VIII
Decision-Making Process

Previous chapters have examined the current state of technology that delimits the options for on-site disposition of tank wastes and analyzed the informational and analytical needs for assessing and monitoring whether a particular disposition plan meets applicable criteria. In this chapter, the committee elucidates the decision-making environment in which the Department of Energy (DOE) operates. Illustrating that environment is important to understanding how the success of the tank remediation program depends on, among other factors, the way DOE approaches its decisions about tank wastes. Specifically, this chapter illustrates the following:

  1. There are considerations in high-level waste cleanup and disposition decisions that extend beyond merely meeting the dose limits and other nonquantitative requirements defined in waste determination criteria;

  2. Site-specific characteristics translate into different site priorities;

  3. DOE operates within an extremely complex and overlapping decision-making structure, which is a source of programmatic risk;1

  4. DOE’s decision-making paradigm has evolved with time; and

  5. Given the programmatic risks outlined previously and those identified in this chapter, a more risk-informed, participatory, consistent, transparent, and consultative decision-making process would make DOE’s waste management decisions more robust, in the sense that DOE is more likely to succeed in its tank remediation mission.

This chapter is not intended to answer the question, How clean is clean enough? for any particular waste stream, tank, tank farm, facility, or site. Instead, it responds to the statement of task by making recommendations that the committee believes will improve management and decision making so that such questions are answered in a consistent, transparent, and scientifically credible manner that protects public health and the environment now and for generations to come.

The legal authorities that apply to DOE are pivotal in shaping the decision-making framework in which DOE operates. Hanford (not located in a state covered under Section 3116 of the Ronald Reagan National Defense Authorization Act [NDAA] of Fiscal Year 2005) operates in a different management and legal environment than the Savannah River Site and Idaho National Laboratory, which are located in states covered under Section 3116. Nevertheless, the considerations relevant to management and decision making under Section 3116 are fundamentally the same as those that apply to a process for determining an acceptable path forward for Hanford, and DOE Order 435.1 is still in effect at all sites. Whether or not Section 3116 applies, a wide range of site-specific regulatory, economic, institutional, and waste management considerations are part of the decision-making process for waste determinations, as explained in the following section.

MULTIPLE DIMENSIONS OF RISK

The dose limits set out in performance objectives and associated guidance are a fundamental and necessary start-

1

Programmatic risk is the risk to cost, schedule, and technical performance of a program. It is associated with all uncertainties, including legal uncertainties that can result in delays, cost increases, and failure to reach the established goals. For example, DOE defines the programmatic risk as high for a given project if the technical approach has not been identified for critical or significant portions of the project; key technologies do not exist for critical or significant portions of the project; current investments do not support the resolution of the project’s science and technology needs; project end point is not determined or supported by stakeholders and Native American tribal nations; waste/material quantities and characteristics are unknown; process operations are not identified or supported by stakeholders and Native American tribal nations; final disposition location for waste/ material has not been identified; activity involves multiple sites; no concurrence has been reached between sites; or a facility does not currently exist and there are no plans for a new facility (DOE, 1998).



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 91
Tank Waste Retrieval, Processing, and On-Site Disposal at Three Department of Energy Sites: Final Report VIII Decision-Making Process Previous chapters have examined the current state of technology that delimits the options for on-site disposition of tank wastes and analyzed the informational and analytical needs for assessing and monitoring whether a particular disposition plan meets applicable criteria. In this chapter, the committee elucidates the decision-making environment in which the Department of Energy (DOE) operates. Illustrating that environment is important to understanding how the success of the tank remediation program depends on, among other factors, the way DOE approaches its decisions about tank wastes. Specifically, this chapter illustrates the following: There are considerations in high-level waste cleanup and disposition decisions that extend beyond merely meeting the dose limits and other nonquantitative requirements defined in waste determination criteria; Site-specific characteristics translate into different site priorities; DOE operates within an extremely complex and overlapping decision-making structure, which is a source of programmatic risk;1 DOE’s decision-making paradigm has evolved with time; and Given the programmatic risks outlined previously and those identified in this chapter, a more risk-informed, participatory, consistent, transparent, and consultative decision-making process would make DOE’s waste management decisions more robust, in the sense that DOE is more likely to succeed in its tank remediation mission. This chapter is not intended to answer the question, How clean is clean enough? for any particular waste stream, tank, tank farm, facility, or site. Instead, it responds to the statement of task by making recommendations that the committee believes will improve management and decision making so that such questions are answered in a consistent, transparent, and scientifically credible manner that protects public health and the environment now and for generations to come. The legal authorities that apply to DOE are pivotal in shaping the decision-making framework in which DOE operates. Hanford (not located in a state covered under Section 3116 of the Ronald Reagan National Defense Authorization Act [NDAA] of Fiscal Year 2005) operates in a different management and legal environment than the Savannah River Site and Idaho National Laboratory, which are located in states covered under Section 3116. Nevertheless, the considerations relevant to management and decision making under Section 3116 are fundamentally the same as those that apply to a process for determining an acceptable path forward for Hanford, and DOE Order 435.1 is still in effect at all sites. Whether or not Section 3116 applies, a wide range of site-specific regulatory, economic, institutional, and waste management considerations are part of the decision-making process for waste determinations, as explained in the following section. MULTIPLE DIMENSIONS OF RISK The dose limits set out in performance objectives and associated guidance are a fundamental and necessary start- 1 Programmatic risk is the risk to cost, schedule, and technical performance of a program. It is associated with all uncertainties, including legal uncertainties that can result in delays, cost increases, and failure to reach the established goals. For example, DOE defines the programmatic risk as high for a given project if the technical approach has not been identified for critical or significant portions of the project; key technologies do not exist for critical or significant portions of the project; current investments do not support the resolution of the project’s science and technology needs; project end point is not determined or supported by stakeholders and Native American tribal nations; waste/material quantities and characteristics are unknown; process operations are not identified or supported by stakeholders and Native American tribal nations; final disposition location for waste/ material has not been identified; activity involves multiple sites; no concurrence has been reached between sites; or a facility does not currently exist and there are no plans for a new facility (DOE, 1998).

