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.
5 A Framework for Evaluating Waste Characterization Activities v ~ . . i In this chapter, the committee describes the elements of an analytical framework to evaluate the effectiveness of specific characterization activities by relating the information gathered through waste characterization to the benefits it produces. The chapter ends with simplified illustrative applications of the proposed framework. 5~1 Statement of the Problem The U.S. Department of Energy's (DOE's) contact-handIed transuranic (CH-TRU) waste characterization program is controversial. DOE as well as the New Mexico Environmental Evaluation Group (EEG) and previous National Research Council committees have suggested that some characterization activities are too extensive, costs are too high, and some of the information gathered is not used to make any decision related to the protection of human health and safely or the environment. On the other hand, stakeholders in New Mexico have argued that all the current characterization activities are necessary to ensure an adequate level of protection. DOE has informed the committee of its intention to propose changes to the characterization program to eliminate or modify activities that do not have an impact on human health and safely or the environment. However, as discussed in Section 5.3, DOE did not provide the committee with a formal analysis of the impacts of altering specific characterization activities on costs or on risks to the public, workers, or the environment. DOE is responsible for making the policy decision to seek regulatory approval for changes to a particular characterization activity. In the committee's view, a structured and quantitative analysis is necessary to determine whether a change to the characterization program is warranted and to justify proposed changes with the regulator~s), state and local officials, and the general public (see Recommendations 1, 5, and 6 in Chapter 6~. In this chapter, the committee provides the elements and illustrative applications (albeit incomplete) of a proposed structured and quantitative analytical framework that could be used to evaluate changes to specific characterization activities. 5.2 An Overview of a Proposed Analytical Framework Below, the committee presents the elements of a generic analytical framework that could be used to iclentify changes to specific characterization activities. The purpose of the proposed analytical framework is to determine the value of the characterization information obtained through a given characterization activity. If the information is used to make decisions about waste handling, transportation, or disposal, then it has an impact on the outcome of these decisions (for instance, on reducing risks, uncertainties, costs, or delays), and thus it has value in the present context. On the other hand, if the characterization information is not used in current or future decisions, then it has no impact, and therefore it has no value. These are the two extremes. In most cases there 58
A Framework for Evaluating Waste Characterization Activities 59 is some value,4 but the question is whether costs to obtain the information exceed the benefits it produces. Figure 5.1 shows a general flow diagram of the characterization and decision- making process. Boxes represent decisions by DOE, such as the decision to review a characterization activity in the current characterization program or to request a permit modification for the Waste Isolation Pilot Plant (WIPP). Ovals represent outcomes of these decisions. The current waste characterization program is composed of characterization activities (see Section 4. ~ ~ that generate a flow of information about the wastes. Some of this information is used to make decisions, such as whether to remove prohibited items or to vent a canister prior to transportation. The characterization activities, along with the decisions made on the basis of the information gathered, lead to impacts in terms of risks to workers and the public, costs, and public concerns. Feedback loops represent updating and improving characterization activities through experience, data collection, and analyses. Experience and learning lead to ideas for improvement, by eliminating unnecessary characterization activities, finding more cost-effective2 solutions, or by adding activities to further reduce possible risks. Once these possible improvements are identified, they form the basis for a permit modification request, which is submitted to the appropriate regulatory agency. The regulatory agency decides whether the proposed change is still in compliance with the regulations and approves or declines request. Within this framework, DOE would communicate interactively with stakeholders about proposed changes to the program and stakeholders would provide input. Stakeholder input could be provided informally during public information exchange meetings (such as the WIPP's quarterly meetings), or formally, through a public comment period during the regulatory process. Stakeholders concerns are part of DOE's analysis to determine whether a permit modification is warranted (see Section 5.6~.3 Within this framework there are several opportunities for structured and quantitative analyses of the value of the characterization information. First, in applying this framework, the characterization information could be recorded and analyzed to determine how the characterization information is used and whether it is used frequently, rarely, or never. For example, information on volatile organic compounds (VOC) or instances of identified prohibited items in waste drums could be recorded and its impact on decisions to vent canisters or to remove prohibited items could be analyzed. Second, the impacts of the characterization information on the decisions made on the basis of this information could be analyzed in terms of their combined impact on risks, costs, and other aspects of the characterization program. For example, the risks of transporting canisters with high VOC levels or with prohibited items could be analyzed to determine the risk reductions achieved by characterization. By comparing risks, costs, and other impacts, such as policy and societal impacts, with and without characterization information, the proposed analytical framework can establish the value of characterization information. Similarly, this analysis can support proposed changes to the characterization program by showing that alternative characterization activities are more cost-effective than the current ones. The analytical framework proposed here denves from a decision-analysis For example, characterization information is used in preparation and archiving of a record of characterization activities and may have value in that context. 2For a definition of cost effectiveness, see Section 5.7. Mechanisms for stakeholder participation in decision-making have been discussed by other National Research Council committees (NRC, 2003 and references therein).
