A Risk-Informed Approach to Low-Activity Waste Practices
Daily life requires constant assessments of risk: Should I drive or fly? Should I eat the fish or the hamburger? Is it safe to cross the street? Should I sell one stock and buy another? How much life insurance should I buy? Sometimes we have some control over accepting a particular risk, but at other times we do not. Generally, we believe that the risks we have some control over are less dangerous or less likely to occur than those imposed on us by others (Slovic, 2000). More formally, risks are averages; that is, they reflect the average likelihood that an event will occur in a population over a given period of time. Because risks are averages, they are of limited value to an individual. Knowing the average risk of death per passenger mile flown per year, for example, might help us decide between flying and driving, but does little to tell us the risk associated with a particular planned flight.
Risk, and decisions about whether to accept or avoid a particular risk, are determined by the probability or likelihood of an adverse result and by its severity. The likelihood of losing a single roll of the dice in a craps game is 51.3 percent; the severity of the consequence of losing depends on the amount of the bet involved. The larger the bet, the greater is the severity of the consequence and the less likely may the player be to accept it. On the other hand, the larger the bet, the greater is the benefit if won. The player’s decision will be a value judgment that involves weighing potential benefits against potential losses. Similarly, the severity of the loss may become more acceptable as the probability of winning increases. Like gambling, decisions about the acceptability of health risks from environmental exposures to radiation or other hazards depend on the likeli-
hood of an effect, the severity of the anticipated effect, who is affected, and the potential benefit, if any, afforded by the source of the risk.
There is a large volume of literature on understanding and evaluating risk, managing risk, and communicating risk (e.g., Kaplan and Garrick, 1981; NRC, 1983, 1994, 1996, 2005b; Garrick and Kaplan, 1995; Risk Commission, 1997). Publications of the National Academies have consistently advocated using risk as a basis for policy decisions and have provided guidance for characterizing risk, for making decisions about the best ways to manage risks, and for including a broad range of involved citizens in such decisions. In concert with this viewpoint, the National Academies’ Board on Radioactive Waste Management1 initiated this study because its members believed that the present, mainly origin-based, regulation and management practices for low-activity radioactive wastes (LAW) do not provide a consistent basis for systematically managing their risks: “The current systems for regulating this waste lack overall consistency and, as a consequence, waste streams having similar physical, chemical, and radiological characteristics may be regulated by different authorities and managed in disparate ways” (BRWM, 2002). In addition, this committee noted in its interim report that the public expresses “considerable lack of trust in the LAW regulatory system due to its complexity, inflexibility, and inconsistency” … raising “doubts about the system’s capability for protecting public health” (NRC, 2003a, p. 5).
During the course of this study the committee came to the conclusion that a “risk-informed” approach would provide the best option for improving LAW regulation and management practices in the United States. A risk-informed approach is based on information provided by science-based risk assessment but includes stakeholders as a central component in decision making. This chapter begins by discussing the concept of risk and concludes with a framework for risk-informed management of LAW.
RISK AND RISK ASSESSMENT
Risk is a common currency that allows regulators and other decision makers to compare different threats to public health from all sources, set priorities among them, choose effective risk reduction strategies, and target those that are most important. Defining the elements of risk involves the “risk triplet,” a series of three questions posed by Kaplan and Garrick (1981): What can go wrong? How likely is it to happen? What are the consequences or outcomes?
Risk assessment is the practice of using observations about what is known to make predictions about what is not known about the nature and likelihood of a risk. Risk assessment provides a framework for organizing information in a form that is meant to provide a useful input—both qualitative and quantitative—to risk management decision making. The quality of a risk assessment intended to inform disposal decisions will rest on the quality of the available scientific data; on the extent to which the underlying physical, chemical, and behavioral phenomena are understood; and on how well that understanding and any related uncertainties are reflected in the analysis. Because risk assessment is not a way to scientifically or analytically measure risks, but rather to make predictions about potential risks, characterizing and managing risks necessarily rely on judgment and policy in addition to science.
Figure 3.1 shows the National Research Council’s interpretation of the relationship between scientific research, risk assessment, and risk management (NRC, 1983). As the figure indicates, scientific data provide the basis for performing a risk assessment, which in turn provides input to a risk management decision. This framework and the basic risk terminology supporting it have served as the basis for environmental health risk assessment, both regulatory and nonregulatory, since the mid-1980s. The results of a risk assessment are used by regulators and other decision makers, along with information about economics, technological feasibility, politics, and the law, to determine how best to manage a risk.
