Executive Summary

Activities of the U.S. Navy over the past century have had substantial impact on the environment. Some of the most significant environmental liabilities the Navy is facing are the 4,448 waste sites at Navy installations across the country. At almost every active and closing Navy base, soils, sediment, or ground water have been exposed to chemical contamination to some extent. At large facilities, there may be as many as 100 individual sites requiring remediation. Cleaning up these sites is a major goal of the Navy's Environmental Restoration Program.

Contaminants and site conditions at Naval facilities are highly variable, with such recalcitrant compounds as metals, chlorinated solvents, and hydrophobic organic chemicals being common concerns. Cleanup efforts involving these contaminants can be frustrating, time consuming, and expensive, and sometimes can have limited results in terms of contaminant removal. Thus, the Navy requested the National Research Council (NRC) to examine how the Navy can speed up the pace of cleanup activities and lower the overall cost of the Environmental Restoration Program. This report presents the results of the first phase of the NRC study, which focused on the use of risk-based methodologies for increasing the speed and cost-effectiveness of the Navy's environmental remediation projects.

Approximately 66 percent of the Navy's hazardous waste sites are subject to regulation under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA). Because the average time for construction of a cleanup remedy is 10.6 years, the CERCLA process is perceived by many as slow and costly. The following problems are partially responsible for this general impression: (1) the way CERCLA is conducted, (2) complex contaminant and site characteristics, (3) unstructured risk management decision making, (4) a lack of technological solutions to address the contamination, and (5) disagreements



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



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

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

OCR for page 1
--> Executive Summary Activities of the U.S. Navy over the past century have had substantial impact on the environment. Some of the most significant environmental liabilities the Navy is facing are the 4,448 waste sites at Navy installations across the country. At almost every active and closing Navy base, soils, sediment, or ground water have been exposed to chemical contamination to some extent. At large facilities, there may be as many as 100 individual sites requiring remediation. Cleaning up these sites is a major goal of the Navy's Environmental Restoration Program. Contaminants and site conditions at Naval facilities are highly variable, with such recalcitrant compounds as metals, chlorinated solvents, and hydrophobic organic chemicals being common concerns. Cleanup efforts involving these contaminants can be frustrating, time consuming, and expensive, and sometimes can have limited results in terms of contaminant removal. Thus, the Navy requested the National Research Council (NRC) to examine how the Navy can speed up the pace of cleanup activities and lower the overall cost of the Environmental Restoration Program. This report presents the results of the first phase of the NRC study, which focused on the use of risk-based methodologies for increasing the speed and cost-effectiveness of the Navy's environmental remediation projects. Approximately 66 percent of the Navy's hazardous waste sites are subject to regulation under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA). Because the average time for construction of a cleanup remedy is 10.6 years, the CERCLA process is perceived by many as slow and costly. The following problems are partially responsible for this general impression: (1) the way CERCLA is conducted, (2) complex contaminant and site characteristics, (3) unstructured risk management decision making, (4) a lack of technological solutions to address the contamination, and (5) disagreements

OCR for page 1
--> among responsible parties, regulatory authorities, and surrounding communities. Because the bulk of cleanup time is spent on initial site characterization, the impression of responsible parties, such as the Navy, is that although money is being spent, little actual cleanup is occurring. Limited financial resources and the inability of remedial technologies to achieve background concentrations have increasingly forced environmental managers to reconsider the concept of resource conservation (that is, restoration of sites to near-natural conditions) as the guiding principle for site restoration. There has been a shift toward considering overall reduction of risk to humans and the environment as the main driver for remediation. At the time of the Navy's request for guidance from the NRC, several studies and other efforts were under way to evaluate the use of risk-based methodologies for cleaning up waste sites. In particular, in 1995 the American Society for Testing and Materials (ASTM) released a risk-based methodology known as Risk-Based Corrective Action (RBCA) Applied at Petroleum Release Sites (''petroleum RBCA"). ASTM RBCA is being rapidly adopted by many states for cleanup of leaking petroleum underground storage tanks. A second ASTM RBCA standard guide, for contaminants other than petroleum hydrocarbons, has also been developed ("chemical RBCA"). Encouraged by RBCA's growing role in cleanup programs, the Navy asked the NRC to evaluate ASTM's standard guides and suggest whether those methodologies, or a similar but simpler approach, could be used to rapidly close out Navy sites. This report examines how risk-based cleanup operates under CERCLA and whether other methodologies can appropriately expedite the cleanup process. Review of Risk-Based Methodologies The committee focused its review of risk-based methodologies on the ASTM RBCA standard guides and the CERCLA process (as described in numerous EPA documents) for two reasons: (1) although not the only approaches available, they are the most widely used and (2) most other methodologies investigated did not differ appreciably from those of the EPA and ASTM. Other frameworks considered by the committee included the Air Force's Enhanced Site-Specific Risk Assessment, the EPA Science Advisory Board's Integrated Risk Project, the Lawrence Livermore National Laboratory/University of California methodology, and numerous state variations on the practices and methods of ASTM and the EPA. EPA Risk-Based Methodologies Because most Navy sites are regulated under the CERCLA cleanup process, CERCLA is the standard to which all other methodologies should be compared. During the initial phase of CERCLA, a preliminary assessment and a site inspection are conducted to collect data and determine the need for further cleanup action. If contamination is found, a detailed remedial investigation (RI) is con-