OCR for page 91
Tank Waste Retrieval, Processing, and On-Site Disposal at Three Department of Energy Sites: Final Report ing place for the dialogue about decision making for disposition of tank wastes. Analysis of the site’s performance and its ability to meet applicable criteria are part of the legal requirements under Section 3116 of the 2005 NDAA, DOE Order 435.1, and other applicable standards, such as the standards for radionuclides in drinking water that are used as groundwater protection criteria. As explained below, the dose limits in the performance objectives do not fully address the multiple dimensions of risk that characterize tank waste management. Risks2 from any disposal facility actually vary over space and time. Examples of different time-space profiles of risks are the following: Risks that are very low for a long time and never really increase above low levels (no “amplitude” over time) Risks that are very low for a long time, but rise fairly rapidly later in time (albeit to levels still below the performance objectives) Risks that are quite elevated near the source (the tank), but decline to acceptable levels at more distant points of compliance Risks can also differ qualitatively, in terms of their mode of impact on health. For example, some risks are immediate or acute,3 whereas others are chronic or delayed in their onset (as in the case of carcinogenic compounds). Different compounds also can harm the body or environment in different ways, some affecting reproductive outcomes and others possibly causing harm only to the elderly or already infirm. These qualitative differences in the types of waste in question may also undermine the usefulness of safety established strictly on the basis of performance objectives. Although each of these hypothetical outcomes might meet the performance objectives, they might engender quite different degrees of concern among the public and decision makers. These kinds of differences in the risk profiles of different waste management choices have important implications for concerns about long-term institutional controls. Some risk profiles may place fewer burdens on assumptions about long-tem institutional controls to ensure that risks are acceptable to the public. The technical complexities of the disposal sites for tank wastes and their interaction with the geosphere may be included where possible in the performance assessments and in the models they use. Therefore, the performance assessment and sensitivity analyses that are performed using different scenarios can show the impact of various decisions on projected risks. Indeed, DOE’s waste determination for the closure of Tanks 19 and 18 (see DOE-SRS, 2005a, Figure 7-14, p. 137) shows how the Savannah River Site used its performance assessment for these two tanks to make tank closure decisions (although the decision tree is somewhat simplified and there is limited supporting information). However, the performance assessment is not intended to provide understanding of the overall nature of the risks that a person at the site would encounter. For example, each of the three sites considered in this report has additional risks other than those represented by the waste in the tanks (e.g., cribs and previous waste leaks at Hanford). It may not be possible to remediate contamination in soils around the tanks until the tanks are closed. The risks from such preexisting contamination may be substantially greater than any risks that might be created by leaving an incrementally larger (yet still relatively small) amount of waste in the tanks to enable their prompt closure. Without considering all of these risks and their relationships in time and space in one comprehensive risk-benefit analysis, it is difficult to see how the programmatic goals fit together and what trade-offs should be made in managing risks. Overall, if all potentially related risks for a site are not considered, decisions may be made that do not reduce the risks from the site “system” as a whole to the maximum extent practical. Thus, considering only whether performance objectives are met for a tank closure or a low-activity waste disposal site may not provide a full picture of risks at the site and how they vary in time and space. Additional risks (e.g., the risk of a leak during waste retrieval, the risk of leaving the tanks emptied for 5 to 10 years before grouting them) and trade-offs among such risks could be identified as part of making a decision for each individual element of the site waste management plan. The ALARA (as low as reasonably achievable) radiation exposure provisions in the performance objectives mentioned in the waste determination criteria require trade-offs among technology limitations; risks to workers, the public, and the environment; and costs. However, the committee has not seen documentation showing detailed risk-benefit analyses of such trade-offs. The waste determinations for Tanks 18 and 19 were considerably better supported than documents the committee reviewed for its interim report, but they still did not contain sufficient information for a complete evaluation of how the trade-offs were made (see discussion in Chapter III on “how clean is clean enough?”). The ALARA analysis required in the performance objectives calls for trade-offs between incremental costs and incremental risk reduction. DOE used such an analysis to decide much tank waste retrieval was enough (see Chapter III). However, the risk-benefit analyses presented in the Idaho National Laboratory and Savannah River Site tank waste determinations were not sufficiently detailed or transparent. The criteria on which trade-off decisions between cost, worker dose, and public dose were made are not 2 In this chapter, risk is intended in a generic sense as risk to workers, the public, and the environment. 3 Except in extraordinary circumstances (such as an operational accident), it is not likely that anybody would receive an acute dose of radiation from the tanks.