In the context of protecting public health from environmental exposures, risk assessment involves combining information about hazardous materials of concern (in this case, radioactive materials), the fate and transport of materials in the environment, the exposure of individuals, and the likelihood of adverse health effects associated with these estimated exposures. The result of a risk assessment—the calculated risk—is an estimate of the probability that a particular type of health effect will occur in an exposed population (e.g., 10−6, or one in a million). In practical terms such a result would mean that in a very large population, such as that of the United States, exposed to some harmful agent, an average of no more than one extra person per one million people would be expected to develop an adverse health effect (e.g., cancer or whatever effect was being estimated).
Environmental health risk assessments are generally performed to evaluate risks that cannot be measured. Risks cannot be measured because they are generally quite low. For example, if the risk of excess lifetime cancer from exposure to a given pollutant were estimated to be 10−6, it would not be possible to determine which individuals, if any among the approximately 42 percent of the U.S. population that gets cancer (including treatable cancers such as skin cancer) from all causes, could attribute
their disease to that pollutant (NRC, 2005a). In addition, the information used to estimate risks is imperfect and often uncertain for many reasons. For example, the possible migration of radionuclides from a waste disposal facility, one of several steps in estimating disposal facility risks, is difficult to estimate over time. Other factors that add to uncertainty are duration and extent of possible exposures and understanding the likelihood of responses at low doses. Even large populations may not manifest enough health effects to serve as the basis of statistically robust predictions of the health risks from exposures to very low levels of radiation (NRC, 2005a).
Risk from LAW is a function of the probability that an event will occur with release of radioactive materials into the environment, the probability that an individual will be exposed to those radioactive materials, the duration and intensity of exposure to ionizing radiation associated with the released radioactive materials in addition to those already present in natural background, and the probability that the exposure will produce a response. The recently published National Academies’ report on health effects from exposures to low levels of ionizing radiation2 concluded that there is a finite risk of health effects from any exposure to radiation (the linear no-threshold model), but for low exposures (“doses”) the probability of inducing a health effect is very low (NRC, 2005a).
Properly done, risk assessment can be a powerful tool for organizing information systematically and understanding the behavior and impacts of LAW under a specified set of conditions and assumptions. Risk assessment of LAW provides a link between the properties of the waste that can be measured or at least estimated (e.g., quantity and concentration of radioactive materials, chemical form, half-life, nonradioactive substrates) and possible adverse effects on human health and the environment (risk), which can only be calculated probabilistically (see Sidebar 3.1; also see Sidebar 3.3 of Appendix A). Risk assessments begin with the measurable properties of a waste material and yield an estimate of the risk associated with its dispositioning (release, storage, disposal) under a given set of conditions. Perhaps more importantly, given the properties of a waste and the level of risk deemed acceptable, a risk assessment can provide guidance about the degree of control needed to achieve that level of risk.
A difficulty in performing risk assessments for waste disposal is developing a comprehensive evaluation of all relevant exposure situations and their associated probabilities during the time period under consider-
Radioactivity is defined as the spontaneous emission of radiation from the nucleus of an atom and expressed in units of disintegrations per second (becquerels [Bq]). In principle, these disintegrations can be registered on a detector. In practice, measuring low rates of disintegrations in LAW may not be practical because of interference from natural background radiation or because the emissions are shielded by the bulk of nonradioactive materials in the waste. Typically samples of LAW must be specially prepared to allow detection of radionuclides of interest possible.
Radiation dose is the term applied to the energy deposited from ionizing radiation an individual receives that is emitted from radioactive material or other sources such as X-rays. Radiation dose is expressed in units of sieverts (Sv) (rem in the United States), with the typical effective dose received by an average individual being about 3.6 mSv (360 mrem) per year from all sources (NCRP, 1987). Doses from LAW can occur to workers handling materials. Doses to members of the public are most often estimated from calculations based on hypothetical events that could breach the isolation system (design failures, intrusion) used to isolate LAW. These calculations can include the probabilities of various events occurring over long time periods commensurate with the radioactive lifetime of the waste. Such calculations are often used in probabilistic risk assessments.
ation, which may be hundreds or thousands of years. To the extent that this can be done, the total risk from all credible processes involving the waste disposal system that may give rise to doses to future individuals can be compared with the level of risk deemed acceptable. Because of the long time periods that must be considered, such a comprehensive evaluation is often not feasible. Hence, as noted in Chapter 2, the International Commission on Radiological Protection (ICRP) suggests a two-pronged approach, where likely or representative scenarios are identified and the calculated doses from these scenarios are compared with the dose constraint. The radiological significance of other less likely scenarios is evaluated from separate consideration of the resultant doses and their probability of occurrence.