OCR for page 1
--> ducted, which involves further site characterization and the elucidation of potential exposure pathways. During the RI, human health and ecological risk assessments are conducted, using procedures outlined in the EPA's Risk Assessment Guidance for Superfund (RAGS). Estimates of risk are determined using models of contaminant fate and transport and human and ecological exposure to those contaminants. When the risk estimate is greater than a specified acceptable risk level, RAGS can be used to develop preliminary remediation goals. For soil contamination only, the EPA has developed the Soil Screening Guidance for closing out low risk sites and arriving at preliminary remediation goals. The Soil Screening Guidance is a tiered approach during which site data are compared to either generic or site-specific screening levels. If concentrations are below the screening levels, the screening process ends, and a site can most likely be closed with no further action. If concentrations are above screening levels, then additional data collection is performed to generate more site-specific cleanup levels. Information from the RI, including the results of all risk assessments and preliminary remediation goals derived from RAGS or the Soil Screening Guidance, is used to conduct a feasibility study to determine the ultimate remedy. During the feasibility study, alternative remedial measures are identified, evaluated, and compared. Although the EPA does allow for a combination of remedial options, it has a preference for options that focus on removal of contaminant sources. The final stages of the CERCLA process include maintaining engineering and institutional controls and conducting long-term monitoring. ASTM RBCA Standard Guides Like the EPA's Soil Screening Guidance, both of ASTM's RBCA standard guides (for petroleum release sites and for sites containing other chemical contaminants) are based on a tiered approach for evaluating the risk of waste sites. However, unlike the Soil Screening Guidance, ASTM RBCA uses the tiered approach for all types of contamination and contaminated media. The first phase of ASTM RBCA is an initial site assessment to gather information about the chemicals of concern (sources), the migration pathways of the contaminants (pathways), and potential human and ecological receptors (receptors). This information allows an initial classification of the risk posed by the site and can direct emergency response actions. A tier 1 evaluation then commences, during which the concentrations of contaminants at the site are compared with generic, conservative cleanup levels. If the contaminants are present in amounts greater than the tier 1 cleanup levels, then either the site can be cleaned up to reduce the concentrations of the contaminants to below those levels, or a tier 2 evaluation can commence to better characterize the site and develop site-specific cleanup levels. At tier 2, additional site characterization is conducted to generate more site-