OCR for page 91
Tank Waste Retrieval, Processing, and On-Site Disposal at Three Department of Energy Sites: Final Report stated. Finding quantitative cutoff points at which further risk reduction is no longer worth its cost may not be possible in such a multiobjective situation. Similarly, a formal cost-benefit analysis or a comparison of incremental (marginal) risk reduction and cost by reducing both to a single, monetary metric may not be necessary. However, if the options for further action (greater risk reduction) are laid out clearly (i.e., risk-benefit trade-offs are clearly explained), it is possible that a consensus, or at least a majority view, can be developed around a preferred option among a set of diverse stakeholders. The cost-risk trade-off is the most salient way to show that the performance assessment is only one source of input to a decision. Stakeholders may consider other nonrisk criteria to be important and part of a comprehensive risk-benefit analysis. This analysis would include the distribution of the burden of the risks to different groups, concerns about permanent (or very long term) risks versus temporary or near-term risks, uncertainties affecting the potential risks in time and space, and the potential to detect and mitigate hazards if they emerge. The result of this analysis may alter societal preferences for one management plan over another and impact the likelihood that a plan will achieve its goal. An illustrative and extremely simplified example of an approach for considering tank closure decisions that directly examines trade-offs among these different forms of risk is presented in Chapter X. SITE-SPECIFIC CONDITIONS CAN LEAD TO DIFFERENT DECISIONS In a consistent process for decision making, the framework for considering risks and the trade-offs among them is the same for all of the tanks at all of the sites. However, different conditions and priorities may exist at different sites. The considerations influencing the trade-offs among relevant risks may be quite different from tank to tank and from site to site. Thus, it is logical to expect that DOE may make quite different choices about the timing of tank closure and the extent of cleanup to undertake before closure based on the specific details at each site. In Chapter V, the committee describes the importance of decoupling tank waste retrieval from immediate closure of some tanks to evaluate whether it is desirable and practical to undertake further cleanup before closing a tank (see also Appendix E). Such a trade-off in timing of final closure may be appropriate for some of the more difficult-to-clean tanks at the Savannah River Site whereas tanks at the Idaho National Laboratory may be appropriately closed right away because the amount of remaining radioactive material is small. An analysis and understanding of the performance objectives is the logical starting place for an analysis of risks posed by tank wastes at the Savannah River Site, Hanford, and the Idaho National Laboratory. To reach decisions about managing tank wastes, DOE should also consider other factors, including the following: The unique risks created by the proposed decision pathways that are not captured in the performance assessment or performance objectives for specific tanks or on-site disposal facilities; Risk considerations at the sites that are altered by cleanup decisions; Changes in, and interactions among, tank residual risks and other waste streams that are associated with the separation of tank wastes into different waste streams; Considerations not captured in risk analyses or performance assessments, such as costs, distributional or equity concerns, and other societal concerns; and The views of the states, Native American tribal nations, and other stakeholders in decision making. To illustrate more fully how this recommendation impacts decision making, the example in Chapter X shows in an extremely simplified fashion how some of these additional factors can affect the choices about cleanup and waste disposition. The analysis also provides an example of how these additional considerations can be incorporated into the decision-making process in a formal, structured manner that permits consistent approaches to decision making but still allows different outcomes to emerge as a result of site- and tank-specific conditions. Setting out the assumptions and time horizons over which risks can impact health and the environment demonstrates the interplay of factors that complicate risk analysis and shows that performance assessments and performance objectives alone do not capture the full range of possible risks. PROGRAMMATIC RISK As shown in previous chapters, the technical challenges that DOE faces in managing tank wastes, tank farm systems, and surrounding environments are difficult, interconnected, and unprecedented. In addition to technical challenges, DOE has faced some legal challenges, due to the litigation concerning provisions for waste determination in DOE Order 435.1. Section 3116 of the 2005 NDAA creates some further complexities in the regulatory environment: DOE plans to declare sodium-bearing waste and waste from certain Hanford tanks to be transuranic waste and dispose of them at the Waste Isolation Pilot Plant (WIPP). However, although it is proceeding down this path, DOE does not have agreement from the State of New Mexico that these wastes qualify for disposal at WIPP. The recent regulatory problems that Hanford encountered in its transuranic waste determinations has a real impact on the entire DOE waste management program (see Sidebar VIII-1 and GAO, 2004). DOE has wastes for which Section 3116 cannot be used, such as all wastes at Hanford, wastes being sent out of the host state for disposal (e.g., sodium-bearing waste), and