The relation between radiation dose and health effects has been studied extensively. For the purposes of radiation protection practice, the relationship is assumed to be linear from highest doses that produce acute effects (radiation sickness, death) to lower doses that might disrupt cellular mechanisms and, on a probabilistic or stochastic basis, lead to effects such as cancer. The health risk associated with exposure to low levels of radiation from LAW is calculated using the observed relationship between radiation dose and health effects (principally cancer) at higher doses, using the linear no-threshold model (NRC, 2005a). Such calculations are used for risk-based regulation of radioactive waste.
Factors Affecting Radiation Risk from LAW
Whether a particular LAW poses a health risk is dependent on many factors, including:
The ICRP view is that this approach does not require precise quantification of the probability of credible but unlikely scenarios occurring in the long term, but rather an appreciation of their probability. Other considerations such as the duration and extent of the calculated doses or risks may be taken into account in determining the significance of such scenarios. The point is that the dose constraint should be interpreted as a comparison with exposure scenarios, and in that context, estimating the likelihood of such scenarios is central. This more qualitative ICRP approach is typical of international risk assessment methodologies.
Based on a particular situation, other aspects of radioactive waste management that could be considered as risks include cultural impacts on resources important to different groups (such as Native Americans);
economic impacts (such as devaluation of property); and psychological damage resulting from fear. Perceived risks can lead to economic impacts and psychological harm even when exposure has not occurred.
RISK AND DECISION MAKING
Risk assessment is only one component of the complex decision-making process referred to as risk management. The risk management process includes consideration of other factors such as economics, technical feasibility, social values or preferences, and legal constraints. Of the various potential approaches to making decisions about how best to manage LAW, there are two that explicitly consider risk: risk-based decision making and risk-informed decision making. This committee endorses a risk-informed, as opposed to risk-based, approach to managing risks associated with LAW, as explained in the remainder of this chapter.
Risk-based decision making relies mainly, if not solely, on the results, most often numerical, of risk assessments. Risk-based decision making is a process of deciding whether, how, and to what extent a risk should be managed based on the magnitude of a quantitative risk estimate (USNRC, 1998). As a hypothetical example of the use of risk estimates, if a risk estimate is 10−5 or less, no action may be required; if the risk is between 10−5 and 10−2, some action may be needed to reduce the risk; and if the risk exceeds 10−2, more extreme action may be necessary. Risk-based decision making tends to be a prescriptive framework that typically does not permit much interpretation. Considerations of cost, feasibility, special sensitivities, or the relative importance of the risk in a given setting generally are not part of risk-based decision making. Risk-based decisions are generally made by technical experts without benefit of stakeholder involvement or public consultation.
Risk-informed decision making evolved from early risk-based concepts into processes that are more flexible and not guided solely by quantitative risk estimates (NRC, 1983; USNRC, 1998). In the context of this report, risk-informed decision making could involve economic considerations, social concerns, preferences of affected citizens, and other factors in addition to a numerical risk estimate. It might include consideration of risk-risk trade-offs such as extended storage to allow decay over time. Risk-informed decision making acknowledges that risk assessment is more than a mathematical exercise and should be a decision-driven activity, guided by risk management goals and directed toward informing choices and solving problems (NRC, 1996). A more recent study found that the biggest challenges to developing a meaningful risk-informed decision process are minimizing disruption to existing laws, regulations,
and agreements and enabling meaningful participation by parties who have few resources (NRC, 2005b).
A general framework for environmental health risk-informed decision making was developed by the Presidential/Congressional Commission on Risk Assessment and Risk Management (Risk Commission, 1997). The Risk Commission was established by Congress through the 1990 Clean Air Act Amendments and was tasked to evaluate and make recommendations about the use of risk assessment and risk management across the federal government. The commissioners were appointed by the President, the majority and minority leaders of the House and Senate, and the National Academy of Sciences.
In its 1997 final report, the Risk Commission concluded that a good risk management decision emerges from a process that elicits the views of those affected by the decision, so that differing technical assessments, public values, knowledge, and perceptions are considered (Risk Commission, 1997). The Risk Commission referred to those affected by a risk or a risk management decision as stakeholders, stating:
Stakeholders bring to the table important information, knowledge, expertise, and insights for crafting workable solutions. Stakeholders are more likely to accept and implement a risk management decision they have participated in shaping…. Stakeholder collaboration is particularly important for risk management because there are many conflicting interpretations about the nature and significance of risks. Collaboration provides opportunities to bridge gaps in understanding, language, values, and perceptions. It facilitates an exchange of information and ideas that is essential for enabling all parties to make informed decisions about reducing risks (Risk Commission, 1997, vol. 1 p. 17).