OCR for page 1
--> specific and less conservative cleanup goals. These goals are the result of risk assessment calculations that take into account an acceptable risk level, models of fate and transport, and models of human and ecological exposure to contamination. ASTM RBCA recommends use of mathematical models similar to those found in RAGS. Contaminant concentrations are compared to the site-specific cleanup levels, and a choice is made to either close the site, clean the site to the site-specific cleanup levels, or proceed to a tier 3 evaluation. Tier 3 contains an additional layer of complexity as more refined site-specific cleanup levels are generated for comparison to contaminant concentrations. Throughout ASTM RBCA, the chosen path is determined by comparing the cost of cleaning the site to the cost of data collection necessary for the higher tiers. Tier evaluations are followed by remedial actions, which eventually lead to closure of the site. Under the ASTM RBCA methodology, the remedy may include a combination of removal actions, treatment technologies, engineering controls, and institutional controls. A committee survey of 20 state regulatory agencies indicated that the use of risk-based methodologies is increasing, primarily for the cleanup of petroleum hydrocarbons but also for more recalcitrant compounds. In many cases, these methodologies are state-customized versions of the petroleum RBCA standard guide. As explained further in Chapter 2, most of the states have started developing the appropriate tools that will enable them to fully implement risk-based approaches. However, the states will require further guidance on issues such as natural attenuation, ecological risk assessment, and the use of institutional controls. Strengths and Weaknesses of Risk-Based Methodologies Risk-based approaches to environmental remediation have certain inherent strengths and weaknesses compared to resource conservation and technology-based approaches, which focus on source removal. These strengths and weaknesses, which must be considered prior to the implementation of risk-based approaches, are manifested differently in the EPA and ASTM methodologies. In general, ASTM RBCA focuses less on source removal than CERCLA. Strengths of Risk-Based Approaches Risk-based approaches tend to be systematic in order to characterize sites and quantify the risks they pose. This organizational structure is often embodied in a tiered approach. In order to be protective of human health and the environment, risk-based approaches depend on extensive data collection and site characterization. The risk assessment calculations that are part of such approaches are

OCR for page 1
--> based on scientific descriptions of contaminant transport and human and ecological exposure and can incorporate new information as it becomes available. Risk-based approaches allow prioritization of contaminated sites based on their relative risk, which can, in turn, aid in the allocation of resources. For example, money can be spent on sites that pose the highest risks, or at sites where the greatest risk reduction per dollar is achievable. Weaknesses of Risk-Based Approaches Risk-based approaches are more likely than resource conservation or technology-based approaches to result in remedies that leave contamination in place. This can be problematic if there are unidentified, potentially dangerous contaminants remaining on site, or if the behavior of known contaminants is poorly characterized. Remedies that leave contamination in place often rely on institutional controls, which, for both legal and physical reasons, are very difficult to enforce. Risk-based approaches may also discourage the development of innovative technologies for source removal. The greater potential for leaving contamination in place introduces significant uncertainties into any risk-based approach. Quantifying the sources of contamination, the pathways of contamination migration, and the potential receptors generates significant uncertainty, much of which cannot be reduced. If recalcitrant contaminants are involved, such as chlorinated solvents and metals, uncertainties increase dramatically, because less is known about the toxicological effects, the bioavailability, and the fate and transport of such contaminants in comparison to petroleum hydrocarbons. There are also inherent uncertainties in the effectiveness of remedial options, including treatment systems, engineering controls, and institutional controls. Technological solutions are often uncertain because of unknowns about how well a particular technology will work at a given site. The effectiveness of institutional controls has also not been well established. Chapter 4 of this report discusses methods for quantifying and reducing uncertainties associated both with risk assessment and risk management. Comparing the Elements of ASTM RBCA and CERCLA As part of this study, the committee compared the characteristics of the ASTM RBCA and CERCLA methodologies. For the most part, the two approaches are remarkably similar. One procedural difference is that with the ASTM methodology remedial options are available earlier in the process, and the tiered approach is used for all types of contaminated media. There is a general perception that ASTM RBCA is faster, less costly, and more user friendly than CERCLA. Finally, CERCLA received more public input during its design and requires more public involvement during implementation than is inferred in the ASTM RBCA guidance documents.

