Risk and Decisions About Disposition of Transuranic and High-Level Radioactive Waste (2005)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

3 Exemption Process The description of the high-level waste (HLW) and transuranic (TRU) waste streams in Chapter 2 and the legal standards applicable to them demonstrate that the Department of Energy’s (DOE’s) existing HLW and TRU waste include a wide spectrum of types and amounts of radioactivity, in many physical and chemical configurations. Conse- quently, while the waste streams that the committee examined pose hu- man health hazards that require active management now and into the fu- ture, they pose a range of hazards in the short and long terms. Thus, one might expect that a variety of disposal or disposition options would be appropriate. Current definitions and regulatory standards, however, are inflexible because they treat almost all HLW and TRU waste identically in terms of disposal options; for example, all wastes defined as HLW must be disposed of in a geologic repository designed for permanent iso- lation of the waste. Therefore, in this chapter the committee recommends 1 the creation of a formal, risk-informed exemption process to provide flexibility in disposal options. 1 “QRA [quantitative risk assessment] results are never the sole basis for de- cision making by responsible groups. In other words, safety-related decision making is risk-informed, not risk-based” (Apostolakis, 2004). 61 

62 RISK AND DECISIONS ABOUT TRU AND HLW In this chapter, the committee first provides a summary of the poten- tial adverse impacts of not being able to consider HLW and TRU waste management strategies that are directly informed by risk, based on the more detailed discussion in Chapter 2. The committee then describes and evaluates alternative approaches to ameliorate the adverse impacts while 2 not sacrificing the core values underlying present approaches to manag- ing HLW and TRU waste. This leads to a recommended approach. Fi- nally, the committee provides an outline of the overall process for im- plementing the recommended approach. As discussed in Chapter 1, the committee does not express an opin- ion on the issues currently being litigated concerning the validity of ex- emptions from the HLW definition. The committee assumes, however, that for both HLW and TRU waste a formal exemption process of the kind recommended here will require some kind of formal authorization by the appropriate authority—legislative, regulatory, or judicial. For these purposes, it does not matter whether the legal location for the ex- emption is new language in the definitions, reinterpretation of existing terms such as “highly radioactive” or “sufficient concentrations,” or implementation of existing but infrequently used exemption authori- ties—the exemption process and its characteristics described below could apply to each of these methods. The committee does not recommend that DOE attempt to adopt these changes unilaterally, either through the classification system or by other means. Unilateral action seems likely to exacerbate the sense of mistrust that has developed between DOE and at least some of the parties that are its partners in seeking site cleanup. The exemption process that the committee recommends must also be implemented in the context of DOE’s existing or renegotiated compliance agreements. While the com- mittee is aware that there is a widely held view that these agreements contain economically or technically infeasible requirements, the commit- tee strongly cautions against any unilateral abrogation by DOE of any existing agreements. The present classification system is the basis for several such agreements; thus, making the case for an exemption is the necessary prerequisite to any changes—including disposal options—in those agreements. Rather, to the extent that DOE can obtain the consent 2 These include, most importantly, adequate protection of human health from radiation hazards, protection over the long term (approaching the period during which most radioactive materials remain hazardous), and accounting for the uncertainties in long-term waste management. 

EXEMPTION PROCESS 63 of its compliance agreement partners to deviate from its current obliga- tions, a risk-informed exemption process, as described below, is the committee’s recommended approach. Put another way, if DOE wants to renegotiate its compliance agreements, it must make a case for renegotia- tion that is informed by risk, sets out clear criteria for an exemption, comprehensively addresses health risks (including worker, transporta- tion, and long-term risk), and follows a transparent process that allows meaningful public input. Finally, the committee’s recommendation of a process for exemptions or exceptions should not be taken to mean that any particular waste stream is suitable for exemption; rather, exemptions must be decided on a case-by-case basis, through the process described herein. 3.1 THE PROBLEM: INFLEXIBILITY IN PURSUING APPROPRIATE MANAGEMENT STRATEGIES The existence of inflexibility concerning disposal strategies for HLW is based on the committee’s understanding of the wastes that were exam- ined, the capabilities of current technologies, and the present regulatory system as interpreted by the courts. The Problem of Inflexibility The need to manage all HLW by disposal in a permanent geologic repository and the strong preference for disposing of TRU waste in the same manner represent extremely expensive options, and, indeed, push or exceed the limits of technical feasibility in some cases. Requiring re- trieval and permanent geologic disposal also entails additional risks from the retrieval work and transportation, for example, that might be quite high. For these reasons, as described in Chapter 2 and above, DOE has in the past sought and obtained from the U.S. Nuclear Regulatory Commis- sion (U.S. NRC) informal or ad hoc exemptions from the definition of HLW to allow for near-surface disposal of HLW and from the U.S. Envi- ronmental Protection Agency (U.S. EPA) to allow near-surface disposal of TRU waste. DOE now seeks the advice of the National Research Council on a risk-informed approach to management of these wastes. All rules generalize, which means that all rules create problems of over- and underinclusiveness. That is, they include some regulated items that probably should not be in a category and omit or exclude others that 