In the case of LAW, stakeholders could include the waste generators, the agencies responsible for regulating waste disposal, the operators of waste disposal facilities and their workers, and citizens residing near waste facilities and waste transportation routes.
The Risk Commission’s framework is shown in Figure 3.2. It is a six-stage process that is circular because decision making should be iterative as new information becomes available. However, the arrow is missing from the last stage to indicate that decision making cannot keep iterating without reaching a conclusion, a process sometimes referred to as “paralysis by analysis.” What makes this framework risk-informed is its other salient feature: placing stakeholders in the center of the decision-making process.
The Risk Commission’s framework has been used in a number of settings, including the strategy developed by a National Research Council committee for cleaning up polychlorinated biphenyls in the Hudson River
(NRC, 2001b). The U.S. Department of Defense, in partnership with various federal, state, and local agencies, relied on the framework to develop and refine a process for assessing and managing risks from unexploded ordnance. Several states have referred to the framework when they wanted to remind the Environmental Protection Agency (EPA) that it needs to be more open and inclusive of stakeholders in its decisions about air toxins (Kelly Rimer, EPA Office of Air Quality Planning and Standards, personal communication to committee member Gail Charnley). The Canadian North-
ern Contaminants Program, a multiagency program led by Indian and Northern Affairs Canada, relies on the framework as a tool to help inform northern communities about local contamination (INAC, 2003). The advantage of the Risk Commission’s framework is that the risk management goals of stakeholders are established at the outset of the decision-making process and are used to provide a context for risk assessment and to guide choices among risk management options.
RISK PERCEPTION, TRUST, AND STAKEHOLDERS
Different stakeholders have different perceptions of the nature, severity, and likelihood of risk. In particular, technical experts’ risk perceptions can vary widely from those of the general public. For radiation issues, risk perception is complex, yet public perceptions about radiation in general and LAW in particular are often important factors in community acceptance of decisions about waste disposal siting and transportation. Most members of the public do not hold consistent perceptions of different types of radiation risks (Slovic, 2000; MacGregor et al., 2002). People see risks from nuclear power or weapons development and production as much higher than risks from X-rays or other medical treatments involving radiation. For example, in a survey of 205 university-age, highly educated adults, a significant number of the respondents believed that exposure to radioactive waste at “even a tiny fraction of current regulatory levels constituted a high or very high risk” (MacGregor et al., 2002, p. 9).
For most citizens, the credibility of and trust in nuclear waste managers is low. In the MacGregor et al. (2002, p. 17) survey, 76 percent of respondents disagreed with the statement that decisions about health effects should be left to experts. Much of the credibility loss for nuclear managers and government regulators has been associated with the Three-Mile Island accident when the public became disillusioned after frequently being assured that nuclear power was safe and that such accidents could not happen (Friedman, 1981), although there were several precursors including the Emergency Core Cooling System hearings in 1972 and the Windscale accident in Great Britain in the 1950s (NRC, 1984). Further losses of credibility occurred in 1988 and thereafter as the media revealed years of mishandled wastes at the nation’s nuclear weapons facilities, and accidental and purposeful radiation releases to surrounding areas (Schneider, 1988).
Such negative coverage continues and reverberates, focusing on costly and drawn-out cleanup efforts, further undermining public trust in governmental or private management of nuclear waste disposal (Friedman, 1991, 2001; Ackland, 2002). Today, the impression (real or perceived) that members of the public were not given the complete truth about their
exposure to radiation and its concomitant hazards is an important source of fear and mistrust. Communicating with the public about radioactive waste issues has been complicated by the public’s lack of trust in those responsible for managing radiation risks and their low credibility.
Risk perception affects trust and the credibility of the risk management decision-making processes. During the early development of risk-based decision making, scientific experts and the organizations they represent dominated risk management. These experts were responsible for estimating risks, and their organizations—often government agencies—were responsible for managing risks that affected the health and safety of individuals. However, as the field matured, the role of experts and technical knowledge in a democracy was frequently debated, particularly in the context of environmental health risk management. The debate centered on conflicts between the “world of values, ethics, politics, and life philosophies” and the “world of information and technical expertise” (Yankelovich, 1991). Scientists were accused of failing to place their efforts in an adequate social context, believing that science is separate from social factors or that social factors play minimal roles (Brown and Mikkelsen, 1990). These differences have been described as technical rationality versus cultural rationality (Krimsky and Plough, 1988).