OCR for page 1
--> Strengths and Weaknesses of ASTM RBCA The strengths and weaknesses of a generic risk-based methodology are more evident in the ASTM RBCA methodology than in the CERCLA process. The tiered approach of ASTM RBCA is highly systematic and can make a large number of diverse sites more manageable. The prioritization of sites is facilitated in both standard guides by a table that categorizes sites into four initial risk levels. Although there is disagreement over the level of site characterization needed for a tier 1 analysis, extensive data collection is required at tiers 2 and 3. There are other important strengths of the ASTM standard guides. ASTM RBCA may accelerate the cleanup process and expedite site closure because (1) management options, including remedial options, are available during each tier evaluation, (2) it allows the use of institutional controls, which are perceived as requiring minimal cleanup effort, and (3) the assessment procedures are standardized for all types of contaminated sites. The petroleum RBCA standard guide's streamlined documentation has facilitated its comprehension and adoption by both states and industry professionals. The ASTM RBCA methodology suffers from the weaknesses inherent in any risk-based approach. Because many remedial options available under ASTM RBCA leave contamination in place, the threat of unidentified compounds posing a risk to receptors remains. Petroleum RBCA does not discuss long-term risk and the need to revisit sites if new sources of contamination are found. Chemical RBCA is generally more cognizant of long-term risks, but it contains a provision that might allow sites to be closed prior to a tier 1 evaluation. The treatment of uncertainty in ASTM RBCA is minimal and inadequate, particularly at tiers 1 and 2. Uncertainty regarding the effectiveness of remedial options is not discussed at all. Finally, there are weaknesses specific to ASTM RBCA irrespective of risk-based approaches. Insufficient consideration is given to the cumulative effects of multiple contaminants and multiple exposure pathways. There is also significant potential for inappropriate application of the methodology. Although not the intention of its designers, the example risk assessment calculations printed in the standard guide, which are intended to serve only as illustrations, have the potential to be inappropriately used during implementation of ASTM RBCA. Petroleum RBCA does not promote public involvement at an early stage in the cleanup process, although this is somewhat corrected in chemical RBCA. Finally, like other ASTM voluntary standard guides, both ASTM RBCA standard guides were designed and approved by a limited group of stakeholders. Strengths and Weaknesses of the CERCLA Process The CERCLA process promotes extensive site characterization and relies on scientifically developed risk assessment calculations. However, CERCLA is not

OCR for page 1
--> as obviously systematic as ASTM RBCA. The tiered approach, which provides organization for examining large numbers of sites, is available only for certain types of soil contamination. Generic soil and ground water screening levels developed by the states do not necessarily impart greater consistency to the CERCLA process. CERCLA does not provide an easy means for prioritizing sites for further action. The Hazard Ranking System, used to determine whether properties should be placed on the National Priorities List, does not distinguish the relative risk between sites on a property. Thus, to some degree, the CERCLA process manifests fewer of the strengths of a generic risk-based approach than the ASTM methodology. CERCLA has a long history of use in environmental remediation at federal facilities and is familiar to Navy remedial project managers. Public involvement occurs at explicit stages of the CERCLA process, which is important for ensuring acceptance of the remedial option. Because CERCLA was generated through a legislative process, it is less likely to be tied to any special interest other than environmental protection. Thus, the public may perceive CERCLA as providing the best assurance of long-term commitments, such as the maintenance of long-term monitoring and institutional controls. As one might expect, CERCLA manifests fewer of the weaknesses of a generic risk-based methodology. This is because CERCLA has a strong preference for removing sources of contamination rather than relying on engineering and institutional controls. Thus, the threat of unknown contamination and contamination remaining in place is considerably lessened at sites using the CERCLA process. The EPA acknowledges the importance of uncertainty during the risk assessment process, although there is still minimal guidance on how to quantitatively incorporate uncertainty into the CERCLA process. Eleven Criteria for a Risk-Based Methodology The committee's evaluation of CERCLA, ASTM RBCA, and other risk-based methodologies illuminated many important features of risk-based approaches, both positive and negative. After comparing these evaluations to the pressing environmental problems faced by the Navy, the committee identified 11 criteria that it believes should be necessary components of any risk-based approach adopted by the Navy. 1.   It should facilitate prioritization of contaminated sites at individual installations. The ability to rank sites according to their relative risk is especially important at facilities that have multiple waste sites, as at Navy installations and other federal facilities. Ranking of sites can help determine which sites pose immediate threats to human health and the environment, which sites should be cleaned up first, and where to allocate funding. Prioritization does not have to occur with the same risk assessment and risk management paradigms that charac-