OCR for page 91
Tank Waste Retrieval, Processing, and On-Site Disposal at Three Department of Energy Sites: Final Report SIDEBAR VIII-1 Recent Government Accountability Office and DOE Inspector General Findings and Recommendations Recent reports by the Government Accountability Office (GAO, 2005b) and DOE’s Office of the Inspector General (DOE-OIG, 2005) point out the legal risks DOE is facing in its accelerated cleanup program. Excerpts from these reports follow. The types of challenges that could increase cleanup costs at [the Hanford, Idaho National Laboratory, and Savannah River] sites include the following: Delays in disposing of highly radioactive wastes. In early 2005, DOE reported that a slip in the scheduled opening of DOE’s planned repository at Yucca Mountain, Nevada, would delay shipment of waste by at least 2 years—and possibly for as long as 7 years—due to technical and regulatory issues. As a result, sites now storing high-level waste and spent nuclear fuel have been reevaluating their waste disposal plans and associated cost and schedule estimates. The sites potentially affected include Hanford, Idaho National Laboratory, Savannah River, and West Valley. Most sites expect costs to increase as disposal schedules slip. In its fiscal year 2006 budget request, DOE estimated that a five-year delay in opening the Yucca Mountain repository could increase costs by as much as $720 million at its three largest sites. This includes building additional storage buildings and added operating costs. Legal obstacles preventing DOE from implementing aspects of its cleanup approach. DOE faces challenges to its planned treatment strategy at the Hanford Site that could potentially increase costs. A 2002 lawsuit challenged DOE’s plans to separate and determine that a portion of its waste could be treated and disposed of as other than high-level waste, and to DOE’s plans to close tanks leaving some radioactive residual in the tanks. In October 2004, a federal appeals court overturned a district court ruling against DOE and held that it was premature to rule on the matter until DOE implemented its strategy. Federal legislation passed in October 2004 provided authority for DOE to carry out its acceleration completion strategy at its Savannah River Site and Idaho National Laboratory. However, the law excluded the Hanford Site. If similar authority is not provided for the Hanford Site, costs at the site could increase significantly—up to $67 billion, according to DOE’s estimate. Similarly, uncertainty surrounds Hanford’s ability to accept waste from other DOE sites as the result of two ongoing lawsuits: one involving a challenge by the State of Washington to DOE’s plan to ship low-level, low-level mixed, and transuranic waste into the state, and one concerning a recent Washington state citizens’ initiative that could prohibit Hanford from accepting additional waste until existing waste is cleaned up. Although DOE believes it will ultimately prevail in these lawsuits, some cleanup activities at the other sites may face delays and increased storage costs until the issue is resolved.a The Office of River Protection (ORP) pursued the Transuranic Mixed Tank (TRUM) Waste Project without sufficiently addressing regulatory and permitting issues. […]The Department has not yet completed the regulatory actions required under the National Environmental Policy Act of 1969 (NEPA) prior to proceeding with the TRUM waste project.[…] On December 15, 2003, the Department’s ORP approved and issued Supplement Analysis for Hanford Tank Farm Contact-handled Transuranic Mixed Waste Treatment, Packaging, and Storage (Supplement Analysis) to the 1996 Environment Impact Statement (EIS). However, the Supplement Analysis did not address key issues which the Department’s Office of Environment, Safety and Health (EH) considered critical to the public. Specifically, EH noted the analysis did not: Clarify the waste classification in light of recent court decisions; Address the cost, feasibility, additional waste generation, and timing issues related to reversing the TRUM waste treatment process if the waste is not accepted for disposal at WIPP; Consider the environmental impact of reversing the action; and, Address potential worker impact for storing the waste above ground. We recommend that the Principal Deputy Assistant Secretary, Environmental Management, direct the Manager, Office of River Protection to: Mitigate regulatory and permitting risks, including the concerns raised by EH before resuming work on the TRUM tank waste project; and Ensure risk mitigation plans are developed in the future that identify project-specific risks and propose appropriate mitigation strategies before initiating projects and resuming the TRUM waste project.b    a SOURCE: GAO, 2005b, p. 27.    b SOURCE: DOE-OIG, 2005, pp. 1-2. wastes that are not covered by a state compliance agreement (radioactive wastes being disposed on-site in burial grounds not covered by state compliance agreements). DOE Order 435.1 is presently the only basis for a waste determination for such waste streams, but an attempt to use it may make the suspended litigation “ripe” (see Chapter II) and open the door for resumption of court proceedings, with the potential for further delays. The concept of “removal of the highly radioactive radionuclides to the maximum extent practical” does not have a clearly defined meaning and, therefore, is open to interpretation and arguments over interpretation.

OCR for page 91
Tank Waste Retrieval, Processing, and On-Site Disposal at Three Department of Energy Sites: Final Report Assessments of whether performance objectives are met can be highly controversial. The choice of point and time of compliance is a decision that has both technical and policy components. This basic fact could render the entire performance assessment outcome subject to controversies, beyond the myriad technical uncertainties and assumptions about human behavior that are inherent in such analyses. These programmatic uncertainties can stop or delay the tank cleanup program just as surely as technical uncertainties can and with the same result—that highly hazardous wastes remain in aging tanks or in the environmental media for a longer period of time, thus increasing present and future risks and remediation costs. The committee is not alone in raising these concerns. The Government Accountability Office in a recent report (GAO, 2005b) expressed concern that DOE’s failure to clarify the legal and regulatory status of the tank wastes could threaten its ability to accomplish its accelerated cleanup plans. In addition, DOE’s Office of the Inspector General recently issued a report (DOE-OIG, 2005) that is critical of the regulatory risk assessment for Hanford’s mixed transuranic tank waste program (see Sidebar VIII-1). As with the tank waste program as a whole, the Inspector General found that DOE failed to identify and mitigate “regulatory and permitting risks” prior to initiating waste management plans, despite its awareness of concerns and controversy, including adverse court rulings, regarding these issues. The result was that DOE had to stop work on the project in midstream (as also happened with the Idaho ruling), and resolution of the tank problem is delayed indefinitely. For waste streams that are to be addressed in the near term, the reasoning that supports waste disposition decisions to be laid out in a manner that is accessible to regulators and stakeholders as part of a transparent, risk-informed, decision-making process. If this reasoning is not laid out properly, there is no way for regulators and stakeholders to evaluate the feasibility of any plans. DOE is taking steps toward this type of process through the waste determinations required by Section 3116 of the 2005 NDAA for the Savannah River Site and Idaho National Laboratory, through the U.S. Nuclear Regulatory Commission consultation process suggested in DOE’s Order 435.1 at Hanford, and at the U.S. Environmental Protection Agency’s urging in relevant environmental impact statements (EISs; DOE-ID, 2005b, response to comments). The committee urges DOE to continue and expand this practice. Some of the programmatic risks to which the committee is referring are not easily evaluated, such as congressional appropriation of money. Some of the risks are probably assessable at a technical and engineering level, such as the likelihood that bulk vitrification would work. The regulatory and legal environment under which DOE operates is complex and contested. Therefore, DOE should acknowledge and account for the programmatic risks in its decision making, because there are legal and regulatory challenges that can— as they have in the past—create major barriers to completion of its mission. DOE needs to recognize these programmatic risks that arise from legal and regulatory challenges, just as it needs to recognize the possibility that bulk vitrification may not work at Hanford or that New Mexico may not allow Idaho transuranic waste to be disposed at WIPP. This is simply a matter of good management. As noted earlier, the committee sees an emerging practice of this kind of analysis by DOE in its waste determinations as required by Section 3116 of the 2005 NDAA. These legal and regulatory difficulties, along with the technical challenges described in previous chapters, increase the programmatic risk that some element of the plan will fail and create problems for other parts of the plan. In the presence of complex technical and regulatory challenges there is the need for a transparent, risk-informed, and participatory decision-making process, as discussed below (NRC, 2005b). The committee recommends that in its planning, DOE identify sources of programmatic risks as soon as possible so that it can seek ways to mitigate or work around them (see Recommendation VIII-2). For waste streams that do not have to be addressed immediately (e.g., calcine at Idaho, pipes and other ancillary tank systems) or for waste streams that are supposed to be shipped off-site, DOE needs to develop programmatic contingency plans in addition to a disposition pathway. RISK-INFORMED, PARTICIPATORY, CONSISTENT, AND TRANSPARENT DECISION-MAKING PROCESS For the past 50 years DOE (or its predecessor agencies) has had the authority and responsibility to manage the high-level waste stored in tanks at the Idaho National Laboratory, the Savannah River Site, and Hanford. During this time, DOE employed available technology and science in addressing treatment, storage, and remediation of waste. As technology and knowledge have evolved and gained in sophistication, the decision making accompanying the use and deployment of nuclear technology and cleanup of wastes has also evolved, becoming more complicated and drawing in parties and partners that were not previously an active part of decision making (see Sidebar VIII-2). States, local governments, Native American tribal nations and other stakeholders ultimately must live with long-term contamination at sites. Therefore, it is important that these parties also be part of the decision-making process concerning site disposition decisions and the associated choices about long-term stewardship (NRC, 2000c). The views of interested and affected parties can have important effects on the way other contextual factors, such as cost and risk, are treated in site disposition decisions. They may influence site disposition decisions at five levels of generality (NRC, 2000c, pp. 73-74):