60 _ ~ Decisions based on Characterization Information ~ Risks, Costs, and an,\ Other Impacts ~ ' ~ Stakeholder Input , 1 I Permit Modification Request . . . \/ Regulator's ~ ? ~: ~ Decision J L Improving the Characterization Program for Contact-Handled Transuranic Waste Current Characterization Activity/Program . Information Generated by - \ / Experience andAnalysis- Characterization A ~ Based Learning Revised Characterization Activity/Program FIGURE 5.1 Features of the proposed analytical framework. Boxes represent decisions; ovals, outcomes; and arrows, the dependencies between them. The dotted line indicates uncertainly in the regulatory decision. The double-headed arrow represents interactive communication. For an explanation of the figure, see text. too! called the Value of Information.4 Other analytical approaches have elements in common with the proposed framework, such as the Data Quality Objectives approach proposed by the U.S. Environmental Protection Agency (EPA) (2000~. An analysis of the effectiveness of characterization activities addresses not only risks and costs, but also programmatic impacts (e.g., delays in processing wastes) and policy and societal impacts (e.g., need for an authorizing document modification request or a shift in public perception clue to a change to the program). These impacts and suggestions for how to address them within this analytical framework are discussed below. 5.3 Assessment of the Risks The first type of analysis in the proposed analytical framework is risk assessment. In the present context, the purpose of risk assessment is to determine the link between waste characteristics and risks related to handling, transportation, and disposal of CH-TRU waste at WIPP. The assessment of risks takes into account risks at a "micro" level (e.g., risks to workers gathering characterization information) anti at the "macro" level of the National TRU Waste Management Program (e.g., risks incurred at 4For example, see Clemen and Reilly (2001~; Winkler (2003~; and, for an application to the characterization of Hanford tank wastes, see Fassbender et al. (1996~.
A Framework for Evaluating Waste Characterization Activities 61 the site, during transportation, and at the WIPP facility). Non-technical risks, such as the risk of decreased public confidence, are also important in the proposed analysis. Waste characteristics that affect health and safely are the most important to characterize. If a characterization activity produces information on waste characteristics that are not linked to human health and safely or the environment, then there are no programmatic decisions to be made with respect to those characteristics. The risk assessment also determines the accuracy and precision needed in knowing a given piece of characterization information. Uncertainties in each analytical measurement are considered and propagated through the analysis so that risks are weighted by the propagated uncertainties. Each estimate of risk therefore has an uncertainly assigned to it. Risks with large uncertainties indicate that those characterization activities that define that risk need more consideration and attention unless the risks themselves are of lithe consequence. Such an analysis may indicate that more characterization is needed or that different characterization activities should be explored. Estimates of risks for which the uncertainties are small indicate that a higher degree of confidence exists for those parameters, and perhaps less analytical data are needed or could be justified. When assessing risks and how they are impacted by characterization information, it is important not only to consider routine and expected situations, but also to inquire how the characterization information might be used in unusual and unanticioated or low orobabililv circumstances. Several situations in which , . ~ . _ · _ _~ ~ ~ _e B_ _ _ ~ _ B_ ~ _ _ _ _ ~ ~ . _ _ ~1 ~ ~ ~ _ : . . . _ _ 1 _ characterization data COUIO ne valuable can oe envisioned. Nor example, IT WaSIe emplaced in WIPP must be retrieved, characterization information might become useful. In addition, different or additional characterization information may become useful for future, as yet undefined, waste streams. Risks in accident conditions would take into account the probability and magnitude of the consequences for each accident scenario in which characterization information could be useful. DOE has prepared several analyses containing elements of risk assessments that were submitted to WIPP's regulatory agencies to authorize operations at the WIPP facility. For instance, the EPA Certificate of Compliance relies on an evaluation of a performance assessment which includes elements of risk assessment (DOE-CGA, 1996~. The Lancl Withcirawal Act requires an analysis of the environmental impact of WIPP operations (DOE-WAC, 2003) and a Safely Analysis Reports (DOE-SAR, 2003~. DOE has also concluctecl safely analyses and performance assessments in connection with transportation of wastes to WIPP (DOE-SARP, 2003~. Many of these studies are now being updated to reflect recent data and experiences and are periodically reevaluated. Although DOE has performed analyses of many aspects of operations related to WiPP performance, including transportation, the committee could find no studies that explicitly, systematically, and quantitatively link its waste characterization program to risks to the public, workers, or the environment (see Finding and Recommendation ~ ). In particular, DOE has not used the studies mentioned above to estimate quantitative relationships between current or planned characterization activities and risks. DOE 5The WIPP Safety Analysis Report, required by agreement with the State of New Mexico, documents the adequacy of safety analysis to ensure that a facility can be constructed, operated, maintained, shut down, and decommissioned safely and in compliance with applicable laws and regulations. The report is based on the facility's safety analysis, which is a documented process that: 1) provides for the systematic identification of hazards within a given DOE operation; 2) describes and analyzes the adequacy of measures taken to eliminate, control, or mitigate identified hazards; and 3) analyzes and evaluates potential accidents and their associated risks.