OCR for page 1
--> terize the complete methodology. However, data collected for prioritization should be useful in the larger risk framework. 2.   It should provide a consistent mechanism for addressing all types of sites. Because Navy facilities include both simple low risk sites and complex high risk sites, the Navy's risk-based methodology must include a systematic approach for conducting site investigations of differing complexity. In facilitating the evaluation of sites ranging from simple to highly complex, both conservative, generic cleanup goals and less conservative but more site-specific cleanup goals should be allowed. Site-specific cleanup goals should be allowed to replace conservative cleanup goals for well-defined contamination problems. To provide equivalent protection for poorly-defined contamination problems, generic cleanup goals should reflect a level of conservatism consistent with the lack of knowledge about the site. 3.   It should provide guidance on data collection needed to support the development of site-specific cleanup goals. Because risk-based approaches are more likely than resource conservation or technology-based approaches to suggest remedies in which contamination is left in place, extensive site characterization and data collection must accompany their use to ensure protection of human health and the environment. A satisfactory risk-based methodology will provide guidance on data collection methods and data quality objectives. It will suggest ways to conduct an initial site assessment that will aid in prioritization of sites, as well as methods of site characterization that can be used to generate site-specific cleanup goals. 4.   It should provide for integrated assessment of sites affecting the same human or ecological receptors. At large facilities containing multiple hazardous waste sites, the risk from exposure to contamination may derive from multiple sources. A risk-based methodology must be able to quantitatively assess all sources that are affecting individual human or ecological receptors. This includes cumulative risk assessment, which evaluates the effects on potential target receptors of multiple chemicals and multiple exposure pathways originating from a single waste site, as well as an "integrated assessment" of sites, which is conducted to determine the overall facility-wide risk encountered by potential target receptors. 5.   It should encourage early action at sites where the risk to human health and the environment is imminent and for which the risks are demonstrably low and remediation is likely to be more rapid and inexpensive. The risk-based methodology should allow for source removal action and pathway interruption to eliminate risk to potential receptors before the remedial option has been chosen. It should facilitate rapid action at sites that are easy and inexpen-

OCR for page 1
--> sive to remediate, as defined by the nature of the contaminants present and the affected subsurface. If all contamination is not removed as a result of these early actions, the methodology should evaluate the long-term consequences of removal actions and pathway interruptions through long-term monitoring and, if necessary, further remedial actions. 6.   It should consider relevant uncertainties. There is significant overall uncertainty in risk-based approaches to remediation (e.g., model and data uncertainty, lack of understanding concerning ecological risk, uncertainty in dose-toxicity relationships, and doubt regarding the enforceability of institutional controls). This uncertainty directly translates to uncertainty in risk estimates, associated cleanup levels, and the effectiveness of the proposed solutions. The Navy's risk-based methodology should allow for the explicit consideration of these uncertainties and suggest options for reducing them, using both long-term monitoring and formal uncertainty analyses. Uncertainty analyses within the methodology should use stochastic modeling approaches, such as Monte Carlo analysis, to represent the degree of uncertainty associated with the risk estimates. These approaches may generate significant cost savings if they lead to less overestimation of the risk than do conservative cleanup criteria. 7.   It should provide a mechanism for integrating the selection of the remedial option with the establishment of remedial goals. It should also provide quantitative tools for developing risk management strategies. Because there are often no technological solutions capable of reducing contamination to background or health-based concentrations, the process by which cleanup goals are set and remedies are selected must be flexible. The risk-based methodology should use an iterative approach during which the responsible party and oversight agencies cycle back and forth between these issues. In addition, the evaluation of multiple remedial options should take place in a quantitative framework to the greatest extent possible. Because short- and long-term effectiveness of the remedy and cost are amenable to quantitative analyses, the Navy's risk-based methodology should contain the appropriate quantitative tools to balance these criteria. 8.   It should have options to revisit sites over the long-term. Risk-based approaches to managing sites must provide a convincing argument for controlling future risks, both short-term and long-term. Thus, a risk-based methodology may call for repeated risk evaluations if site conditions change over time. Long-term considerations of risk will enhance the credibility of risk-based approaches with both the regulatory agencies and the affected public. 9.   It should be implemented in a public setting with all stakeholders involved. Implementation of a risk-based methodology should take place in a public forum. The science and assumptions used during the risk assessment pro-