To include both citizens’ concerns and technical knowledge in risk management decisions, decision-making processes involving communities or others affected by risks were increasingly recommended and implemented. For example, the 1996 National Research Council report Understanding Risk noted that risk management processes must have an appropriately diverse participation or representation of the spectrum of interested and affected parties, of decision makers, and of specialists in risk analysis at each step (NRC, 1996). The report defined “affected parties” as people, groups, or organizations that may experience benefit or harm as a result of a hazard, of the process leading to risk characterization, or of a decision about risk. The report noted that to be considered affected, such parties need not be aware of the possible harm. “Interested parties” were defined as people, groups, or organizations that decide to become informed about and involved in a risk characterization or decision-making process (who may or may not be affected parties).
A difficult situation that arises often in dealing with environmental issues is when the benefits go to one group and risks are borne by a different group—for example, persons living next to an airport. When members of the public feel they get no tangible benefit personally, they are reluctant to bear any perceived risk. It is perceived benefits versus perceived risks that drive acceptability, and this perception differs among stakeholders. Compensation of some kind is possible in some cases. In South Korea, communities that earlier resisted hosting a LAW disposal
site competed for the facility after financial compensation was offered by the government (BBC Monitoring, 2005). Residents of Andrews County, Texas, view the proposed site there in terms of jobs and economic development and are generally supportive of the site (West, 2004).
Despite their common sense appeal, stakeholder-based processes have been criticized for several reasons: the substantial investment of time and resources required; the likelihood that they will heighten, not alleviate, conflict; the difficulty in identifying and facilitating the inclusion of truly representative stakeholders; and the possibility that they are actually counter democratic because of increased involvement of special interest groups (Risk Commission, 1997).
Clashes between the technical and cultural rationalities also draw criticism. Some experts may be concerned that when nontechnical people are included in decision making, the scientific or technical and factual basis of a problem or solution will be distorted, trivialized, or ignored. This problem arises partly because of the difficulty scientists have in communicating technical information as part of stakeholder deliberations and partly because decision makers often perceive nontechnical stakeholders as being more legitimate representatives of social values (EPA, 1995). This clash can also be attributed to nontechnical stakeholders’ beliefs that science can be distorted to support different stakeholders’ points of view. According to one citizen, “sound science is whatever some expert tells you that supports his or her point of view.”
Lack of two important priorities for the sponsoring government agency also can work against successful public and stakeholder participation outcomes and increasing public trust. The first is if the agency initiating the public participation process is not willing (or able) to make the kinds of commitments needed to make the process successful. In such a situation, agency decision makers would not be flexible and open-minded about the nature of the participation and its outcomes. For example, they would not welcome desires of public participants to “redefine problems, focus on different issues, or otherwise change the nature of questions that agencies ask.” The second would be if agency decision makers do not recognize the legitimacy of public values and understand that those values may lead to priorities and conclusions that agency personnel, who have their own understanding of what the public interest is, find wrong. According to a recent meta-analysis of 239 cases of public involvement in environmental decision making that had occurred over the past 30 years, a failure to commit to these two important priorities by an agency threatens the legitimacy of the public participation process and whatever public trust the lead agency may have (Beierle and Cayford, 2002, pp. 63-64).
Another problematic factor relates to who should represent “the public” in these processes. Many times, government planners call on a
relatively small group of people representing various interests groups to act as “proxy for the larger public.” Such smaller groups are needed for a deliberative process, particularly if the issues and goals are complex and require resolving conflicts. However, if the issues involved affect a broad section of the public, then, according to the meta-analysis, broader participation is needed for information sharing and educating the public (Beierle and Cayford, 2002, p. 65). In some complex situations both deliberative and information-sharing activities are required.
Another issue that complicates public and stakeholder involvement is how much influence should be given to the public and stakeholders. Many people agree that the public participation process requires some level of public influence. Yet, in most public meetings, citizens may only provide information and comments and agencies may have little obligation to act on these contributions. According to the meta-analysis, “One of the principal reasons offered for low levels of participant motivation was a perception that the public had little influence over agency decisions.” Such beliefs work against building public trust. The analysis showed that “the goal of incorporating public values, which essentially measures the public’s influence, is highly and significantly correlated with the goal of public trust. In low-trust situations, then, the public may need to be granted more influence to convince them of the legitimacy of the public participation process” (Beierle and Cayford, 2002, p. 68).
Despite these obstacles, there are many ways that governmental organizations, members of the public and stakeholders can and do work together, ranging from traditional public hearings and public comments procedures to policy dialogues, stakeholder advisory committees, citizen juries, and facilitated mediations. The meta-analysis found that among the important factors leading to a successful outcome was a participatory process that starts early in the discussion of an issue. It also found that the type or process of participation is quite important. For example, public hearings and meetings might be quite useful in improving the quality of decisions, but they are not very effective at either resolving conflicts among competing interests or, more importantly, building trust in institutions. The best participatory process for trust building was found to be negotiations and mediations with advisory, stakeholder, or similar types of committees (Beierle and Cayford, 2002, p. 66).