OCR for page 91
Tank Waste Retrieval, Processing, and On-Site Disposal at Three Department of Energy Sites: Final Report SIDEBAR VIII-2 The Evolution of DOE’s Decision-Making Paradigm The Atomic Energy Commission (AEC) was established by the Atomic Energy Act of 1946 to develop nuclear energy for purposes of national defense. In 1954, revisions of the Atomic Energy Act allowed private industry to participate in the development and uses of nuclear technology for peaceful purposes and gave the AEC regulatory powers in the areas of public health and safety and national security as related to nuclear energy. In response to developing concerns that no single agency should be responsible for promoting and regulating nuclear energy activities, including radioactive waste management and disposal, the Energy Reorganization Act of 1974 replaced the Atomic Energy Commission with the U.S. Nuclear Regulatory Commission and the Energy Research and Development Administration (Public Law 93-438, 88 STAT. 1233). The U.S. Nuclear Regulatory Commission’s primary mission is to regulate nuclear reactors, materials, and waste facilities in the commercial sector. The Energy Research and Development Administration was given responsibility for managing nuclear weapon, naval reactor, and energy research and development programs. In 1977 the U.S. Department of Energy was created to provide a comprehensive national energy plan by centralizing the responsibilities of the Energy Research and Development Administration and other energy-related government programs. Part of the mission of the new agency included nuclear weapons research, development, and production. DOE both managed and regulated these facilities, and although an environmental impact statement was required as part of the National Environmental Policy Act (NEPA) process for major federal actions, the public was provided only limited access to information about federal activities at the sites and input into decisions. In 1984, the Legal Environmental Assistance Foundation (LEAF) sued DOE alleging that DOE allowed unpermitted discharges of nonradioactive pollutants from the Y-12 Plant at Oak Ridge National Laboratory in Tennessee (LEAF vs. Hodel, 1984). In this litigation, the court held that DOE is subject to the provisions of the Resource Conservation and Recovery Act. As a practical matter, states thus gained leverage over some DOE waste management decisions. In the late 1980s as the Cold War was winding down, DOE began to shift part of its mission from weapons production to environmental remediation of the nuclear weapons complex scattered throughout the United States. In 1989, DOE created the Office of Environmental Restoration and Waste Management, later renamed the Office of Environmental Management, to manage the process of environmental cleanup while protecting the health of workers and the public. The office was also responsible for working with a wide range of stakeholders, including states, Native American tribal nations, other federal agencies, interested and affected members of the public, and public interest groups. At each of the major sites, DOE, the host state, and the U.S. Environmental Protection Agency signed legally binding Federal Facility Agreements and Consent Orders that describe the roles of each federal and state agency in meeting specific schedules for environmental cleanup (see discussion of policy background in Chapter II). Native American tribal governments are not parties to these Federal Facility Agreements and Consent Orders and instead turned to the courts as a means to influence DOE decisions that affect tribal peoples, lands, and resources. In addition, public interest groups and other concerned parties have used the courts and the media to try to exert influence over DOE’s waste management decisions. Thus, the decision-making paradigm for DOE and its predecessor agencies gradually changed from operating unilaterally within a self-regulating context of building nuclear weapons and expanding nuclear technologies, to involving multiple federal agencies, states, and other interested and affected parties in the decisions for managing and cleaning up wastes at DOE sites. Stakeholders may help to define risk levels specified in regulations; They may influence priorities about which sites within a facility are addressed first and to what extent (thereby also influencing the management of other waste sites within the facility); They may help specify a desired future state for a site, particularly in terms of its preferred future uses; They may help decide the relative balance of contaminant reduction, contaminant isolation, and stewardship activities to be used in achieving a desired future state for the site; and They may influence choices concerning specific approaches and techniques (e.g., a preference for vitrification over grouting, a desire to have deed restrictions as well as zoning, an objection to the use of on-site incineration). Given the decision-making environment in which DOE operates now—and in the presence of the programmatic risks mentioned earlier in this chapter—the committee recommends a more risk-informed, consistent, transparent, and participatory process, as recommended in a previous National Research Council (NRC, 2005b) report. This process would make DOE’s tank waste management program more robust, in the sense that it is more likely to succeed in its mission, and more transparent, so that regulators, Congress, and the public have a clearer idea of the challenges and choices that DOE faces (see Recommendation VIII-3). The NRC (2005b) report—the Risks and Decisions report— describes the basic elements of a risk-informed approach that is compatible with the needs and legal requirements of this system and is capable of encompassing the nontechnical considerations discussed in Finding VIII-1. The committee authoring that earlier report found that an effective and

