y12 is the main source of mercury contamination at and around ORR. Twenty-four million pounds of mercury were vouchered/accepted at Y12. It is estimated that 0.5 million pounds of mercury remain in buildings; 0.5 million pounds were released to East Poplar Creek and nearby soil; and approximately 1.2 million pounds are “unaccounted for.”12 Mercury continues to leak and seep from storm drains. In 1954, the volume of mercury released to East Poplar Creek was approximately 72,000 pounds/year. In 2000, this amount had been reduced to about 30 pounds/year.
Initial studies and experiments on reprocessing and extracting/purifying plutonium were performed at X10. ORNL is located where X10 used to operate and is the main source of the cesium contamination. Many radioactive and hazardous by-products from these experiments were released to White Oak Creek, which feeds directly into the Clinch River. Cesium has been detected in sediment throughout the Clinch River and Lower Watts Bar Reservoir (and other places). A sediment retention dam was built at the mouth of White Oak Creek; it has been effective in reducing the rate of cesium releases into the Clinch River but has not stopped them completely.
The gaseous diffusion plant at K25, currently the ETTP, used vast amounts of electricity to enrich uranium. PCBs were used in the power system components (e.g., condensers, capacitors, transformers). Because
12 Barrels of mercury were counted and vouchered upon delivery at Y12. However, not all of the barrels were full (i.e., a percentage of the barrel could have been partially full), which was not noted in the vouchering process. Some fraction of 1.2 million unaccounted for mercury was missing upon delivery.
of the many PCB oil spills on site, K25 is the main source of PCB contamination for Oak Ridge. PCB contamination spread from K25 to Poplar Creek, which also feeds directly into the Clinch River and Lower Watts Bar Reservoir.
The area downstream of Oak Ridge was split into two operable units (Clinch River/Poplar Creek and Lower Watts Bar Reservoir) to address remediation challenges. Costs for full remediation of the sediment within both operable units were estimated to be between $30 and $60 billion (1995 dollars), a large fraction of EM’s overall cleanup budget of $200 billion.
Alternatives to full remediation of sediment were explored. It was determined that cesium was the largest human exposure pathway for sediment with acceptable risk set at 11 picocuries per gram (pCi/gm) (The risk calculation assumed that sediment is dredged from the rivers or creeks and placed onto land. This land was then used to plant vegetables and crops for human consumption). Sediment maps were generated by sampling for cesium along the reservoir. The presence of dams affects flow rates and directions, which affects sediment deposition patterns within the reservoir.
As mentioned previously, the IAWG was established to protect the public from exposure risk from sediment disturbances. Any activity on the reservoir that potentially disturbs sediment requires IAWG approval. The application for approval requires the following information:
- Review the proposed action (size, location, elevation)
- Collect historical data on vicinity of project
- Evaluate levels of cesium in samples
- Decide whether further investigation is needed
- Determine whether DOE action is required to reduce risk
DOE’s involvement is not needed for the vast majority of the applications. However, DOE action is sometimes required. For example, the Tennessee Department of Transportation had taken sediment samples from a bridge excavation and left its samples in the yard of a resident. DOE investigated and determined that cesium activity in the sediment was below the established acceptable risk levels.
What happens if the risk exceeds acceptable levels? This depends on whether the land is public or private. Neither DOE nor the IAWG has authority to restrict citizens from activities on private property. In a hypothetical example, a citizen plans to build a dock that would disturb sediment contaminated above the 11 pCi/gm level. In this case, DOE would have to confirm that workers were protected while building the dock and would be responsible for disposing of the contaminated debris from the project.
The IAWG has determined that mercury and cesium contamination are
co-located in the sediments and exposure pathways. Therefore the protective measures established for cesium (i.e., institutional controls of sediment) also apply to mercury. However, the exposure pathway for the third representative contaminant, PCB, is through ingestion of fish. The record of decision requires that samples of fish, surface water, and sediment be tested annually for PCBs. The fish sample is collected from several species including largemouth bass, striped bass, hybrid (striped and white) bass, and snapping turtles. Subsistence fishermen make up a small percentage of the population along the Lower Watts Bar Reservoir, but environmental justice requires that their protection be considered. Therefore fishing advisories have been issued along the Clinch and Tennessee Rivers to reduce the risk of PCB exposure.
There are several challenges for remediation: multiagency consensus, identification of contaminant source terms, long-term funding, and public acceptance. Two groups within the IAWG, the Corps and TVA, are not signatories to and are not bound by the Oak Ridge Reservation FFA, which can add difficulty to building consensus for decision making. The fish advisories are established by the Tennessee Department of Environment and Conservation (TDEC), which is not required to make them consistent with the Lower Watts Bar Reservoir’s Record of Decision (this example also highlights the challenge of managing multiple and potentially conflicting remediation decisions within a single state agency because TDEC is also a signatory to the FFA).