62 Improving the Characterization Program for Contact-Handled Transuranic Waste provided cost information for ongoing characterization activities but the committee could not find any discussion of the costs and benefits of current characterization activities compared to alternative ones. The committee is not aware of any analysis showing the risks of handling, transporting, or of disposing of some of the prohibited items listed in the Hazardous Waste Facility Permit (HWFP). Such items are currently handled by workers when they are identified in waste characterization and removed from waste streams to attain compliance. To the best of the committee's knowledge, DOE's risk assessments address only waste characterized in compliance with Waste Acceptance Criteria and HWFP requirements. In these risk assessments it is assumed that waste characteristics , ~ , are known based on the current characterization program. For instance, the performance assessment submitted for EPA Certification is based on the assumption that waste activity and volume are within the limits established in the Land Withdrawal Act and that the waste does not contain any water or other prohibited items. Only recently (September 2003) has DOE submitted to the committee studies that could be viewed as the initial elements of an analytical approach in connection with a request for a chance in the headscace Gas samolino-related activities (Boalwriahi. . ~ , ~ . ~ ~ ~ , 2003; Myers, 2003; and McCutia and Van Soest, 2003~. These studies include en evaluation of WIPP room-based VOC monitoring as an alternative to analyzing the headspace gas of each drum. Additional material submitted address Data Quality Objective-related issues, and the accuracy of Acceptable Knowledge (AK) on waste characterized to date (HWFP, 2003; Kehrman and Most, 2003~.6 Examples of issues to be included in the proposed analytical framework to determine the linkage between waste characteristics and risks in handling, transportation, and disposal of CH-TRU waste at WIPP can be found in Appendix F. Risks considered are the following: risks related to waste handling; risks related to transportation; and risks related to waste disposal. Within its constraints of time and information, the committee could only point out some of the risks to be included in this analysis; a complete risk assessment is a major endeavor that the committee was not prepared to undertake. 5.4 Costs of Waste Characterization Analysis of the impacts of waste characterization on costs would include "hard" as well as "soft" costs. Hard costs are the costs associated with gathering characterization information. Table 4.1 shows that the most expensive characterization activities involve breaching waste containers and waste analyses. This is due to the ~ _ ~ . _ . _ costs of protecting workers from radiation. Other hard costs Include: costs of equipment, data management, overhead costs (including that for quality assurance), characterization method certification, staff training, record keeping, as well as the costs of submitting a permit modification. Soft costs are costs that are difficult to quantify, such as the costs associatecl with a possible loss of stakeholder confidence; the costs of delaying shipments to WIPP if the permit modification request is denied; or costs incurred in the case of an accident 6NMED found that these analyses lack correlation between the characterization issues associated with waste emplaced to date and the remaining inventory projected for disposal at WIPP (NMED, 2003c).
A Framework for Evaluating Waste Characterization Activities 63 (e.g., costs of cleanup if reduced characterization information leads to an accident during waste handling, transportation, or disposal). According to DOE's cost analysis, it is not possible to predict whether characterization costs will increase or decrease with time: A comparison of the unit costs developed [beivveen October 2002 and June 2003] showed no significant trend (either up or down) for unit costs over time. While some costs did in fact decrease, others increased. Many factors impact the costs, some of which cannot be controlled, or reduced, at the activity level. In some cases, a reduction in quantities may even lead to an increase in unit cost. Site-specific priorities such as stakeholder interaction, or regulatory impacts, directly impact unit costs. Increases or decreases in facilities and infrastructure costs affect the cost structure relating to characterization activities. Sites differ in the types and amounts of TRU waste, schedules for shipping, contractual agreements for characterization, and the maturity of the program. All of these factors, whether positive or negative, affect the total costs. While processes may be streamlined and shipping procedures made more efficient, costs may not be reduced proportionately. Therefore, reducing future costs must be pursued in other areas, such as regulatory relief and technology development (DOE- CABE, 2003; page 23~. ~ s _ · · _ - - The committee believes that, as the experience base continues to grow, it is possible to analyze costs sufficiently to produce at least a credible and useful prediction of cost trends under the most likely scenarios. Changes to the characterization program that would improve the characterization process, together with an expanding operational experience base, could help reduce both risks and costs. However, if new and more difficult or complex waste types have to be dealt with in the future (see Sections 2.3 and 2.4y, or more stringent cleanup standards are imposed by the regulators, then characterization costs could increase. Scenarios in which characterization needs and costs may increase are the following: radiographic techniques and other special characterization techniques and technologies for large containers may have to be developed and may cost more than the observed costs to date; remote-handIed transuranic waste with dose rates greater than 1,000 rem per hour may be authorized in WIPP possibly requiring additional worker protection measures; previously buried TRU wastes may include more compromised containers requiring more elaborate handling procedures; future waste streams may require new or different characterization information; additional TRU waste not accounted for in current inventories may be approved for disposal in WIPP increasing total characterization, transportation, and disposal costs; or characterization requirements may be added or strengthened for public confidence purposes, in particular if an accident occurs.