Although it may seem at first glance to take power away from those with legislated responsibility, by involving stakeholders an agency can better identify serious public concerns and be better prepared to deal with them. While in fact the agency retains all responsibility for the decision, if it ignores consensus recommendations of a stakeholder process then future processes are weakened. The committee recognizes that there are some activists who are completely dedicated to thwarting any change and
there is no way they can be accommodated, co-opted, or convinced. Including the appropriate stakeholders to participate in risk-informed decision making is difficult. Still, if a decision is to be made and the agency really listens to stakeholders with the goal of minimizing their irritation and anxiety, there is more of a chance that decisions will go forward and not be mired in controversy.
When risk assessment and risk management are conducted by analysis and open deliberation, scientists and technical experts have an opportunity to interact and work with the public meaningfully by establishing a dialogue about a potential hazard and creating a neutral framework for discussion and collaboration (NRC, 1996). Working together can contribute to increased transparency of the decision-making process, more trust among the involved parties, and some reestablished credibility for managers and government officials who have contributed to the process. Although some involved citizens may remain hostile toward agency decisions, achieving broader acceptance is a reasonable expectation. However, this is not an easy process and it requires genuine respect for public values, careful planning, and a commitment to make the public participation process work from all involved parties.
Sidebar 3.2 provides an example of how improving stakeholder involvement in risk management decisions has helped those decisions to be more reflective of social values and public concerns.
ATTRIBUTES OF A RISK-INFORMED LAW REGULATION AND MANAGEMENT SYSTEM
A risk-informed LAW regulation and management system would combine the principles of risk, risk assessment, and risk-informed decision making and apply them to control LAW according to their actual radiological hazards (see Sidebar 3.1). A waste’s potential to cause harm would guide decisions about its regulation, management, and disposal. These decisions would be further informed by views, needs, and constraints from all stakeholders—recognizing that the current system of origin-based regulation and management is deeply embedded in legislation, the current regulatory framework, and commercial infrastructure. The system would operate according to the Risk Commission’s risk management framework or a comparable framework: involve stakeholders, agree on risk management goals and on a definition of “acceptable risk,” establish a system of LAW classifications consistent with their risks, and identify appropriate disposal options. This section sets out the committee’s vision of a fully risk-informed system. The steps toward implementing this vision, given practical constraints and currently available mechanisms, are developed in Chapter 4.
Native American communities are examples of how stakeholder involvement has led to risk management processes more reflective of social values and public concerns. Until recently, the Native American cultures had little understanding of radioactive material or radioactive waste. Risk assessment and risk communications strategies failed to adequately address the complex issues relating to Native American communities. This failure is of particular concern because Native Americans have been exposed to radiation from fallout during atomic testing, from living adjacent to nuclear waste sites, and from working in the uranium industry as miners, millers, and ore transporters. Yet, policy decisions on how best to manage radioactive waste failed to consider Native American values such as respect, balance, containment, moderation, and reverence. Because Native Americans link their own psychological well-being to environmental stability, the role of psychological healing in decisions about environmental restoration should not be underestimated. More recent efforts to address remediation of radioactive waste-contaminated sites affecting Native Americans have begun to include their perceptions of their environment by including them in the decision-making process.
SOURCES: Dawson et al. (1997); Markstrom and Charley (2001).
At the outset of the committee’s study, the National Council on Radiation Protection and Measurements (NCRP, 2002) proposed a very general risk-based system for classifying essentially all hazardous wastes (radioactive, chemical, biological, or mixtures of these wastes) into categories of (1) exempt waste that could be disposed in a landfill or equivalent facility; (2) low-hazard waste suitable for near-surface disposal in an appropriately regulated facility; and (3) high-hazard waste that requires geologic disposal or equivalent confinement. The NCRP’s general approach and framework served as useful guidance as the committee focused its attention more narrowly and specifically on the attributes of a risk-informed system for LAW. The committee also found international perspectives and initiatives to be useful in developing its views.
In proposing its risk-based classifications, the NCRP developed the following definitions, which are qualitative but nonetheless useful for LAW (NCRP, 2002):
Unacceptable risks are intolerable. Such risks must be reduced regardless of the cost or other circumstances. The NCRP also considers risks unacceptable if they are “tolerable” but not as low as reasonably achievable (ALARA).
Acceptable risks are below intolerable (i.e., they are tolerable) and they are ALARA.
Negligible risks are so low that further risk reduction using the ALARA principle is not warranted. The NCRP notes that achieving negligible risk is not the goal of ALARA. Acceptable risks that are ALARA might not be negligible.