OCR for page 1
--> cess, the proposed remedies, and all relevant value judgments must be presented to and understood by interested stakeholders. To facilitate public scrutiny and debate, the methodology must be explicit and traceable. 10.   Its guidance document should undergo both external, independent scientific peer review and public review. Any proposed risk-based methodology must be the product of rigorous, broad, and external scientific peer review. In addition, all interested stakeholders must be allowed to review the approach and supporting policies during a public comment period prior to its implementation. 11.   It should comply with relevant state and federal statutory programs for environmental cleanup. Any risk-based methodology must comply with relevant laws and regulations. Thus, it must be flexible enough to incorporate the different regulatory policies of the states and the EPA. Comparing ASTM RBCA to the Eleven Criteria The ASTM RBCA methodologies, as strictly interpreted from the guidance documents, minimally satisfy only five of the eleven criteria outlined above. ASTM RBCA is able to prioritize sites by risk and thus meets the first criterion. Ranking of sites takes place during the initial site assessment and leads to site classifications of immediate threat, short-term threat, long-term threat, or no demonstrable threat. The tiered approach of RBCA provides a mechanism for addressing all site types and for replacing conservative cleanup goals with site-specific cleanup goals, and thus this process meets the second criterion. Generic standards are used for tier 1 cleanups, while tiers 2 and 3 require site-specific data. Data collection and site characterization are addressed in the ASTM documents, particularly in the chemical RBCA standard guide. However, guidance on monitoring methods, minimum sample size, and method performance criteria is lacking. Therefore, the ASTM methodology minimally satisfies the third criterion. ASTM RBCA allows for source removal actions and actions to interrupt pathways of contaminant migration, meeting the fifth criterion. Once an early action is completed, the user returns to the main cleanup process. This ensures that early actions do not lead directly to site closure. The seventh criterion is partially satisfied by ASTM RBCA, which allows for flexibility in the setting of cleanup goals and the selection of remedial options. During each tier evaluation, either remedial action can be taken or more site-specific cleanup goals can be generated. The RBCA methodology does not provide a quantitative framework for analyzing the cost and effectiveness of different remedial options. The ASTM RBCA methodology does not satisfy six of the eleven criteria. Although chemical RBCA calls for cumulative risk assessment during a tier 2 evaluation, it does not require an integrated assessment of all sites affecting indi-

OCR for page 1
--> vidual receptors and consequently does not satisfy the fourth criterion. The ASTM RBCA methodology does not sufficiently consider uncertainty in the data and the modeling used in the risk assessment process, a requirement of the sixth criterion. The ASTM RBCA approach to modeling is inherently deterministic and ignores the uncertainties associated with subsurface and exposure pathway modeling. Monte Carlo analysis, which ASTM RBCA suggests as a part of a tier 3 analysis, should be used for tier 2 sites as well. Finally, none of the uncertainties associated with remedial options are mentioned in ASTM RBCA. ASTM RBCA's handling of long-term risk varies between the two documents. Petroleum RBCA provides almost no guidance on long-term monitoring, and it does not discuss assessing the long-term risks of a hazardous waste site. Chemical RBCA, on the other hand, clearly states the purpose of long-term monitoring and outlines actions that should be taken if remedial options are ineffective or if there are future changes in land use. However, chemical RBCA contains a provision in its tier 1 evaluation that, depending on its interpretation by the states, might lead to site closure prior to any comparison between site conditions and generic cleanup levels. Thus, neither RBCA standard guide fully satisfies the eighth criterion. Public involvement is not discussed in the petroleum RBCA standard guide. Chemical RBCA mentions public involvement during the collection of site assessment information, following site classification, and during remedy selection, but public involvement does not appear at explicit points in the RBCA flowchart. The ninth criterion, therefore, is not adequately satisfied by either RBCA standard guide. The ASTM standard guides were created and approved by a limited group of stakeholders consisting of oil and chemical industry representatives, some state regulators, and the EPA (which abstained from voting in favor of chemical RBCA). The RBCA standard guides did not undergo external scientific peer review nor a public review period, and therefore they do not satisfy the committee's tenth criterion. Finally, although the RBCA methodologies outline a framework that encompasses all stages of cleanup, from site discovery to site closure, it is not clear that this framework can be integrated into existing state and federal environmental statutes. Especially for those sites regulated under CERCLA, the use of either RBCA standard guide may present legal and administrative problems for the Navy. Thus, the ASTM RBCA methodology does not satisfy the eleventh and final criterion. Recommendations The characteristics of Navy facilities that make them different from other hazardous waste sites include the wide range of activities that generate waste, the large amounts and types of chemicals that have been disposed of, the poor record-keeping associated with hazardous waste disposal, and the rushed time line on which many of the facilities are expected to close. These features of Navy instal-

