10
CONCLUSIONSANDRECOMMENDATIONS

NEED FOR PRIORITY-SETTING PROCESS

The enormous costs and technical limits to cleaning up hazardous-waste sites highlight the need for a more comprehensive, explicit, and systematic approach to setting cleanup priorities.

From a review of the hazardous-waste cleanup problem in the United States, and the technological limits and enormous costs to reaching the original goals for cleanup stipulated by Congress, it is apparent that the original assumption of a few sites needing remediation was incorrect. It is also apparent that there are tens of thousands of sites potentially costing hundreds of billions of dollars to cleanup. Thus, it no longer



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 267
Ranking Hazardous-Waste Sites for Remedial Action 10 CONCLUSIONSANDRECOMMENDATIONS NEED FOR PRIORITY-SETTING PROCESS The enormous costs and technical limits to cleaning up hazardous-waste sites highlight the need for a more comprehensive, explicit, and systematic approach to setting cleanup priorities. From a review of the hazardous-waste cleanup problem in the United States, and the technological limits and enormous costs to reaching the original goals for cleanup stipulated by Congress, it is apparent that the original assumption of a few sites needing remediation was incorrect. It is also apparent that there are tens of thousands of sites potentially costing hundreds of billions of dollars to cleanup. Thus, it no longer

OCR for page 267
Ranking Hazardous-Waste Sites for Remedial Action suffices to have ranking models that only attempt to identify the bad sites. There are too many of them. Faced with this reality, society needs to take the next step and develop an overall priority-setting system that helps define a more comprehensive, explicit, and systematic cleanup strategy, addressing such questions as where the available funds should be spent, and how they should be spent. Toward this end, the committee reviewed and compared the current models for ranking hazardous-waste sites and the overall priority-setting systems being used or proposed by various federal agencies and the states. An approach toward a unified national priority setting system was developed. Conclusions and recommendations reached from the committee's efforts are contained in the following. The committee's general conclusions and recommendations are presented below: first, for the overall priority-setting process for remediation of contaminated sites and, second, for the mathematical models that are used as part of the overall process for ranking sites. Conclusions and recommendations for the priority-setting processes and ranking models of specific agencies are presented in earlier chapters. CURRENT PRIORITY-SETTING The current priority-setting processes for hazardous-waste site cleanup are not well deigned and appear to lack adequate evaluation, sufficient consistency, and effective oversight. Confronting the great number of waste sites and potential hazards that have been identified, EPA, DOE, and DOD had to develop ranking and priority-setting systems for remediation. The missions or mandates for these systems are diverse and complex. The scope and scale of the national effort to remediate waste sites

OCR for page 267
Ranking Hazardous-Waste Sites for Remedial Action in the private and public sectors is much larger than was conceived by any of the agencies. Resource requirements parallel those of other major societal activities and challenges such as national debt reduction, the savings and loan crisis, and infrastructure renewal. Setting priorities for remediating the sites requires a well-organized and well-defined national approach and commitment. This has not yet been achieved. None of the agencies have developed its overall priority-setting process in a manner that is explicit, adequately documented, and sufficiently open to scientific and public scrutiny. The overall process for setting priorities for remediation of hazhazardous-ardous waste sites was found generally to involve three major tiers of activity: (1) site screening to determine which sites will receive a detailed evaluation for further decisions; (2) preparing a detailed technical evaluation of the situation at each site chosen in Step 1; and (3) setting of priorities and procedures for remediation. Most federal and state agencies follow an overall process of setting priorities for site remediation that includes the three phases in some manner. However, much of that process remains opaque and thus potentially lacks credibility. In considering the procedure used by EPA, DOD and DOE, the committee has determined that many of the steps in determining priorities are external to the ranking models that serve as aids in the process, have not been explicitly articulated, and therefore remain obscured from public scrutiny. The DOD and DOE priority-setting processes for their own sites have inadequate independent oversight. Such a process leaves the polluter itself (DOD or DOE) responsible for discovering the pollution, investigating and characterizing the extent of pollution, selecting remedial approaches, setting priorities for remediation of sites, and executing and monitoring the remediation, a situation which cannot help but undermine the credibility of the process.