OCR for page 91
Tank Waste Retrieval, Processing, and On-Site Disposal at Three Department of Energy Sites: Final Report credible risk-informed decision-making process has the following characteristics: It is (1) iterative and participatory; (2) logical; (3) consistent with current scientific knowledge and practice; (4) transparent and traceable; (5) structured with reasonable independence of the decision authority from the petitioner; (6) subjected to thorough, independent peer review; (7) technically credible, with believable results; and (8) framed to address the needs of the decision process. The Risk and Decisions report defines the term “risk-informed approach” as “[a]n approach in which risk is the starting point but still only among several factors in a decision process” (NRC, 2005b, p. 208). More specifically, Risk and Decisions points out that: Risk assessments (e.g., the scientific evaluation of known or potential hazards) can and should be integrated into risk-informed decision making; Because risk-informed decision making cannot be applied in a cookbook fashion, the risk assessment underlying decision making should account for the complexities and uncertainties of the underlying processes being modeled; Risk analyses carried out to support exemptions from HLW disposal requirements should be logical, well founded, transparent, and traceable; and Risk-informed decision making and the analyses that support it should be structured to inform a specific and well-defined decision based on criteria formulated well in advance of modeling or computation. A full implementation of a risk-informed approach acknowledges that the process of analysis could be more important in achieving transparency, trust, and understanding than an elegant and complicated analysis that is presented as a completed package. Risk-informed decision making thus demands that a suitable process be established and followed. A rote progression through a series of steps would not meet this obligation. Emphasis should be placed on establishing a useful and meaningful process that involves stakeholders from the outset. Using an iterative, staged process brings involved parties along as the complexity and sophistication of the analysis and data increase and is more likely to create confidence in the final disposition decision. In summary, the critical elements for implementing a risk-informed approach are Identify a specific set of options; List the information or data needs for deciding among these options; Establish a set of criteria for determining the best option (before analysis occurs); Carry out rudimentary risk calculations (minimal complexity) to help separate options based on decision criteria; Perform risk calculations that take into account alternative views of physical processes, using ranges of parameter values that reflect the state of the science; Determine which uncertainties affect the ranking of disposition options, and explain them and their significance for review by experts and stakeholders; Make a decision; Monitor; Compare predicted performance with data collected; and Make refinements as necessary through an iterative, staged process. The Risk and Decisions report sets out a six-step process for implementing a risk-informed approach (see Sidebar VIII-3). In the past there have been concerns about the transparency of DOE’s decision-making process. Some of DOE’s decisions concerning tank waste management did not have a clear description of risks, alternatives, or rationale for choices. Some decisions had too little supporting information (e.g., choice of glass as a low-level waste form at Hanford), or conversely, while some had an overwhelming amount of supporting information, that was still incomplete in some important areas and so large that time was inadequate to review it in detail (e.g., the first Section 3116 waste determination at the Savannah River Site for salt waste) so stakeholders did not have the opportunity to understand what was being proposed and why. DOE has better chances to reach its programmatic milestones if it adopts a more risk-informed, participatory, consistent, and transparent decision-making process as described in Risk and Decisions (NRC, 2005b). DOE has taken steps to improve its transparency and detail in its most recent waste determinations (Sams, 2004; DOE-SRS, 2005a; DOE-ID, 2005a) and performance objectives demonstration documents (Buice et al., 2005), which describe how DOE reached its decisions and provide supporting data and analyses making it easier for others to understand. The documents supporting Section 3116 waste determinations are a worthy effort and should be pursued for other waste determinations. The states of Idaho and South Carolina also found the Section 3116 process helpful in reviewing DOE’s waste determination plans because of the greater transparency in DOE’s decision-making process (SCDHEC, 2005a, 2005b; Trever, 2006). The recent Savannah River Site and Idaho National Laboratory waste determinations, and the Hanford exemption for tank C-106 show the beginning of trade-offs among public risk, worker risk, and cost (e.g., Davis, 1998; Gilbreath, 2005; and Sams, 2004). However, the requests for additional information issued by the U.S. Nuclear Regulatory Commission and state comments show that DOE has further opportunities to improve the waste determination process. For example, in the case of the draft Section 3116 waste determination for Tanks 18 and 19, it was not clear whether waste had been removed to the maximum extent practical from