64 RISK AND DECISIONS ABOUT TRU AND HLW should be included. In fashioning a rule, therefore, the question is not whether to generalize, but how much over- and underinclusiveness to tolerate as the cost of simplicity. Like all rules, an inherent part of the definitions of HLW and TRU waste is that they are under- and overinclu- sive with respect to particular waste streams. For example, the definition of TRU and the consequent requirement of permanent geological disposal may be overinclusive with respect to some buried TRU waste in very arid and controlled environments where radionuclide migration is expected to be small as a result of the absence of water and intrusion not being expected because of institutional con- trols and minimal pressures from human activities. Conversely, it may be underinclusive, at least with respect to human health risk, as compared to high-activity, long-lived low-level wastes. An example might be the waste storage silos at Fernald, in which radon-226 levels exceed 100 nanocuries per gram (nCi/g). The so-called K-65 waste resulted from processing of exceptionally rich uranium ore that was obtained during and very shortly after the Second World War. Although radium is not transuranic, the K-65 wastes produce a substantial external dose due to gamma-ray emission and the risks they pose may even exceed those posed by some transuranic wastes and are at least similar based on the intrinsic toxicity of the isotopes involved. Under- and overinclusiveness are in the nature of all rules of general applicability. The drafters of general rules are rarely able to anticipate all of the circumstances to which they will apply, and new circumstances often appear that make previous assumptions obsolete. Moreover, gen- eral rules often build in under- or overinclusiveness as a way of making a point of principle or as a means of simplifying administration of and compliance with the rule. The question, therefore, when over- or under- inclusiveness is found is not how to eliminate it, but how to manage it. Managing Over- and Underinclusiveness in Regulation There are two decision points that create the inflexibility in management of HLW and TRU waste. The first is the classification decision, in which the waste products of certain processes and materials with certain atomic compo- sition are automatically designated HLW and TRU waste, respectively, with little or no real opportunity for adjustment based on other factors, such as risk or hazard. The second is the disposal decision, about which the legal structure is extremely prescriptive: if the waste is classified as HLW or TRU waste, it must be disposed in a geologic repository unless, in the case of TRU waste, 

EXEMPTION PROCESS 65 an exemption is applied for and approved by the U.S. EPA. The result is a highly inflexible system, as described in Chapter 2. Logically, one could cre- ate flexibility in one or both of two places: the classification decision or the disposal decision. The committee has chosen to consider flexibility in the classification decision only, for four reasons: 1. It more readily accommodates the kind of exemption that the committee recommends. 2. Classification has been for some time the focus of concern with the disposal provisions for HLW and TRU waste. 3. Such an approach requires the least change in the fundamental structure of the existing HLW and TRU waste disposal require- ments. 4. Most importantly, it poses a familiar regulatory problem of over- inclusive categories, which has familiar solutions. Additionally, if the focus is the disposal decision, the inevitable re- sult of an approved exemption would be the existence of at least two types of HLW—for example, “near-surface” HLW and “deep-geologic” HLW—which is sure to result in confusion. It would be a mistake, how- ever, to see the two areas of flexibility as entirely separate. As is plain in the discussions of the exemption process in Section 3.2 of this chapter and of a risk-informed process in Chapter 4, flexibility in classification will require careful consideration of the characteristics, especially as they relate to risk, of alternative disposal options. The existing system for managing HLW and TRU waste is inflexible because it is based on a system of categories with predetermined conse- quences. It is a common, if not universal, characteristic of regulatory sys- tems that they establish certain categories at the outset and then specify the different consequences that flow from being placed in one category or another. This is a conceptually simple and administratively manage- able way to apply general rules to variegated reality. Moreover, since regulators are not omniscient, it is a way to limit the universe of consid- erations to those for which the information is obtainable at a reasonable cost (Karkkainen, 2001). It is also common, if not universal, that catego- ries are over- and underinclusive in the sense that they cover either more or fewer—or, often, both—instances than their rationale would indicate should be covered. 

66 RISK AND DECISIONS ABOUT TRU AND HLW There are a few generic ways to handle the problem of over- and un- derinclusiveness. First, one can choose a classification system that is more precisely tuned to the rationale for the differential treatment that flows from the classification. The more fine tuned a classification system is, however, the more information it requires to make classification deci- sions and hence the more expensive and time-consuming it is to operate. Second, one can accept overgeneralizations as the cost of an easily un- derstandable and predictable regulatory system. Third, one could retain the general classification, because of its predictability and ease of ad- ministration, but permit exemptions where the classification demonstra- bly fails to fit the situation. This option can be subdivided into two cases: (1) exemptions can be granted on an ad hoc, highly discretionary basis, or (2) exemptions can be granted according to well-defined, preexisting standards. 3.2 OPTIONS FOR ALLOWING NON-GEOLOGIC DISPOSAL OF HLW AND TRU As stated above, the adverse impacts of over- and underinclusiveness can be handled or at least moderated in four ways: 1. create new categories, 2. accept existing categories, 3. permit ad hoc exemptions, and, 4. establish formal standards for exemptions. In this section, these options are discussed mainly as they apply to HLW, because no valid exemption currently exists for some HLW (a valid exemption does exist for TRU waste, as discussed above). The con- clusions, however, are fully generalizable to TRU waste, as described below. Option 1; Create New Categories: Establish a Generally Applicable Risk-Based Definition of HLW One obvious solution to some of the shortcomings of the existing waste classification system is that the current definition of HLW could 