There are multiple sources of contamination. TVA’s Kingston Fossil Plant, for example, is an upstream source of mercury as well as ORR.13 It is estimated that 25 percent of the PCB contamination came from DOE activities at ORR, the remaining 75 percent from other industries. The main contaminants are long-lived, and the current remedies require long-term funding. Cesium-137 (Cs-137 is the main contaminant) has a half-life of 30 years, and PCBs and mercury do not decay or degrade. Current remedies assume the federal government will continue to maintain protective actions for hundreds of years.
Finally, public opinion and perception impact the final remedies. Risk may inform proposed remedies, but public values determine the remedy selections. Understanding and interpretation of risks vary across citizens groups. For example, the following risks of exposure are presented at a local town hall meeting: There is a 10-4 to 10-6 risk of cancer due to contaminated sediments and there is a fish advisory against eating striped bass. Parents and grandparents may interpret that risk to mean, “My child/grandchild will get cancer if he or she swims in the reservoir.” On the other
13 In 2008, a retaining wall at the Kingston Fossil Plant, collapsed releasing 2.5 million cubic yards of fly ash across hundreds of acres. The fly ash contaminated the river and homes. Fly ash contains mercury among other hazardous chemicals.
hand, commercial fishermen and new property/land owners may interpret the risk to mean, “My job is in jeopardy,” or “The value of my home has dropped.”
In the final record of decision for the Lower Watts Bar Reservoir, no one group was happy but perhaps that means it was the right decision for everyone.
3.8 SUMMARY OF DISCUSSION SESSION
Protectiveness of the Remedies. Michael Kavanaugh (Geosyntec) asked if the CERCLA 5-year reviews have established that the remedy is protective and whether any group questions the protectiveness. Mr. Petrie confirmed that the reviews have shown the remedy is effective at protecting human health and the environment. However, commercial fishermen feel the remedy is overly protective and that it adversely impacts the economic value of the reservoir; this has not been seen in the economic analyses.
The following example provided by Mr. Petrie highlighted the intersection of public opinion, awareness, and knowledge and the protectiveness of proposed remedies: The Lower East Fork Poplar Creek Flood Plain Record of Decision14 had established a reference acceptable dose limit of 500 parts per million (ppm) of mercury based on ecological risk. The public was concerned about the overall disruption and cost of the proposed remedy and asked for a less disruptive and consequently less protective solution. The original reference dose was calculated assuming the exposure to mercury was due entirely to its most toxic form (organic mercury or methylmercury). A more realistic model assumed the majority of the mercury is in the form of elemental and inorganic mercury (other naturally occurring forms of mercury) not solely as methylmercury. A re-calculation of reference dose determined the acceptable limit to be 400 ppm. The remedy for this dose resulted in less disruption of the sediments and lands; as a consequence, the overall costs were reduced. A 5-year review indicates the mercury uptakes in the environment are larger than expected, so the modified remedy may not have been as protective as originally thought.
A workshop attendee asked about the Lower East Fork Flood Plain decision and asked for additional information on the public’s decision to push for a less protective remedy. Mr. Petrie responded that the most vocal opponents had property that would have been affected by the initial remedy. Other members of the public thought that the cost of the remedy was not justified. In general, the public’s initial reaction to an ecological risk is to assume that a deer (e.g., Bambi) or similar animal species will suffer from the uptake of contamination. In reality, the assessments indicate that
less sympathetic species will be impacted the most; the animals identified in the analysis of mercury uptakes for the Lower East Fork Flood Plain were spiders and starlings. The public is not as concerned about the risk to these creatures as they are to some others.
Fish Monitoring Program. A workshop attendee asked about the fish monitoring program. Mr. Petrie responded that fish monitoring was previously performed annually by multiple groups—TVA, DOE, U.S. Fish and Wildlife Service, the State of Tennessee—resulting in comments by local stakeholders that “the biggest threat to fish in the reservoir was the scientists performing environmental monitoring.” A Fish Tissue Committee was established to review analyses, assign work, and share data among these groups. This has increased the amount of data available to any one of the tissue committee partnering agencies for analyses. Eric Pierce (ORNL) noted that the fish analysis has evolved to the point of catch-and-release after a tissue sample is taken.
Best Practices. Dr. Kavanaugh asked what best practices might be identified on risk-informed remedy selection. Mr. Petrie responded by focusing on the novel aspect of the decision-making processes that were developed to establish the remedy. The risk-informed decision process was a necessity because of several factors: initial impractical cost estimate of $30 billion to $60 billion, sediments being buried and continued to be buried, and no existing standards for contaminated sediments in Tennessee. Notably, the fish advisories were based on risk formulae for ingestions. But other factors beyond risk were weighed in the decision-making process—such as public and community values.
A workshop attendee noted that all of the pieces of a sustainability approach—consideration of risk to health and environment weighed against economic and societal values—can be found in this case study. It was suggested that the sustainability approach was “backed into” in the process. Mr. Petrie agreed and noted that similar sustainability approaches are currently being used to guide remediation decisions for new sources of contamination (such as mercury contamination from the Kingston plant fly ash release).