OCR for page 91
Tank Waste Retrieval, Processing, and On-Site Disposal at Three Department of Energy Sites: Final Report SIDEBAR VIII-3 The Six Steps of a Risk-Informed, Participatory, Consistent, and Transparent Decision-Making Process Step 1: Initiate the Process, Laying Out Viable Options and Potential Decisions. This process initiation phase consists of (1) defining the problem and issues; (2) engaging partners and regulators to discuss and refine issues; (3)defining presumptive and alternative disposition alternatives; (4) defining criteria that are relevant to decision making; (5) developing a process plan by which consultation and analysis will proceed; and (6) seeking review and feedback from stakeholders and partners. Step 2: Scope Information and Analysis This step begins the process of estimating the specific risks that will be compared. It includes (1) sketching out the structure of the risk analysis; (2) identifying parameters, datasets, and models required; (3) collecting and reviewing needed information; (4) performing a scoping risk assessment and sensitivity studies to identify critical parameters that require attention; (5) describing data gaps and a data collection plan; (6) conducting review and feedback from experts and stakeholders; and (7) finalizing the work plan and moving forward. Step 3: Collect Data and Refine Models This step is a straightforward implementation of the data collection plan developed in Step 2, as refined by the results of the scoping analysis. It consists of the following activities: (1) collecting quality data that describe the waste and the site; (2) describing and collecting data regarding engineering remedies; (3) refining model logic; (4) disclosing new information collected; and (5) submitting the new data and any new risk calculations to externalreview. Step 4: Prepare a Refined Risk Assessment This assessment involves taking the initial risk assessment (see Step 2) and improving it by applying new data, advanced understanding of the underlying processes (i.e., better modeling), or more sophisticated uncertainty analyses. The refined risk assessment is used to inform the disposition decision. Preparing a refined risk assessment involves the following steps: (1) defining the range of uncertainty by, making use of collected data; (2) conducting analysis, including uncertainty analysis, and producing risk estimates; (3) performing a validity check to make sure that the results are reasonable in light of real world experiences; (4) performing a thorough quality assurance-quality control check of model logic and data inputs; (5) summarizing the results of the risk assessment, paying attention to the risk estimates and the uncertainty; (6) obtaining peer review of the model and its results; and (7) releasing the results to the public, in accordance with the agreed upon plan. Step 5: Conduct Additional Analyses and Data Collection as Needed to Support Decisions If additional analyses or refinements are needed, they would be done following a plan agreed on with stakeholders. It is likely that such planning will result in analyses and data collection being iterative; additional data collection might be necessary as learning progresses and the need for additional analyses is identified. Step 6: Finalize the Decision At this point, DOE has determined that it has sufficient information for decision making and it will use this information to seek final authorization for disposition of the waste stream in question. Tank 19 (SCDHEC, 2005a). The State of Idaho and the U.S. Nuclear Regulatory Commission raised questions about the residual waste characterization of some of the tanks at Idaho and the state of knowledge of the contamination in the sand pads (Trever, 2006; USNRC, 2006). Management of Long-Term Risks and Long-Term Stewardship The committee recognizes that it is not feasible, and does not advocate, removing wastes in tanks “up to the last molecule.” This means that in many cases, sufficient amounts of radioactive materials will remain on-site to prevent release of the site for unrestricted use. In the presence of long-term environmental liabilities, a form of “defense in depth” involves establishing institutional controls (also called long-term stewardship). Nevertheless, the committee believes in the importance of (1) not relying on institutional controls exclusively (in lieu of waste removal or other engineered and natural barriers) and (2) assessing the consequences of the failure of long-term institutional management in light of the uncertainties (e.g., present and future behavior of contaminants in the environment, future developments in society and technology, model limitations).

OCR for page 91
Tank Waste Retrieval, Processing, and On-Site Disposal at Three Department of Energy Sites: Final Report 10 CFR Part 61 provides guidance for long-term stewardship of land disposal sites for low-level waste. The U.S. Nuclear Regulatory Commission (USNRC), as provided in 10 CFR 61.59, does not usually allow reliance on active institutional controls for more than 100 years.4 Therefore, the USNRC requires that DOE demonstrate in its performance assessment that a site meets the performance objectives of 10 CFR 61 and does not depend on active institutional controls beyond 100 years. All three sites rely (at least for portions of the land) on long-term stewardship and institutional controls in perpetuity to help define land use and intruder scenarios. In addressing its responsibilities to reduce risks and protect the environment over long periods, DOE’s long-term stewardship, monitoring, maintenance, and/or solidification or immobilization plans use terms such as “in perpetuity,” “1,000 years,” and “10,000” years. Long-lived man-made physical structures or institutions are extremely rare. A few obvious examples of long-lived man-made physical structures and a long-lived institution are the Egyptian pyramids (approximately 5,000 years old, although most were violated long ago) and the Roman Catholic Church (approximately 2,000 years old), respectively. Very few structures have lasted intact and can fulfill their initial design purpose at even 1,000 years, such as some of the roads, bridges, aqueducts, and amphitheaters built by the ancient Romans. In DOE’s performance assessments reviewed by the committee, the assumption of no loss of institutional control before 100 years is hardly challenged. Although it is impossible to predict what changes will occur centuries from now, it is reasonable to predict that changes will occur, by merely considering the significant changes that have occurred in the past 100 years. The committee acknowledges that DOE is already laying out some of the alternative scenarios in its performance assessments and environmental impact statements (see Chapter VI). An assessment of the consequences of changes in the assumptions (e.g., concerning the effectiveness of institutional controls) used in the performance assessments and an evaluation of the cost, risks, and environmental impact of taking action to mitigate the consequences would be desirable. For instance, if assumptions about land use based on the site being zoned “industrial use, in perpetuity” prove to be incorrect, what would be the consequences in case no action is taken and in case the tanks are re-remediated? When making tank waste determination decisions, DOE must demonstrate that it has removed the highly radioactive nuclides from the waste to the maximum extent practical. This term refers to technical, safety, and financial considerations. As the committee points out earlier in this chapter, the concept of “maximum extent practical” does not have a clearly defined meaning and, therefore, is open to interpretation and argumentation. DOE takes into account the balance between the costs of worker exposure and the effect on risks of removing additional waste from the tanks or in leaving some waste in the tank as the heel. Financial considerations address the costs of retrieving additional waste from the tanks compared to the corresponding reductions in risks while taking into account additional risks to workers to remove the waste. For example, it is more cost-effective to meet performance objectives by using tank isolation and land-use restrictions than by conducting expensive, more complete contaminant reduction measures. However, a future state that includes stewardship is not the same as a future state reached via more complete contaminant remediation, particularly if the latter would allow unrestricted access. For example, it would be very costly, and perhaps not possible, to remediate the Savannah River Site to allow unrestricted access to the entire site given the existing groundwater contamination. DOE relies heavily on limiting the future use of the Savannah River Site to a “high-security mission” (or at least heavy industrial use) as a rationale for not cleaning up many areas where it intends to maintain active institutional controls for the foreseeable future to levels suitable for unrestricted access. In addition to considering the costs of using additional retrieval technologies and protecting workers from additional radiation exposure, other cost elements to consider include the costs of monitoring the site as well as the costs of installing and maintaining engineered barriers and sustaining institutional controls. Thus, a decision to end waste removal operations and grout a tank now may appear economically practical, but when long-term stewardship costs are factored in, that may no longer be the case. A large body of work has been produced on long-term stewardship. The National Research Council published three reports on long-term stewardship relevant to the DOE weapons complex, which includes Hanford, Idaho National Laboratory, and the Savannah River Site (NRC, 2000c, 2003b, 2003c). The main findings and recommendations from the previous NRC committees can be summarized as follows: effective long-term stewardship will likely be difficult to achieve; engineered barriers and institutional controls will eventually fail; great uncertainties remain in assessing the effectiveness of a long-term remediation plan; and no plan developed today is likely to remain protective for the duration of the hazards. 4 “The period of institutional controls will be determined by the Commission, but institutional controls may not be relied upon for more than 100years following transfer of control of the disposal site to the owner” (10CFR 61.59). Note that DOE Order 435.1 has different guidance, “In the intruder assessment, institutional controls should be assumed to be effective in preventing intrusion for at least 100 years following disposal facility closure; longer periods may be assumed with justification (e.g., land-use planning, passive controls)” (DOE, 2001a).