64 Improving the Characterization Program for Contact-Handled Transuranic Waste Changes to the characterization program also have an impact on the schedule. The proposed analytical framework would consider time and effort required to: acquire alternative characterization information (to the required accuracy and precision); manage the data (including quality assurance activities at the data generation level and project management level); prepare a permit modification request; obtain approval of the modification, obtain certification; and train personnel to use new characterization equipment. 5.5 Policy Impacts Most7 changes to the characterization program must be approved by the regulator~s) to ensure continued regulatory compliance. Regulatory compliance is grounded in risk considerations. Regulatory requirements are derived from assessments of the risks to the public, workers, or the environment and the need for records documenting waste characteristics, waste characterization activities, and oversight of related operations. Therefore, while characterization is not a risk assessment in itself, its purpose is to provide information used in determining if criteria derived from risk assessments have been met. DOE and its regulators worked to translate the intent of various environmental laws (such as the Land Withdrawal Act, the Resource Conservation and Recovery Act) into specific regulatory requirements for waste characterization. These regulatory requirements are set forth in 40 CFR 194, the HWFP, and the Certificate of Compliance for transportation packages (see Chapter 3 and Appendices C and E). The details of characterization activities are prescribed in W! PP's Waste Accectance Criteria and in the HWFP. The Waste Acceptance Criteria can be modified ... . . . . . . . . ,.. . ... _ _ ~ .. . . . Without regulatory approval It changes are not In contact with Thea, New Mexico Environment Department (NMED), or U.S. Nuclear Regulatory Commission (USNRC) regulations; any change to the HWFP has to be approved by NMED. Experience shows that requests for authorization document modification are most likely to be approved if they are supported by careful analysis and detailecl recorcis of characterization experience. As a result of substantial interactions with the regulators and thorough technical justification in specific areas, DOE has succeeded in obtaining many revisions of the WIPP program thus far (see Chapter 3~. The political feasibility of a change in the characterization program is a further element to be considered in the impact analysis. The HWFP is the only direct control that New Mexico has on the characterization process and WIPP operations. Any proposed change to the characterization program wit! be examined carefully by the state with regard to its effects on previous agreements and on the state's role in regulating WIPP. Moreover, continuing to use the established regulatory processes for making changes to the characterization program couicl help build cooperation and confidence among all parties (see Finding and Recommendation 6~. The proposed analytical framework could be used to make a technically defensible case before the regulator and the public that a given change to the program is still protective of human health and the environment. The regulator would make the final determination whether the proposed change would still lead to compliance with regulations. 7Some changes to characterization activities are under DOE's purview, such as the addition of fast-scan radiography at the Savannah River Site to screen for prohibited items (see Section 4.3~.
A Framework for Evaluating Waste Characterization Activities 65 5.6 Societal Impacts Although societal impacts are difficult to quantify, the impact of a change to the characterization program on stakeholder confidence in DOE, not only in New Mexico, but also in generator and corridor states, cannot be ignored (see Findings and Recommendations 5 and 6~. The public, state and tribal officials, and organizations such as the Western Governors Association and citizen's groups watch the permit modification process closely. Evidence suggests that some members of the public and state officials lack trust in DOE. Some stakehofJer concerns are listed in Section 4.6.4. Unless supported by technically defensible analyses showing that no increase in health, safety or environmental risk will ensue, a request for change in a permit may be interpreted as leading to a decrease of safely, with concerns of increased risk to workers, the public, and the environment. Changes to the characterization program that would accelerate waste shipment to WIPP (or elsewhere) may be popular at generator sites from which the wastes would be removed. However, they may have a negative societal impact in the recipient state, New Mexico, or in corridor states through which waste is transported, if it is believed that such acceleration reduces safely. Providing a defensible and credible technical analysis and characterization record to the public in support of programmatic changes could build confidence that the changes do not weaken protection of human health, safely or the environment, and help avoid charges of secrecy that might otherwise be made. Characterization information could be disseminated in a variety of ways, including through published reports, public access to the WIPP Waste Information System, or summaries thereof. The analytical framework proposed by the committee, or any comparable analysis produced in an open and responsive process, can help reduce or mitigate any lack of trust and lead to greater tolerance of changes, where warranted. The societal impact of changes to the program, provided they are technically defensible, may even be positive overall in light of savings in taxpayer dollars. 5.7 Cost-Effectiveness Analysis To determine whether a characterization activity has value, information on the risks, costs, and other impacts with and without the characterization activity is needed. For example, to determine the value of headspace gas sampling and analysis and how it reduces risks, data on the flammability and risks of worker exposure to high VOC concentrations if headspace gas sampling did not occur are necessary. There is a simple rule to identify which characterization activity needs to be re- evatuated through a structured and quantitative analysis: if the information gathered by this activity is never used for making decisions about waste handling, transportation, or disposal, it has no value. This is because the risks, costs, and other impacts of this characterization activity would be the same without collecting this information. Ultimately, the choice of modifying or eliminating a characterization activity from the program depends on a cost-effectiveness analysis. Proposed changes to current characterization activities could be justified by clearly demonstrating that the changes improve the program in the following ways: 1. the proposed change is less expensive, involves less worker risks, less public risk, and less other impacts than the current characterization activity; 2. the proposed change is less expensive or has less worker risks or both and has the same public risks and other impacts than the current characterization activity, particularly if the current characterization activity generates information that is not used for decision-making purposes; or