In a risk-informed system, determining what level of risk is acceptable includes nontechnical risk perceptions along with science-based risk assessments and technical and economic constraints. The risk of a given LAW, manifested by its radiological hazards, would be balanced against the control measures applied—including regulation and physical barriers. Determining the balance point—the acceptable risk—is a public policy decision (see Figure 3.3).
While emphasizing its position that determining acceptable risk is a matter of public policy, the committee recognizes the considerable efforts that have gone into developing semi-quantitative guidelines for acceptable risk, often with considerable public input. For example, the EPA uses an upper bound on acceptable risk over the lifetime of an exposed individual of around 10−4 in its Comprehensive Environmental Response, Compensation, and Liability Act cleanup requirements. The ICRP guidelines for “constrained optimization” are equivalent to a risk of about 10−5 per year (ICRP, 1998). The committee does not advocate reassessing these guidelines. Rather, these dose-based risks may serve as a starting point for developing risk-informed practices. Risk-informed regulation and practice would then ensure that all LAW is controlled to a consistent level of acceptable risk. Furthermore, establishing a consistent level of acceptable risk would provide a way to harmonize regulation of LAW with that of other hazardous materials as envisioned by the NCRP.
The risk-informed approach would therefore combine information about the risk arising from inherent properties of a given waste and information about technical, economic, and social issues related to a particular disposal approach in a context of what is considered acceptable. Sidebar 3.3 provides a qualitative illustration of this approach and the basic attributes of a risk-informed system for LAW management.
A waste classification system is necessary for regulating and managing LAW consistently and predictably. For risk-informed LAW regula-
tion and management, the classification system reflects the wastes’ actual radiological hazards irrespective of the wastes’ origins. Waste classification categories would be developed gradually as the decision-making process evolves away from origin-based to risk-informed. As noted previously, current initiatives in the United States that would lead to more disposal options for truly low-activity wastes and impose greater controls on more hazardous wastes are important initial steps in this evolution. International approaches (e.g., International Atomic Energy Agency and European Commission) to waste classification provide good models for a risk-informed system, as does the overarching system envisioned by the NCRP.
The objective of waste management is to isolate hazardous wastes away from humans and the environment. No isolation system is perfect, especially for a very long time period, so risks are unavoidable. A useful way to express the concept of risk is to answer the three questions of the risk triplet (Kaplan and Garrick, 1981): What can go wrong; how likely is it; what are the consequences? The triplet can be expressed mathematically, which forms the basis for the science of probabilistic risk assessment. The following qualitative discussion, however, illustrates the basic concepts of a risk-informed approach to LAW management.
Consider a drum of waste placed in a below-surface vault for long-term isolation. One can list what could go wrong: the multilayered cap leaks, the ceiling or walls of the vault crack, someone intrudes, and so forth. Based on experience or other information, one can also assign a probability to each of these events. Thus, at least to a first approximation, the first two questions of the triplet: what can go wrong? How likely is it? are independent of what is in the waste.
The consequence of something going wrong depends on the nature of the waste: is it flammable, radioactive, pathogenic, or inert? For LAW, risk is related principally to the radioactive properties of the waste.a This is the basis of the committee’s argument that the current system for regulating and managing LAW is based on the wrong premise: controlling the waste according to the enterprise that produced it rather than its actual radioactive properties. Some wastes may be overregulated relative to their actual risk; others that pose equal or higher risks may be escaping the regulatory net.
The mathematical methods of risk assessment allow one to estimate a number for the risk associated with a given system for isolating a given waste. Acceptable risk may, for example, be in the range of 10−6 to 10−4 as discussed in the previous sections. If the estimated risk exceeds that deemed acceptable, then more robust measures to isolate the waste are required.
As discussed previously in this chapter and in other studies, determining the level of risk that is acceptable is a matter of public policy, not purely technical analysis. While not discounting expert analysis, risk-informed decisions explicitly recognize perceptions by all stakeholders, including informed publics. In a facility-siting decision, public officials representing their constituencies may demand a lower risk for a radioactive waste disposal facility than, for example, a chemical plant or a prison, irrespective of the technical assessment. Conversely, decision makers may decide that limited financial resources are better used for reducing risks associated with other activities.