EXEMPTION PROCESS 67 be replaced with a basic provision related to the factor of greatest con- cern—human health risk—plus a provision for case-by-case exceptions. Structuring the definition in this manner would make it similar to that of TRU waste. Such an approach would be the logical outgrowth of a gen- eral regulatory system that emphasizes human health risk as a measure across numerous areas of environmental, safety, and health regulation. Systems based solely on risk per se—that is, systems that are finely tuned to the fundamental reason for the rules—are inherently costly to implement because they require that each instance be analyzed in detail to determine precisely how to apply the rule to it. Moreover, such screen- ing requires more detailed information; so the data demands of a fine- tuned system are high, often challenging and sometimes exceeding the available information. A system based on risk per se would likely exhibit this drawback, not so much because of the toxicity component of the risk equation (which for radionuclides is relatively well known), but because of the exposure component, especially for long-term exposure scenarios. In addition, with a wider range of possible classifications (no longer in- out or yes-no), the outcome is far more difficult for either regulators or regulated entities to predict. Therefore, the committee does not recommend the implementation of a purely risk-based definition of HLW as an alternative approach for management and disposition of TRU and HLW at this time for the fol- lowing reasons: 1. A high degree of reliance has built up with the existing system— disposition decisions, treatment methods, investment in infra- structure (in the billions of dollars), compliance agreements, and so on. Many of these agreements have been the subject of litiga- tion and judicial interpretation. An entirely new system would place all of these agreements and decisions—many or even most of which have not been completely fulfilled—in question, result- ing in chaos in DOE’s cleanup program. Whatever efficiencies would be gained by changing the management of a few, special wastes could be outweighed by the loss in efficiency of reopen- ing dozens of completed decisions, many of which required years of data collection, analysis, negotiation, and public in- volvement. 2. As has been amply demonstrated with the existing definition, changing the definition of HLW could have other unanticipated consequences. Thus, such revision is not advisable without 

68 RISK AND DECISIONS ABOUT TRU AND HLW detailed and comprehensive study to develop a reliable and defensible definition and a large amount of political will to put it in place. 3. It has not been demonstrated that revising the definition of HLW is worth the effort. Such an endeavor is worth pursuing only if a study to identify waste streams that might benefit from a thor- oughly revised definition of HLW shows that the adverse impacts of the existing HLW definition are sufficiently pervasive so that the effort is warranted. 4. To a significant extent, the issue that has to be addressed is mak- ing a site cleanup decision (e.g., what amount of “heel” can be left in a tank) not a waste classification decision. While waste classification and site cleanup decisions are tightly coupled, the classification of waste as such is not directly relevant to potential risks from various cleanup alternatives. Alternatively, one could adopt a system that, even though it was not based on risk per se, was based on a closer surrogate of risk. However, even a HLW definition more closely based on risk—for example, one based on the level of radioactivity—would be imperfect. It would replace one generalization—method of production—with a better but still impre- cise surrogate of risk. It is a surrogate, because risk depends on the de- tails of the radioisotope (its half-life and the type and energy of radioac- tive decay), the physical and chemical form of the specific waste stream, and disposal path. Consideration of risk itself requires consideration of exposure, which requires consideration of the actual disposal environ- ment, which includes a consideration of the hydrology, geochemistry, population distribution, and so forth. Whereas, all in all, such a surrogate may now be considered a better proxy for what people most care about, it is still incomplete and susceptible to over- and underinclusion. Such situations could presumably be handled by adding exemption provisions similar to those in the definition of TRU waste. Wholesale reclassification, however, would pose several dangers of its own. It could create its own set of unintended and irrational results due to unforeseen circumstances. No choice of words perfectly matches the exter- nal world they describe. So it is likely that new loopholes and ambiguities would creep into a wholly new system. Moreover, as noted above, substan- tial parts of DOE’s environmental management program depend on agree- ments and decisions based on existing categories. As a result, it is clearly 

EXEMPTION PROCESS 69 appropriate to consider the second option, that is, to accept an imprecise system built on generalities as the price of predictable (if inefficient) out- comes and an administratively manageable system. Option 2; Accept Existing Categories and Retain the Current Process-Based System Theoretically, it might on the whole be easier to work with the cur- rent process-based definition of HLW. Although risk may be the ultimate benchmark, the generating process may be sufficiently closely linked to risk that the additional effort and unpredictability may not justify the change. It is easy to dismiss the administrative costs of a classification system, but a project of the magnitude of DOE’s cleanup consumes hun- dreds of millions of dollars each year in administrative costs. In addition, predictability is not simply a question of the ability of regulators and regulated entities to plan; it is also some assurance of fairness and even- handed treatment of the regulated entities by the regulators—a clear 3 definition treats everyone equally in a formal sense. Moreover, overin- clusiveness can be a hedge against the uncertainties inherent in the gen- eral definition; that is, can err on the side of protection. The committee declines to follow this route, however. For reasons stated in Chapter 2, the general definitions of HLW and TRU waste may well be too rigid, at least for some waste streams, so some effort to be more precise is probably “worth it” in terms of administrative complex- ity. Indeed, arguments for administrative efficiency and equity lose some of their force in the particular DOE context. The sheer vastness of the project and the nature of the wastes make it one of a kind. DOE is the only party responsible for this work, from generation to management and disposal—this is not an industry of thousands of licensees. Also, HLW and TRU waste constitute a relatively limited universe of wastes, and DOE already spends substantial resources in administering them, so the ability to make generic decisions may be less important here than in an ordinary regulatory system. 3 Certainly, formal equality does not mean equal impact. Formally equal treatment can have intentionally or unintentionally disparate impacts. 