66 Improving the Characterization Program for Contact-Handled Transuranic Waste 3. the proposed change costs much less or has much less worker risk or both and affects public risk and other impacts in a minor way. The first case is the most convincing one: there is no reason to reject a change that is better in all aspects. The second case involves a cost-benefit analysis that determines whether it is possible to obtain the necessary information for less cost and worker exposure. The third case is the most complex one, since it requires quantification of the terms "much" and "minor," which is usually done in a cost-effectiveness analysis. A cost-effectiveness analysis differs from a cost-benefit analysis in that the former relates the money spent to the benefit (in whatever units) gained, while the latter requires that all costs and benefits be expressed in monetary terms. Cost-effectiveness does not require monetization of "effectiveness," just quantification of a measure of effectiveness. Cost-effectiveness quantifies all costs as direct costs and all benefits or effectiveness measures in non-monetary terms. This allows making judgments about whether it is worth spending an incremental amount of money for an incremental amount of "effectiveness" without assigning dollar values to the effectiveness measure. In this context, a cost-effectiveness analysis allows assessment of a broader range of impacts (including soft costs, policy, and societal impacts) without attaching monetary values to them. ; . ~ 5.8 Implementation Considerations The proposed analytical framework is conceptually powerful, but demanding in practice. A technically complete, formal cost-effectiveness analysis, as defined in risk and decision analysis, requires a "well-defined problem." That is, the decisions and outcomes are defined, an appropriate "measure of value" for the problem is specified, the relevant measurable parameters are identified, and the degree of certainly or knowledge about each of these is quantified. The complete and formal application of such an analysis requires that: first, the connections between waste characterization and impacts to workers, the public, and the environment be identified; second, that actual information about risks, costs, worker exposure, and benefits of characterization information be available; and third, that all consequences (environmental, health, programmatic, policy, and societal risk, costs, and benefits) be converted into a single number specifying equivalent cost or utility. Given its complexity, a complete analysis of the benefits and costs of each waste characterization activity is beyond the commidee's capability in this study. In any event, neither DOE nor the regulators provided any information on the linkages between characteristics, waste characterization activities, and protection of human health and environment. The committee also had limited information on worker doses associated with some specific activities. Absent the linkages between waste characterization and system performance, the committee could not identify specific changes to characterization activities. To identify the most promising areas in which to evaluate the effectiveness of the characterization program, a less elaborate analytical framework than the one described above can be used. If an alternative characterization activity is compared to the current one, and the alternative is found to be better from the point of view of risk and cost and at least equal on all other parameters, then these activities are candidates for a quantitative cost-effectiveness analysis.
A Framework for Evaluating Waste Characterization Activities 67 5~9 Applications Based on the information provided by DOE and on findings by a previous National Research Council committee, three characterization activities appear to be candidates for re-evaluation using the proposed analytical framework: 1. headspace gas sampling and analysis; 2. homogeneous waste sampling and analysis; and 3. visual examination to confirm radiography results. , According to the information gathered by the committee, these three characterization activities are carried out with their current sampling frequencies mainly for regulatory compliance and do not appear to directly reduce risks or costs. These activities could be analyzed through the proposed analytical framework to determine their value and impacts, and whether there are alternative characterization activities (or modifications to the current one, such as different sampling frequencies) that would still be in compliance with regulations and whose risks and costs to gather the information would be more commensurate with the risks entailed. If the alternative has positive impacts on risks and costs, as well as on policy and societal factors, then DOE may decide to apply for a permit modification. For each example below, the committee attempts to analyze the value of the characterization information qualitatively. The information and conclusions provided are based on information gathered during the study. Different, more appropriate examples might become evident if a more complete risk and impact assessment were implemented. 5.9.1 Headspace gas sampling and analysis Heacispace gas sampling and analysis was chosen as an example on the basis of information submitted by DOE, on cost considerations, and on the findings of a previous National Research Council committee. Although not the most expensive activity per unit cost, headspace gas sampling and analysis is one of the most expensive activities when averaged on the entire waste inventory because it involves sample collection and analysis on 100 percent of waste containers. During information gathering meetings, DOE has publicly stated that this activity is performed only to ensure compliance with the HWFP. A previous National Research Council committee found that there was no specific regulatory requirement in the Land Withdrawal Act, the Resource Conservation and Recovery Act, or in 40 CFR 264 to sample headspace gases in the totality of waste. DOE proposed to NMED to do so while drafting the HWFP application (NRC, 2001~. NMED accepted the proposal and this characterization activity is now codified in the HWFP and therefore has become a regulatory requirement. The HWFP now mandates that the concentration of 29 VOCs be measured in all drums (see Section 4.3 for exceptions) before shipment to WIPP to ensure that their concentration in WIPP waste disposal panels is within the limits allowed in the HWFP. The alternative characterization activity considered by the committee is room monitoring in WIPP for VOCs coupled with flammability tests at generator sites for compliance with transportation regulations. The suitability of this alternative activity could be perioclically checked by headspace gas sampling and analysis on a statistical basis. VOC concentrations in W!PP are currently monitored in W!PP's exhaust shaft, as required in the HWFP, to confirm the absence of VOC release into the repository.