OCR for page 1
--> lations require the adoption of a highly flexible, yet protective, risk-based methodology, if one is to be adopted at all. Significant weaknesses of existing risk-based methodologies prevent the committee from fully endorsing their use at Navy facilities unless those weaknesses are corrected. Use of the ASTM RBCA Standard Guides The committee does not recommend the adoption of either the ASTM petroleum or chemical RBCA standard guides at Navy facilities unless they are modified to satisfy the 11 criteria outlined above. As currently formulated, the ASTM RBCA methodology does not satisfy all of the 11 criteria listed above as characteristics of an effective risk-based methodology. Second, the perceived time and cost savings associated with the use of ASTM RBCA at petroleum underground storage tanks are less likely to accrue to the Navy because of the complexity of its hazardous waste sites. This is especially true for those sites that would have to be evaluated with a tier 3 evaluation. Navy facility cleanups are overseen by a wide range of federal and state regulatory agencies, often with inconsistent environmental policies. The reopening of Federal Facilities Agreements to incorporate RBCA could present legal difficulties and public opposition, and ultimately delay cleanup. A Risk-Based Approach for the Navy The committee recommends that the Navy develop a risk-based methodology to be used specifically for its Environmental Restoration Program that satisfies the 11 criteria listed above. These criteria overcome many of the weaknesses associated with the risk-based methodologies evaluated by the committee. In particular, if a risk-based methodology gives sufficient attention to long-term risks, the presence of potentially dangerous unidentified compounds and the use of institutional controls become less problematic. Public trust in such an approach is likely to be greater because of the continued involvement of the responsible party through long-term monitoring and enforcement of institutional controls. Because of the diversity of Navy facilities and sites in those facilities, the committee recommends developing a broad framework that can be customized for use at individual facilities. A possible starting point for constructing this framework could be the ASTM RBCA methodology, but this is not a requirement. The framework should be pilot tested at a single facility before being adopted on a widespread basis. The 11 criteria identified as important for any risk-based approach to environmental remediation point to several avenues for further study. First, the process by which the remedial option is chosen requires significant analysis, explanation, and refinement. One possible study could expand on the second

OCR for page 1
--> statement in the seventh criterion. The goal of this study would be to develop a quantitative analytical tool for evaluating the cost and effectiveness of remedial options. As currently envisioned by the committee, this tool would be based on the paradigm of optimizing risk reduction per dollar spent for a suite of available remedial options. Another important aspect of a risk-based approach is its ability to assess risk over the long-term. Improving long-term monitoring at sites where contamination is left in place will be critical to the use of a risk-based approach at Navy facilities. Techniques that would allow for successful and cost-effective monitoring of the site's responses to natural processes, treatment technologies, containment technologies, and institutional controls should be the focus of such efforts. Finally, assessing and reducing uncertainties associated with both risk assessment and risk management should accompany the implementation of a risk-based approach to environmental remediation. It will be important for the Navy to develop guidelines aimed at remedial project managers for (1) using more sophisticated analytical techniques during risk assessment, (2) reducing the uncertainty associated with site characterization and treatment technologies, (3) conducting uncertainty analyses when appropriate, and (4) enforcement of institutional controls. Success Metrics The committee recommends that the military develop better terminology and metrics of success. Inconsistencies in the language used to describe stages in the Environmental Restoration Program have made it difficult for the Navy to measure the success of its remediation efforts. The military has recently suggested clarifications of terminology for the end stages of the cleanup process, including more precise definitions of the terms ''response complete" and "site close-out." The committee recommends that the military reach consensus on processes that use or adapt this terminology. The committee further recommends that project managers—for regulatory agencies as well as the military—be quickly apprised of the terminology and associated goals. In the committee's opinion, the Navy's Environmental Restoration Program focuses too heavily on "site close-out" as its metric of success. If a risk-based approach is to be developed, the committee recommends that the Navy put more emphasis on "response complete" as the most important metric of success. "Response complete" more adequately addresses progress at sites where contamination remains in place by rewarding the effective operation of the remedial option(s). Such a metric would reward progress toward long-term monitoring, containment, and enforcement of institutional controls, but it would also imply that continued monitoring or other responses may be necessary. Developing improved metrics of success for sites in which contamination remains in place may

OCR for page 1
--> increase the public acceptability of risk-based approaches to environmental remediation. In summary, the committee does not fully endorse the ASTM RBCA standard guides for implementation at Navy facilities. However, recognizing that a consideration of risk during environmental remediation can help to better allocate limited resources, the committee suggests 11 criteria that the Navy should incorporate into a risk-based methodology for use at its facilities. These criteria overcome many of the weaknesses of risk-based methodologies in general and afford a level of scientific credibility and protectiveness of human health and the environment appropriate for the Navy's Environmental Restoration Program.