70 RISK AND DECISIONS ABOUT TRU AND HLW Sidebar 3.1: Unintended Consequences of Environmental Policy: The Case of Brownfields A good example of unintended consequences can be found in the Superfund program under CERCLA as it applies to the environmental cleanup of former industrial property in urban areas—“brownfields.” The broad liability provisions were intended to speed cleanup efforts, to as- sign cleanup costs to industries that are generally responsible for the environmental harm, and to provide a deterrent to future environmental sloppiness. One consequence of the liability scheme, however, was that industrial and commercial developers avoided the reuse of existing in- dustrial areas, even though such areas had preexisting infrastructure and other advantages, because they feared Superfund liability. As a result, urban industrial areas decayed as the industrial base changed, and new industries were built in suburban and rural “greenfields,” destroying prime agricultural land, requiring new roads and highways, and contributing to the general blight of urban centers. Recent legislation has attempted to reduce the brownfields effect to remedy the unintended consequence, by providing exemptions from or limitations on liability for specific, defined formal industrial (brownfields) sites (BRERA, 2002). Back-End Adjustments: Ad Hoc Exemptions or a Formal Exemption Process The third and fourth options involve an exemption or “back-end ad- justment” process for obtaining relief from the disposal requirements of 4 the HLW definition (Shapiro and Glicksman, 2003). Exemptions— which come in many forms—are a means of making case-by-case ad- justment to general policies, because there was limited understanding of the full implications at the time a general rule was adopted, new informa- tion became available, or circumstances have otherwise changed. It is the nature of “bounded rationality” that human beings cannot know the full implications of regulatory actions they take at the time they take them. They recognize “that formal rules are unlikely to capture the infinite va- rieties of empirical reality and that increased flexibility in the rulemaking 4 The discussion of exemption processes generally draws heavily on Shapiro and Glicksman (2003, pp. 158-77). 

EXEMPTION PROCESS 71 process is necessary.” 5 Without such flexibility, the regulatory action is often undesirable or even unacceptable because it is unnecessarily ineffi- cient, unfair in application, or even inconsistent with the basic purpose of the rule. Additionally, sometimes following one goal too far can conflict (in unexpected ways) with other goals (see Sidebar 3.1). A familiar example of an exemption process is “delisting” of materials classified as hazardous wastes under the Resource Conservation and Re- covery Act (RCRA). U.S. EPA maintains a broad classification to sweep a broad range of materials into the hazardous waste management system be- cause, following the clear directive of Congress to be protective, it wants to err on the side of overinclusion (regulating wastes that are not very hazard- ous) instead of underinclusion (missing truly hazardous materials). A broadly inclusive classification system also permits U.S. EPA to obtain a Sidebar 3.2: Variances: An Example From the Clean Water Act A good model for exceptions is Congress’ response to a decision of the Supreme Court that ratified a measure of discretion that the U.S. EPA claimed for granting exemptions for toxic water pollutants (CMA v. NRDC, 1985, 470 U.S. 116). Under the Clean Water Act, the general standards for most pollutants can be varied on a number of grounds. For toxic water pol- lutants, however, Congress rejected such variances. The U.S. EPA never- theless permitted a certain type of variance—for “fundamentally different factors” (FDFs) of production method. The idea is that variances for cost, for example, should not be allowed to undermine the protections against toxic water pollutants, but that the variance based on the technology used by the polluter was appropriate for technology-based standards. The U.S. EPA had created these variances on its own, without congressional guid- ance, and environmental groups and Congress worried that it was uncon- strained. So Congress amended the Clean Water Act to recognize the need for FDF variances, but it laid out a process and the precise conditions for obtaining them: Any variance must be (1) technically rigorous, (2) democ- ratically responsible, and (3) considerate of the long term (Clean Water Act § 304(n)). 5 Breger, quoted in Shapiro and Glicksman (2003). 

72 RISK AND DECISIONS ABOUT TRU AND HLW better understanding of the overall hazardous waste problem. However, some wastes just are not as dangerous as initially thought or changes have been made in the process generating the waste. Adjustments can be made in two ways: as ad hoc exceptions to the general rules (the third group), or as a structured process (the fourth) (see Sidebar 3.2). Option 3; Permit Ad Hoc Exemptions: Allow Informal Exemptions Ad hoc exceptions constituting the third option are very common in regulatory systems. Exercising enforcement discretion, delay in en- forcement, deadline extensions, informally agreed upon changes, or sim- ply ignoring legal requirements are ways to avoid the application of gen- eral rules to unanticipated or ill-fitting circumstances. Some of these ex- ceptions are unofficial, back-room deals, not designed to see the light of day. Others are unavoidable—for example, the kind of discretion that police officers and prosecutors exercise on a daily basis—but are ex- tremely situation-sensitive and not susceptible to generalization. Con- gress has occasionally resorted to this technique with appropriations rid- ers—sometimes inserted into complex or essential legislation—designed to excuse a particular project, for example, from the strictures of general rules. Whatever their status, however, ad hoc exemptions all suffer from certain drawbacks. They can be lawless, even if legal, in the sense of subverting the general intention of the existing rules. At the extreme, too many exemptions can swallow the rule. If the exemptions are handled in a carefully documented, public process, the need for a multitude of ex- emptions can be a warning sign that the general rule is dysfunctional. A secretive or uncontrolled system simply undermines the general rule and is subject to abuse. Similarly, ad hoc exemptions do not express a coher- ent policy, even a policy for exemptions. They can be random, lack a firm factual basis, be inconsistent or perverse, or even reflect favoritism and agency capture. Exemptions can create their own uncertainty. Fi- nally, they can “fly under the radar screen,” avoiding public participa- tion in important and difficult public policy choices. DOE and the U.S. NRC have sought to use what are, arguably, ad hoc variance procedures to deal with some of the waste streams that the committee is considering. Under its self-regulation powers, DOE sought to create a process whereby it could declare some waste from HLW tanks to be “waste incidental to reprocessing” (DOE Order 435.1) that would not require permanent geologic disposal. This was overturned by a fed- 