68 ~ . .. . . Improving the Characterization Program for Contact-Handled Transuranic Waste 5. 9. 1. ~ Assessment of risks This characterization activity is designed to help assess risks from the presence of VOC in W! PP above the allowable limits, presence of flammable gas above flammability limits during shipping and handling drums in W! PP, and presence of hazardous compounds in WIPP not identified with AK. The objective of headspace gas sampling and analysis is to prevent shipment of flammable or explosive gases, exposure of WIPP workers to VOC concentrations above applicable limits, and VOCs release in the atmosphere. Room sampling would measure directly VOC concentrations in WIPP.8 Flammability limits could be tested at generator sites using alternative characterization methods to track hydrogen and methane (for example using gas chromatography with a flame ionization detector). The proposed alternative characterization activity would produce equivalent risk information in a potentially more reliable fashion. According to DOE, VOC concentrations measured in the stagnant air behind the curtain used to isolate each disposal room after it has been closed correlate well with results from headspace gas sampling and analysis at generator sites (Boalwright, 2003~. Therefore, this method could be used instead of headspace gas sampling and analysis at generator sites. However, the information would be at a different level of specificity, as current headspace gas sampling and analysis provides data on each drum, while room sampling would not identify individual container sources for any VOCs detected. 5.9.1.2 Value of characterization information The concentrations of the 29 VOCs in each drum are used to project VOC concentrations in WIPP and to confirm AK by comparing the VOCs found in the headspace gas to those expected from historical knowledge of the waste. In its review of CH-TRU characterization activities, the EEG writes (EEG, 2003b; page 28~: [Headspace gas sampling and analyses] do provide additional information on the contents of waste containers. Additional waste streams have been defined because of the results of these analyses, and on occasion additional RCRA [Resource Conservation and Recovery Act] hazardous waste numbers have been added to waste streams. The importance of these functions has occasionally been denigrated, primarily because this additional information is not used to control quantities of VOCs coming to WIPP other than to show compliance with the room based concentration limits. [...] This information is used (in conjunction with acceptable knowledge) to assign hazardous waste numbers to each container. However, EEG is not aware that these hazardous waste numbers are used to exclude waste from the WIPP or to otherwise control the hazardous waste. These data probably provide the incidental benefit of confirming AK and ensuring the various Waste Acceptance Criteria (WAC) requirements for stability of waste are met. The EEG believes that it is desirable to maintain a comprehensive headspace gas sampling and analysis program for CH-TRU wastes as an additional confirmation method for AK. However, it adds (EEG, 2003b; page 58~: To date, VOCs have not been observed in the exhaust shaft at WIPP, due both to the high rate of air flow through the mine and because waste received to date has contained low levels of VOCs in the headspace gas.
A Framework for Evaluating Waste Characterization Activities it should be possible to require less than 100°/O sampling in some cases. Our primary concerns are with organic sludges and older waste containers where information may be less reliable. 69 Headspace gas data and records are also used to ensure that the concentration of flammable gases (in particular, methane and hydrogen) does not exceed flammability limits for transportation purposes. Headspace gas information affects transportation decisions (e.g., purging the container prior to transportation) and consequences (e.g., the risks and possible consequences of deflagration during transportation). The information gathered with the alternative characterization method could be used to make the same decisions about transportation as above. 5.9. ].3 Cost impacts Details on the efforts and costs to obtain the headspace gas sampling and analysis are in Chapter 4. The sampling and analysis requires approximately 4.4 person- hours and costs, on average, $620 per waste container. Most of the characterization effort and time is due to the data management for 29 VOCs. DOE informed the committee that, although reliable cost estimates are not available, the technical effort to obtain hydrogen and methane concentrations through gas chromatography using a flame ionization detector is similar, in terms of equipment and time, to that for VOCs (Nelson, 2003~. However, the data management effort is substantially lighter because there are only 2, rather than 29, VOCs to track. The same quality assurance requirements apply to the monitoring of VOCs in WIPP, as established in the HWFP. 5.9. 1.4 Policy impacts Any change to the headspace gas sampling and analysis activity requires a HWFP modification. In several occasions, NMED has stressed the importance of this characterization activity to satisfy the regulatory requirement of 40 CFR §264.~3(a)~1 ~ to "obtain a detailed chemical and physical analysis of a representative sample of the wastes" before such waste is accepted for disposal. NMED has provided comments to the committee on the DOE's analyses supporting the request for relief from this characterization activity. According to NMED, DOE did not go far enough in the analysis of the relationship between VOC data of waste emplaced thus far and those of future waste streams (NMED, 2003c; cover leHer): tthe conclusions reached in the AK accuracy analysis] do lithe to address the uncertainties associated with older, poorly documented waste streams generated fifteen to thirty years ago that have yet to be characterized. The committee believes, based the information it has received, that the proposed alternative can be shown to be in compliance with regulations and to ensure the same level of protection of human health and the environment as the present characterization activity. Should VOC concentrations measured in disposal rooms begin to rise, steps (e.g., increasing ventilation rates) can be taken to control worker exposure or releases to the atmosphere. Concerning the policy impacts on transportation requirements, flammability characteristics of waste can be monitored by any method or equipment that complies with the quality assurance requirements (calibration, records, training, and so