The committee did not attempt to develop its own classification system for LAW because deciding what levels of risk may be considered acceptable, unacceptable, or negligible in particular situations is a matter of public policy. Nevertheless, it is possible—and desirable—to describe the attributes of a risk-informed LAW classification system. To be consistent, any classification system should attempt to balance the intrinsic hazards of the wastes with the extent to which they are controlled (see Figure 3.3). Overall, a risk-informed waste classification system would include the following attributes:
A lower limit for the radioactivity concentration below which the waste could be exempt from regulation or could be cleared from being regulated as a radioactive material (see Sidebar 3.4).3 Such wastes present only a negligible risk to health, so they could be disposed of along with other nonhazardous wastes. The committee has not concluded that such exemption or clearance should necessarily imply release into general commerce—rather, conditional release options such as those being explored
Presently the only way for wastes to exit the regulatory system is via the little-used case-by-case process under 10 CFR 20.2002, as discussed in Chapter 2.
Fly ash from coal combustion typically contains 2 to 10 pCi/gram of NORM, which falls under the regulatory authority of states. In the United States, just over 30 percent of coal ash (14 million metric tons in 2001)a is reused and the remainder is landfilled or disposed of in ponds as slurry. Fly ash is being incorporated into cement products, grouting mixes, and asphalt paving. Applications include highway and runway construction, livestock feedlot and hay-storage pads, and a cellular concrete product that can be used as an alternative to wood in floors and ceiling panels. Decisions about whether or not construction products are labeled with information on radioactive content are the responsibility of states.
Under the current origin-based system for regulating LAW, materials with radioactive material concentrations similar to coal ash that are regulated by the USNRC must be disposed in a facility approved for low-level radioactive waste and cannot be released into commerce except on a case-by-case basis.
by the Nuclear Regulatory Commission’s (USNRC’s) initiatives on alternative disposition of slightly radioactive solid materials would be available (see Chapter 2).
A class of very low hazard radioactive wastes that could be disposed of in appropriately controlled (regulated or permitted) landfills or similar disposal systems. Of particular interest are very low level radioactive wastes from reactor decommissioning and site cleanups considered in the EPA’s Advanced Notice of Proposed Rulemaking (see Chapter 2).
Classes of wastes that are suitable for near-surface disposal but, due to their radioactive material concentration, total quantity of radioactive materials, and other factors that can affect the risk assessment, require increasingly stringent packaging and disposal methods to meet the agreed-upon acceptable risk objective. The USNRC waste classification system in 10 CFR Part 61 is an example of such waste classes (i.e., Classes A, B, C), and the committee does not advocate changing it. As noted previously in this report, however, the Nuclear Waste Policy Act’s definition of low-level waste, which underpins the 10 CFR Part 61 system, is not risk-informed.
Classes of wastes that require special consideration for near-surface disposal in order to meet the acceptable risk objective. Portions of nuclear fuel reprocessing waste that may remain at Department of Energy (DOE) sites constitute one example. These wastes are subjects of another National Academies’ study (NRC, 2005c). Greater-than-Class C wastes, which will be the subject of a future DOE Environmental Impact Statement, are another example.
Classes of wastes that are hazardous to the extent that they require geologic disposal to meet the acceptable risk objective. Most of these wastes have specific definitions under the Atomic Energy Act (AEA; e.g., spent nuclear fuel, high-level waste, transuranic waste). The committee did not discuss these wastes, but includes them here to recognize that they would fit into an overall risk-informed structure (see NRC, 2005b).
It is worth emphasizing that the above attributes of risk-informed categories are independent of the wastes’ origins or their current regulatory designations. Regardless of whether they are naturally occurring radioactive materials (NORM) or AEA wastes, some uranium-bearing wastes, for example, might be discharged with little or no control while others would require disposal in a licensed site with engineered barriers—with the degree of control depending on their radiological hazards. As another example, wastes containing about the same amounts and concentrations of 60Co would be subject to the same requirements regardless of whether they arose in a nuclear power reactor (currently AEA controlled) or in a research accelerator (currently under state authorities). As noted above and throughout this report, there are substantial portions of current U.S. legislation and regulation that are consistent with a risk-informed approach to LAW disposal, as are many current or proposed international standards and practices.
A risk-informed system for regulating LAW—one that regulates radioactive materials based on their radiological hazards rather than their origin—would help simplify and standardize the decision-making process. A simpler, easier-to-understand process will be more open to public scrutiny and participation. The present LAW regulatory system is difficult to understand, even for experts. Because the current system is not easily understood, public participation in decision making is more difficult, which may engender a lack of trust.
A gradual transition from the current regulatory patchwork to a risk-informed system would be an understandable process for all stakeholders and increase their opportunities to participate in the decision making. It
could allow people to know which agency is making important LAW decisions, why these decisions are being made, and which agency is accountable if the decisions are wrong. Citizens could be able to better evaluate public health protection and cost-effectiveness to determine whether enhancements are needed. Reducing complexity could also provide more credibility for government officials, showing that they are taking the problem seriously by looking for a more rational and effective way to manage LAW.