EXEMPTION PROCESS 73 eral court in Idaho as lacking legal authority and meaningful standards, although the ruling itself was overturned because the court of appeals deemed that the issue was not yet ripe (see Section 1.2). Even without the Idaho ruling, actions under the DOE order would have suffered from the lack of credibility that often attends self-regulation: lack of a public process, lack of clear standards, and lack of public confidence in the de- cision maker. This variance does not have the kind of sustained technical basis that would be expected in a regulatory initiative with a complete administrative record. The effect is that the existing system is both tech- nically and democratically indefensible. Option 4; Establish Standards for Exemptions: A Formal Exemption Process The fourth option, and the one the committee endorses, is a formal, defined, and transparent exemption system that uses risk to human health as a basis or starting point. A clearly defined exemption process with clearly defined standards can help avoid irrational or counterproductive results, while at the same time retaining the basic efficiency of general rules. It can “accommodate unique or anomalous situations without sac- rificing regulatory objectives” (Shapiro and Glicksman, 2003). Having exemptions is consistent with protective regulation, as long as the ex- emptions are limited to situations that have previously been defined as exceptional, are different from the general rule (i.e., truly exceptional), and are consistent with the level of protection that Congress and DOE have properly and previously selected. A well-defined exemption process has two other benefits. It provides an incentive for the regulated entity to generate reliable information to justify deviation from the strict baseline rule. Also, if there turn out to be too many meritorious applications for adjustments, this provides a useful warning that the baseline rule itself must be fixed. None of these benefits, however, are provided by ad hoc exemptions, because they do not entail limitations or require analysis. As noted above, delisting provisions for RCRA provide one model for an exemption process. Under the statute, a solid waste is subject to the strict management provisions applicable to a hazardous waste if the solid waste either (1) exhibits particular hazardous characteristics, or (2) is listed as a hazardous waste. The technique of listing is itself an exam- ple of the advantages and disadvantages of general rules that have al- ready been discussed. It proved to be extremely burdensome for regu- lated entities and the regulator to test each of the thousands upon thou- sands of industrial sources of solid wastes for hazardous characteristics; 

74 RISK AND DECISIONS ABOUT TRU AND HLW therefore, U.S. EPA was authorized to list certain waste streams based on certain constituents or certain processes to simplify administration and predictability. It was always clear that such listings would be overinclu- sive in the sense that they would cover waste streams at particular sites that were not as hazardous as the generality of such wastes. Therefore, Congress provided an exemption process, known as delisting, to remedy the inherent imprecision of the listed waste categories. The delisting process is set out generally in the statute, 42 U.S.C. § 6921(f) and de- scribed as follows in U.S. EPA’s RCRA Orientation Manual (U.S. EPA, 2003): The RCRA regulations provide a form of relief for listed wastes with low con- centrations of hazardous constituents. Through a site-specific process known as delisting, a waste handler can submit to an EPA Region or state a petition dem- onstrating that even though a particular waste stream generated at its facility is a listed hazardous waste, it does not pose sufficient hazard to merit RCRA regulation. For example, a waste generated at a specific facility may meet a list- ing description even though the process uses different raw materials than EPA assumed were used when listing the waste, thus the waste may not contain the contaminants for which it was listed. Similarly, after treatment of a listed waste, the residue may no longer pose a threat to human health and the envi- ronment. Specifically, the petition must demonstrate that the waste does not: • Meet the criteria for which it was listed • Exhibit any hazardous waste characteristics • Pose a threat to human health and the environment by being hazardous for any other reason (e.g., does not contain additional constituents that could pose a threat). If the EPA Region or state grants a delisting petition, the particular waste stream at that facility will not be regulated as a listed hazardous waste. This process contains the key features, also discussed in the text, of criteria for application and constraints (single facility, factors for listing), burden on applicant, relative difficulty of obtaining approval (factors in addition to listing), and a designated decision maker. In this report, the committee uses the term decision maker to refer to the entity with final approval authority, i.e., the regulator. 

EXEMPTION PROCESS 75 Reasons to Establish Exemption Standards for HLW Finding 5: Without a formal, well-structured, decision-making proc- ess, less desirable, ad hoc approaches will emerge. Given the costs and difficulties of sending all waste that could be classified as HLW or TRU waste to a deep geologic repository, some approach will arise for deciding what waste requires geologic disposal and what does not. A formal and well-defined exemption system offers several specific benefits for the management of HLW. First, it is what might be called a minimalist approach; that is, it avoids root-and-branch change in the basic system for regulating HLW. DOE does not challenge, to the committee’s knowledge, the general policy of permanent geologic disposal for HLW (and TRU waste), and it has not yet been shown that the adverse impacts of the current definition are sufficiently pervasive to justify a recommendation to modify the basic definition of HLW. Impor- tantly, Congress has a long-standing commitment to permanent geologic disposal as the method of choice for HLW and TRU. At a minimum, therefore, established public policy is that permanent geologic disposal is the starting point for disposal decisions. An exemption system respects this established public policy: it accepts the starting point and then devi- ates from this only where it can be justified. A baseline rule of geologic disposal with defined exemptions would therefore continue to signal Congress’ firm commitment to, in this case, strong protection against radiation hazards. Considering the characteristics of the waste itself, the three wastes identified by the committee suggest, on the basis of a relatively cursory examination, the utility of alternatives to permanent geologic disposal for some waste forms, because they exhibit the following characteristics: • They appear to be relatively low in radioactivity, and/or hazard compared to other HLW and TRU waste that DOE manages, and perhaps could be managed in some manner other than disposi- tion in a deep geologic repository. • For some of these wastes it is infeasible to recover and dispose of every last bit of waste that might conceivably be classified as TRU or HLW. 