70 Improving the Characterization Program for Contact-Handled Transuranic Waste forth) listed in the Quality Assurance Program Plank However, a change in the flammability test may require a modification to the USNRC Certificate of Compliance for shipping containers. , 5.9. 1.5 Societal impacts According to the records of public comments to HWFP modifications, stakeholders in New Mexico have reacted strongly to proposed changes to the headspace gas sampling and analysis requirement. Their concern is that any change to this requirement could jeopardize the protection of human health and safely and the environment (Nuclear Watch of New Mexico, 2002~. A change to this requirement, if not fully supported by a structurecl, creclible, and quantitative analysis, could cause a loss of confidence, in particular in the case of an accident involving VOCs or hazardous waste components that were not detected prior to shipping the waste to WIPP. Corridor states also need assurance that the proposed alternative does not increase transportation risks. 5.9. 1.6 Cost-effectiveness analysis The following cost-effectiveness analysis for this characterization activity is only qualitative and is based on partial information received by the committee during this study. The purpose is simply to illustrate the use of the value of information framework in a structured and quantitative analysis of characterization activities. Based on the comparison between the status quo (headspace gas sampling and analysis) and the proposed alternative, it appears that the information on VOC concentrations obtained at generator sites is redundant with that obtained through WIPP disposal room monitoring, could potentially be less reliable, could be obtained with a cheaper alternative, and therefore has low value. In other words, the value of the information gathered with the proposed alternative appears to be higher. In situ monitoring coupled with measurements of hydrogen and methane at generator sites for transportation regulation purposes is a more cost-effective solution that could ensure the same, or a potentially higher level of protection of human health and the environment. Moreover, the committee believes that EEG and NMED concerns with organic sludges and older waste containers can be addressed with the proposed alternative. To decide whether a permit modification would improve the program a more complete analysis is needed in which the advantages of using the alternative characterization activity are considered along with the cost, time, and effort to prepare a permit modification request to NMED (and possibly a USNRC Certificate of Compliance modification), as well as with the risks of decreasing the confidence of the public in New Mexico and that of corridor states in the safely of WIPP operations. 5.9~2 Homogeneous waste sampling and analysis Homogeneous waste sampling and analysis is the most expensive unit characterization activity even though it applies to only a small fraction of the waste (see Table 4.~ ). A previous National Research Council wrote the following about this requirement (NRC, 2001; page 80~: 9The requirements for gas testing are listed in the "Quality Assurance Program Plan for the Gas Generation Test Program" (available at <www.wipp.ws/library/gasgenqapp/DOE-WIPP- 01-31 87Rev1 .pdf>).
A Framework for Evaluating Waste Characterization Activities No operational decisions are made based on these data; that is, the results of the sampling and analysis do not affect how waste is handled' so it is not clear what justifies the additional radiation exposure risk and cost of this procedure. In the committee's view, this sampling and analysis applied only to homogeneous waste is unnecessary: If acceptable knowledge documentation...provides sufficient characteriza- tion information for heterogeneous waste, the committee can identify no technical reason why acceptable knowledge should not also be adequate for homogeneous waste. , .. 71 The proposed alternative is a reduction or elimination of homogeneous waste sampling and analysis to confirm AK. 5.9.2. ~ Assessment of risks, costs, anc] other impacts According to the HWFP, the risks addressed by this characterization activity are related to the potential toxic characteristics of homogeneous waste due to the presence of metals or other hazardous compounds. Risk information is needed to determine the health and safely consequences to humans or the environment related to the sampling and analyses of homogeneous waste. Details on the efforts and costs of performing homogeneous solids sampling and analysis are in Chapter 4. Characterization data produced involve measuring the concentration of 30 VOCs. 11 semi-VOCs. and 14 metals. Other hazardous components may be added to the characterization data requirements for a waste stream if they are found in more than 25 percent of the sample for a Given waste stream. To the best of the committee's knowledge' the information ~;~ ~Z~ , acquired through this characterization activity is not used to make any decision on handling, transportation, or disposal of CH-TRU wastes. Table 4.1 shows that this is the most expensive characterization activity (on average) because of the need to prevent contamination of workers and of the workplace during sampling (drilling into the solid waste) and analyses. r ~ ~ ~ ~ ~ I hese waste characterization activities require from 8 to 40 person-hours and cost an average of $87,000 for each unit on which they are performed. That averages to $430 per waste container for the sampling and analysis over all units characterized. A change in this characterization activity would require a HWFP modification. Previous modifications to this requirement have already been approved by NMED (see Section 3.3~. These changes concern statistical quality control methods, quality control requirements for two semi-VOC analyses, and the reduction from three to one sample from each core (EEG, 2003b). A societal impact of a change to this characterization activity has a potential to decrease of public confidence in DOE particularly if an accident occurs during handling, transportation, or disposal of uncharacterized homogeneous waste. 5.9.2.2 Cost-effectiveness analysis It appears that the information gathered with homogeneous waste sampling and analysis has low or no value. In that case, reducing or eliminating this activity could reduce costs without affecting the protection of human health and safely and the environment. The New Mexico Environmental Evaluation Group independently reached the same conclusion and wrote (EEG, 2003b; page 60~: The EEG continues to believe that the homogeneous sampling and analysis are unnecessary waste characterization requirements in the HWFP. Our principal reason for this position is that the data are not used