OCR for page 91
Tank Waste Retrieval, Processing, and On-Site Disposal at Three Department of Energy Sites: Final Report When setting up a long-term stewardship program, DOE should Plan for uncertainty; Plan for fallibility; Undertake scientific, technical, and social research and development; and Seek to maximize follow-through using a phased, iterative, adaptive long-term approach, in which monitoring to confirm the performance of the site is a key element. The National Research Council reports urged DOE to look at the very long term in the stewardship of its environmental legacy. Along with the U.S. Environmental Protection Agency and the U.S. Nuclear Regulatory Commission, other agencies (e.g., Department of Defense, Department of Interior) face similar long-term stewardship challenges and could benefit from a common methodology for dealing with them. FINDINGS AND RECOMMENDATIONS Finding VIII-1: Basing tank management and waste disposition decisions only on performance assessment results to demonstrate compliance with performance objectives is inadequate because such assessments do not take into account all of the factors that could be important to decisions such as the evolution of a full-risk profile (risks across the site under different exposure scenarios) over time; compliance with Federal Facilities Agreements; changes in costs and changes in the way people value health and the environment; progress to build confidence in the program; and other site risks. All of these factors become increasingly uncertain as people attempt to anticipate conditions further into the future. Recommendation VIII-1: Site-specific characteristics such as the type of wastes, tank types, and amounts and location of contamination are important to tank waste management decisions. Differences in these characteristics can lead to different choices among cleanup options at each site and even among tanks at the same site. Despite these differences, DOE’s decision-making process should be consistent, even if this leads to different outcomes for different tanks. In addition to performance assessment results, the decision-making process should take into account site-specific factors, including the following: The unique risks created by the proposed decision pathways that are not captured in the performance assessment or performance objectives; Risk considerations at the sites that are altered by cleanup decisions; Combined risks associated with tank residues and other waste streams that are associated with the separation of tank wastes into different waste streams; Considerations not captured in risk analyses or performance assessments, such as costs, distributional or equity concerns, and other societal concerns; The views of the states, Native American tribal governments, and other stakeholders in decision making. Finding VIII-2: DOE operates in an extremely complex and overlapping regulatory and governance structure. Thus, in addition to the technical and health risks that have been detailed to this point in the report, DOE’s conceptual approach to tank cleanup poses a number of “programmatic” risks, that is, risks that the cleanup program will not be carried out as planned. Some of these risks are due to purely technical and engineering challenges (e.g., Will bulk vitrification work?), some are budgetary challenges (e.g., Will Congress appropriate sufficient funds to implement the approach?), some are regulatory challenges (e.g., Will New Mexico accept waste from Idaho and Hanford?), and some are legal challenges (e.g., Will DOE be allowed by the courts to rely on DOE Order 435.1 for its waste determinations with respect to locations and materials to which Section 3116 of the 2005 NDAA does not apply?). Recommendation VIII-2: In its planning, DOE should identify sources of programmatic risks as soon as possible so that it can seek ways to mitigate them. This process will make DOE’s planning much more robust, in that success is more likely in its tank cleanup mission; and more transparent, in that regulators, Congress, and the public will have a clearer idea of the challenges and choices that DOE faces. Finding VIII-3: The National Research Council’s Risk and Decisions report (NRC, 2005b) describes the basic elements of a risk-informed approach that is compatible with the framework and legal requirements in which DOE operates and is capable of encompassing the nontechnical considerations discussed in Finding VIII-1. Recommendation VIII-3: DOE should pursue a risk-informed, participatory, consistent, and transparent approach for making decisions under Section 3116 of the Ronald Reagan National Defense Authorization Act of Fiscal Year 2005 and DOE Order 435.1, which governs tank waste decisions at Hanford. DOE is taking steps toward this approach through the waste determinations required by Section 3116 for the Savannah River Site and Idaho National Laboratory. The committee urges DOE to continue and expand this practice.