76 RISK AND DECISIONS ABOUT TRU AND HLW • The effort, exposures, and expense associated with retrieval, im- mobilization, and disposition in a repository may be out of pro- portion with the risk reduction achieved (if any). An alternative to permanent geologic disposal, therefore, might be preferable for these waste types. The bases listed above provide criteria for considering an exemption. The criteria for granting an exemption must be developed through the exemption process. Procedurally, a formal exemption system provides transparency and discipline to exceptions. There is some history of ad hoc exemptions, and the committee heard concerns from stakeholders that this undermined their confidence in the safe management of these wastes. Moreover, without a default or presumptive disposal method, regulators and the regulated cannot know or predict how to manage these wastes. What is the tipping point between permanent geological disposal and near- surface (or other) disposal methods—a certain half-life, a certain level of activity, a certain level of risk? If so, who will set that bright line? These are rhetorical questions, of course. Either Congress or an agency must do this in some very generic way, or the system would be totally flexible and give no guidance. Finally, one of the strengths of an exemption sys- tem is that it does not require detailed review of every tank or fraction of waste stream, but only those that initially appear to be exceptional (in the sense of inappropriate to the general rule). As described above, establishing criteria for exempting some HLW from the requirement of geologic disposal could proceed from either of two starting points: (1) reinterpreting key phrases in clause (A) of the current definition of HLW (e.g., “highly radioactive” or “sufficient con- centrations,” see Section 2.1), or (2) from a “blank slate” in the form of an integrated and consistent set of technically defensible conditions that must be met for a particular waste to be exempted. The committee favors the second approach for two reasons. First, recent litigation made it clear that the definition of HLW is, if not ambiguous, at least capable of dif- ferent interpretations by different parties. It seems unlikely, therefore, that unambiguous agreement could be reached on inherently vague terms such as “highly” or “sufficient.” Second, it is clear that the key phrases in clause (A) of the current definition (see Section 3.1) do not provide an adequate technical basis for defining exemption criteria without a sub- stantial amount of administrative elaboration, for which the statute gives little guidance. In sum, the committee judges that a clear set of techni- cally defensible, formalized standards and well-defined procedures for 

EXEMPTION PROCESS 77 implementing them would be the best way to avoid future legal disputes of the kind that resulted in the recent Idaho litigation. Facilitating TRU Waste Disposal Exemptions As discussed previously, the current definition of TRU waste in- cludes a provision for exempting wastes from the requirement for deep geologic disposal on a case-by-case basis. However, this provision does not appear to have been used almost at all, although there are instances in which DOE clearly must seek exemptions to follow through on its plans. The committee believes that DOE would benefit from using the criteria and process described in the remainder of this report if it applies for such exemptions in the future. In other words, an exemption system is already provided for in the case of TRU waste, so the need is not to create an exemption, but to provide a transparent process and clear, risk-informed criteria. 3.3 AN EXEMPTION SYSTEM The committee offers the following structured exemption system as one that accommodates the considerations discussed above. The system has four parts: the bases of permissible exemptions, the criteria for grant- ing exemptions, the burden of proof, and the process for obtaining ex- emptions. Bases for Exemptions To begin with what an exemption system should not be, exemptions should not be available simply for DOE’s convenience, or simply to save time or money. Nor should they undermine either the classification sys- tem or the preference for permanent geologic disposal embodied in pre- sent laws and regulations. The first step, therefore, is to identify the rea- sons for which an exemption may be sought (see Sidebar 3.3). Based on 

78 RISK AND DECISIONS ABOUT TRU AND HLW Sidebar 3.3: ARARs CERCLA provides an example of a detailed exemption process. As discussed in Section 1.2, cleanup remedies selected under CER- CLA must meet, among other things, any “applicable or relevant and appropriate requirements” (ARARs) of federal or state law (42 U.S.C. § 9621(d)(2)). In some cases, however, compliance with ARARs is im- practical or counterproductive, so the National Contingency Plan (U.S. EPA’s regulatory blueprint for remedy selection) permits ARAR waiv- ers in carefully specified circumstances (40 CFR § 300.430(f)(1)(ii)(C)) and by following specified procedures (40 CFR § 300.515(d),(f)). Of the permitted reasons for an ARAR waiver, three are of direct rele- vance to the difficulties of managing the HLW and TRU waste streams discussed in Chapter 2: • Compliance with the requirement will result in greater risk to hu- man health and the environment than other alternatives, • Compliance with the requirement is technically impracticable from an engineering perspective, • The alternative will attain a standard of performance that is equivalent to that required under the otherwise applicable stan- dard, requirement, or limitation through use of another method or approach (40 CFR § 430(f)(1)(ii)(C)(2)-(4)). There may be other relevant reasons for DOE to seek a waiver, but they are beyond the scope of this report. The ARAR waiver standards and process could serve as a model for the exemption process that the committee proposes, to encourage a consistent and transparent ap- proach to departures from general rules when there are special circum- stances for particular waste streams at specific locations. the committee’s review of the waste streams identified in Chapter 2, four grounds for seeking an exemption from the general rule of permanent 6 geologic disposal appear to merit consideration: 6 At this, the gatekeeping stage of the exemption process, it is not necessary for DOE to prove this or any other of these criteria. Instead, DOE simply has to have a sufficient basis for believing that this might be the case to justify the time and expense of seeking an exemption. A final determination of risks and other 