72 Improving the Characterization Program for Contact-Handled Transuranic Waste for any additional regulatory control (metals releases from accidents or long-term processes would be controlled by radionuclide control requirements and VOCs and semi-VOCs by theadspace gas sampling and analysis] or the Confirmatory VOC Monitoring Plan). To decide whether a permit modification would improve the program, the advantages of performing less or no homogeneous waste sampling and analysis are to be considered along with the time and effort to prepare a permit modification request to NEED, as well as with the risks of decreasing the confidence of the public in New Mexico and that of corridor states in the safety of WIPP operations. 5.9.3 Visual examination to confirm radiography results Visual examination as a quality control check on radiography results on existing waste (or reirievably stored waste, according to DOE's terminology) is the second most expensive activity per unit container. A previous National Research Council committee wrote (NRC, 2001; page 80~: there is no requirement fall {air '`~ri~i~ti~^ ~{ real lows r-~i~r~h`` , IVI Vail Ill~aLlUI I Vl I OOl-Lll 11= 1 O~lV8l Oral Iy results.... line visual examination confirmation is a self-imposed procedure that yields no benefit but results in increased risk of exposure and cost. . The proposed alternative is to decrease the size of the statistical sample or to eliminate this activity completely to reduce waste handling. 5.9 3. ~ Assessment of risks, costs, and other factors Visual examination provides information designed to assess risks related to the hazard of mix-certification, that is the risk of handling and disposing waste that is not suitable for transportation to and disposal at WIPP or estimating an erroneous waiting time to comply with the drum age criterion. Experience gathered to date could provide statistics on how often and which items were mix-identified using radiography, how often visual examination has produced new information, or what is the test-retest or inter- operator reliability of radiography operators. Risk information is needed on the impact on human health and safely of mix-certification, miscalculating a drum age criterion or material parameters weights. Information is also needed on worker exposure during this activity to compare benefits and risks of performing this characterization activity. Details on the efforts and costs to perform statistical visual examination to confirm real-time radiography results for existing (or retrievably stored, in DOE's terminology) waste are in Chapter 4. The information gathered through this activity is used to determine the mix-certification rate and to verify the aging calculated by radiography (the latter is done by visually examining the packaging configuration, type and number of fillers, and inner layers of confinement). Operational decisions (i.e., whether to ship the waste or to open containers found within the waste) are also made on the basis of visual examination information. A change in this activity would require a HWFP moclification. As previously noted, DOE has already obtained some relief from this requirement from NMED (see Section 3.3.2~. The societal impact of a change to this requirement couicl be a decrease of public 4°The committee referred to requirements in the Land Withdrawal Act, RCRA, Titles 40 CFR 191, 40 CFR 194, or 40 CFR 264. Visual examination is now a regulatory requirement because it has been included in the HWFP.
A Framework for Evaluating Waste Characterization Activities 73 confidence in DOE particularly in case of an accident due to an overlooked item that visual examination and segregation work could have eliminated. 5.9.3.2 Cost-effectiveness analysis Y Real-time radiography is generally an efficient and effective characterization activity. Operators are trained to recognize prohibited items, although they cannot read labels or estimate weights through radiography. This is done by opening the drums and physically handling objects. Radiography is already a confirmation technique for AK; visual examination is now a second type of confirmation. Given the high costs of this activity, the potential for worker exposure, the existence of redundancies in the characterization process to ensure that only waste complying with Waste Acceptance Criteria is shipped to WIPP and the (apparently) marginal benefits of checking relatively accurate radiography results on a small percentage of containers, it appears that visual examination to confirm radiography results on existing waste has tow value. Concerning this characterization activity, the New Mexico Environmental Evaluation Group writes (EEG, 2003b; page 59~: The [visual examination] process has the potential for slightly greater radiation exposure than the other waste characterization requirements, although the EEG has not been provided any data from the DOE to indicate that exposures are significant enough to justify reducing the requirement. The DOE has been successful in modifying the HWFP on reirievably stored visual examination and this would be the preferred process for seeking further reductions. To decide whether a permit modification would improve the program, additional risk information is neeciecl along with an analysis of the policy and societal impacts of a change to this characterization activity.