EXEMPTION PROCESS 79 1. It is impractical, and in some cases may be technically impossible, to comply with the requirements at an acceptable level of worker safety, ecological impact, and cost, balancing all of these factors. 2. An alternative disposal method would provide long-term protec- tion of human health and the environment, and would not pose unacceptable risks (for example, by meeting existing environ- mental standards for near-surface disposal). 3. The cost of permanent geologic disposal is extremely dispropor- tionate to the risks actually posed by the waste. 4. The waste stream is a promising candidate for delaying perma- nent geologic disposal because it can be managed confidently in situ in the near-term future with no or negligible increase in risk, and either new, safer management technologies are a reasonable prospect or the waste will experience a substantial decay of dan- gerous radionuclides in the short term in a safe and stable con- figuration. The committee recognizes, of course, that all of these descriptions involve judgments upon which reasonable persons could differ. It will perhaps be helpful to keep in mind two guiding principles in evaluating all requests for exemptions. First, the regulatory system as a whole starts from a preventive, protective baseline that prefers permanent geologic disposal as the technique for reducing present and (especially) future risks of highly hazardous wastes. Second, the exemptions should be spe- cial, unusual circumstances. If most or even a major portion of the HLW or TRU waste streams were legitimately granted exemptions, it would call into question the validity of the general rule, as discussed below. To speak colloquially, the exemptions are intended to apply only to situa- tions in which permanent geologic disposal simply makes no sense. Fi- nally, the strong technical basis and the open process, both discussed be- low, provide a further bulwark against these exemptions’ being misused. Mindful of the concern that the exemptions could swallow the gen- eral rule, the committee seriously considered whether to recommend lim- iting the number of exemptions available under this process to ensure criteria would be part of another entity’s evaluation of DOE’s application for an exemption. 

80 RISK AND DECISIONS ABOUT TRU AND HLW that the exemptions are truly exceptions. The committee decided against this course of action, however, because any upper limit would necessar- ily be arbitrary. It also seems unnecessary, given the strict and limited grounds and criteria for granting exemptions. RCRA, for example, does not limit the number of permitted delistings of hazardous waste, but the standard and procedures for delisting are such that it is a rarity—not be- cause Congress wanted delisting to be a rarity as such, but rather because Congress protectively wanted the definition of hazardous waste to sweep a broad range of wastes into the RCRA management system. Finally, if it turns out that most of the waste streams in fact meet the exemption crite- ria, then this should be a signal that the baseline rule is fundamentally flawed, and a responsive regulatory system will take that as a signal that a more thorough review of the classification system is required. DOE can assess the anticipated number of HLW and TRU waste exemptions and, if it appears exemptions are not few in number, indicate that a more gen- eral fix may be warranted. The committee suggests that DOE begin with the three waste types identified herein, because they are illustrative of the reasons for considering alternatives to deep geologic disposition. 3.4 CRITERIA FOR EXEMPTIONS Finding 6: Human health risk is a good basis or starting point for considering whether a waste stream should be granted an exemption, but it is not a sufficient basis for deciding these questions. At a minimum, costs, work-related risks, risks to ecosystems, technical feasibility, cultural and societal impacts, land use implications, pre- existing agreements, and other, site-specific factors are also relevant in what is called a risk-informed approach. Risk-informed approaches can provide valuable information for use in an exemption process. Risk analysis provides a fundamental basis for exemptions because risk reflects one of the basic values being protected— human health—and therefore is a sensible starting point. Human health, however, is not the only value at issue, so the approach must be risk in- formed and not exclusively risk based. Risk assessment is the analytical basis of the risk aspect of a risk-informed approach, because it is a pow- erful, structured, and highly developed analytical tool for organizing (and revealing the absence of) relevant information. Moreover, its theoretical 

EXEMPTION PROCESS 81 and practical strengths and limitations are—or should be7—well known, permitting a clearer understanding of what risk analysis does and does not tell us. The process for using risk analysis is described in Chapter 4. Consideration of risk must be applied to alternatives: both the risks of the waste materials themselves in their present configurations and the risks of managing them in a different way. Perhaps even more important, given the nature of the wastes at issue, it is essential that the long-term risks of the materials in their current and alternative configurations be thoroughly explored. For example, the National Research Council (NRC, 2000b) and others (e.g., Applegate and Dycus, 1998) have previously warned that reliance on institutional controls to reduce long-term risks is highly problematic. As discussed in Chapter 5, the assessment of long- term risk can be particularly difficult because of the challenges of model- ing long-term geochemical and hydrologic processes and because of un- certainties in future land use and other human behaviors. Even with these guidelines, a risk-informed approach poses major challenges for DOE. One lesson to be drawn from DOE’s many attempts to employ risk as a central planning and decision-making tool— described in Appendix B—is that their results in fact gave little useful guidance to the program. Risk alone offered no firm standards for action, embodied no agreed-upon methodology, and had no basis in existing agreements or legal requirements. Risk, in other words, must be consid- ered along with other relevant considerations such as cost and ecological impacts in the context of a focused decision-making process. In addition, risk analysis is subject to both theoretical and practical limitations in its ability to measure precisely what should be done with these wastes (see Chapter 5). These limitations are also evident in many of DOE’s previous efforts to use risk. The results were in no case defini- tive and, in any event, suffered from data limitations, inconsistency and incommensurability, incomplete factors or considerations, and failures to consider intergenerational and long-term risks. Also, even if one could measure perfectly with risk analysis, it is very clear that there is no socie- tal consensus on the appropriate level of residual or acceptable risk (Babich, 2003). However, risk analysis can identify the exceptional cases (outliers, in other words) in which permanent geological disposal either does not decrease or even increases risk. 7 The National Research Council has explored these issues in detail in numer- ous reports (see Appendix A). 

82 RISK AND DECISIONS ABOUT TRU AND HLW 3.5 BURDEN OF PROOF AND INDEPENDENT DECISION MAKER The basic concept of an exemption process is that exemptions are deviations from the normal rule and that they are exceptional. It follows that the party seeking an exemption must justify the departure from the general rule. The foregoing give the committee’s judgments of permissi- ble grounds for seeking a departure and the criteria by which such a de- parture must be supported. Placing the burden of proof on the party seek- ing an exemption puts that applicant in the position of having to (1) come forward with sufficient credible data to support its claims and (2) per- suade a decision maker that the grounds and criteria for an exemption have been met and that an exceptional situation has been demonstrated. Finding 7: The credibility of DOE’s planning and decision making is reduced by the apparent conflict of interest created by DOE’s au- thority both to propose and to approve disposition plans for radioac- tive waste. Allocating the burden of proof to DOE is meaningful only if DOE is not also the decision maker. That is, the burden of proof would be weak indeed if it was simply a matter of DOE convincing itself that it is right. DOE’s status as a self-regulating agency is problematic because of the perceived and real conflict of interest: DOE is both petitioner and deci- sion maker. Outsiders might reasonably question whether DOE is able to separate these functions so that the agency is neutral in the latter role. Having DOE’s application for exemption subject to the judgment of an independent arbiter would make the process more credible to skeptics, of which, in this area, there are many. Therefore, the burden of proof implies, and the committee here makes it explicit, that a separate federal entity is needed as the regulatory decision maker for exemption purposes. DOE is, of course, regulated by a number of different federal and state entities. In the licensing process for Yucca Mountain, Congress developed a three-party arrangement in which DOE is the applicant, U.S. EPA sets the standards, and U.S. NRC issues the actual license. This or some other arrangement could be adopted, in which U.S. EPA or U.S. NRC is the decision maker. Persua- sive arguments could be made for each agency as regulator, because both have significant expertise in the regulation of radioactive materials. The 

EXEMPTION PROCESS 83 committee does not have a basis for making a recommendation for either agency but offers some observations on the merits of each for this role. The U.S. EPA would appear to be the most obvious regulator for TRU waste, because it is already the decision maker identified by law and has worked extensively with such waste at the Waste Isolation Pilot Plant. U.S. EPA also has been the principal regulator for cleanup at the sites at which HLW and TRU waste is found and U.S. EPA has extensive experience with stakeholder interaction under several statutes; probably more experience than U.S. NRC has. The U.S. NRC, on the other hand, is the agency mentioned in the current definition of HLW. U.S. NRC will rule on DOE’s license application for a HLW repository and is the regu- lator for the cleanup of waste, including HLW, at DOE’s West Valley site, which is perhaps the experience that is technically most similar to the management and cleanup of HLW at Hanford, SRS, and INEEL. Also, U.S. NRC is legally an independent agency and has some distance from the administration in power. At the same time, however, U.S. NRC is perceived by some to be a captured regulator, serving the interests of the nuclear industry. Further, coming as it does from the same parent agency (the Atomic Energy Commission), U.S. NRC is perceived by some as being too close to DOE and therefore having an institutional bias for DOE. Finally, the committee notes that it is desirable, but not essential, for the sake of efficiency and consistent application, that the same agency be the exemption decision maker for both HLW and TRU waste. 3.6 PROCESS The process for seeking an exemption need not be complex, but it has four requirements. First, the process must be transparent, in the sense that the proceedings are open to the public from the outset and that the application and the data on which it relies are publicly accessible. In ad- dition, the decision to grant or deny the application must be based on identifiable, publicly available data and its reasoning made clear. Second, there must be adequate opportunities for stakeholders and other inter- ested persons to participate in the process from the outset, not only by receiving information (as above), but by submitting information and views, at a time and in a manner that makes it possible to influence the decision. Third, there must be a neutral decision maker, as discussed above. Fourth, approval of an exemption application by cognizant regu- latory authorities must be required before any action inconsistent with 

84 RISK AND DECISIONS ABOUT TRU AND HLW the general rule is taken, in particular irretrievable actions. For example, it would be unacceptable for DOE to take an action that would preclude permanent geologic disposal of a particular waste stream, in advance (or even in anticipation) of receiving an exemption. A five-step application process would meet these requirements: 1. DOE would publicly announce its intention to apply for an ex- emption for a specific waste stream. This would give regulators and stakeholders the opportunity to collect relevant information. 2. DOE would then implement the risk-informed process described in Chapter 4 to generate the information to support an applica- tion. The end result of the DOE process, would be an application for an exemption. 3. As part of the internal DOE process, the risk analysis would be thoroughly, independently, and externally reviewed before sub- mitted with the application. 4. DOE would submit its application to the relevant decision maker. 5. If approved, DOE would be authorized to approach other regula- tors, in particular the states, to seek any necessary changes in compliance agreements. The final step bears emphasis. As stated at the beginning of this chapter, the committee expressly rejects the unilateral abrogation or amendment of existing compliance agreements by DOE (nor does the committee believe that DOE seeks to do so). The exemption process is the beginning, not the end, of renegotiation of compliance agreements, although the process the committee envisions would involve the states from the outset, which should make renegotiation simpler if the exemp- tion is granted. Conversely, if DOE is unable to convince the exemption decision maker that an exemption is justified, it has no basis for ap- proaching the states to renegotiate. Recommendation 1: The nation should pursue a formal, well- structured, risk-informed approach to consider what parts of the waste types enumerated in Finding 3, if any, should be disposed in some manner other than deep geologic disposal. 

EXEMPTION PROCESS 85 Recommendation 2: DOE should not attempt to adopt these changes unilaterally. Likewise, the exemption process that the committee recommends must be implemented in the context of DOE’s existing or renegotiated compliance agreements.