OCR for page 267
Ranking Hazardous-Waste Sites for Remedial Action There is no Consolidated Ranking of Sites at the National Level. The federal agencies with hazardous-waste sites needing remediation are approaching ranking and setting priorities of sites differently. The ranking and priority-setting processes used by DOD, DOE, and EPA have been considered in detail in this report, but other agencies have different approaches as well. The other agencies include the Department of the Interior, the Department of Transportation, the Small Business Administration, and the National Aeronautics and Space Administration, all of which have hazardous-waste sites on the NPL, and other agencies that have sites on EPA's Federal Facility Docket. Many contaminated sites are not included in the present ranking processes because they are under different programs. An example of this is the exclusion of sites and facilities covered by the Resource Conservation and Recovery Act. Other examples include DOD sites for storing and destruction of chemical and biological weapons (demilitarization program); sites covered by DOD's Defense Environmental Restoration Program; DOD's Base Realignment and Closure sites; DOE's Environmental Restoration Program sites and Waste Management sites, as well as other DOE hazardous-waste sites not covered by either of these programs, e.g., sites under FUSRAP (Formerly Utilized Sites Remedial Action Program). IMPROVING THE PRIORITY-SETTING PROCESS To the maximum extent possible, the overall priority-setting processes, including the mathematical models used, should be similar across the various federal agencies. Escalated remediation costs (actual and projected) and insufficient public accountability of desperate efforts suggest that the existing assortment of processes for priority setting should be

OCR for page 267
Ranking Hazardous-Waste Sites for Remedial Action drastically changed. A unified national process-based on the use of similar information and models at every site and is consistent for all states and areas within states—is recommended. In view of the enormous direct public investment required for remediation of contaminated federal facilities and indirect payment for remediation at private sites, a single well-developed and documented process is needed that not only ensures use of funds on a consistent basis and a proper return to the nation from this investment but also engenders confidence within the scientific community and the public. This uniform national priority-setting process should be more scientifically based, explicit, and open and accessible to the public than has been the case for the three major tiers of the overall priority-setting process discussed in Chapter 9. Openness is at least as important as scientific validity. The complete priority-setting process should be well documented, it should be subject to review by the scientific community and the affected public, and it should explicitly address not only risks to human health and the environment, but also social and economic issues. This consistency and openness should apply to everything from data requirements to requirements for addressing social, economic, and cultural factors. A specific practical implication is that each tier of activity should be similar in scope and content across the various agencies. The priority-setting process should have a common mechanism for identifying serious immediate hazards or emergency conditions and pulling them out of the longer-term priority-setting process; all of the systems the committee studied have some such feature. A unified approach should also include a formal site-discovery program, which is currently lacking. It should also include a process for tracking site remediation progress and monitoring sites that may pose dangers far into the future. Such a national plan or protocol would greatly benefit by a support program, including

OCR for page 267
Ranking Hazardous-Waste Sites for Remedial Action Expanded research and development in the advancement of remediation technologies, analytical methods, knowledge of contaminant movement and fate, methods to measure and estimate health, socioeconomic and environmental impacts, and analyses of institutional barriers retarding the remediation process. Technical and scientific education for federal, state, local and private operatives, as well as foreign partners. Outreach extension, and technical transfer for the federal, state, private, and public sectors. CURRENT RANKING MODELS USED IN PRIORITY-SETTING The formal mathematical models developed to aid in the priority-setting process play little role in determining which sites are ultimately remediated. Much attention has been given to scores from mathematical models developed for site- ranking, which are used as aids in the priority-setting process. However, they are only one of the factors that are ultimately used to determine whether or not a site will be remediated, the degree to which it will be remediated, and when. Key decisions are made external to the models through negotiations and the political process. Site-ranking models would play a greater role in the priority-setting process if they incorporated to a greater extent social and economic values, and if users and the public were more confident in model outcomes. A strong scientific base now exists upon which to build a sound ranking model that could play a larger role in the overall priority-

OCR for page 267
Ranking Hazardous-Waste Sites for Remedial Action setting process. No large apparent gap exists in this base that would require a major research effort for improvement. This scientific base has already been used to a degree in the development of at least parts of the different ranking models. However, all ranking models were found to fall short on several important attributes of model development, including adequate documentation, proper validation, completeness, transparency, and adequate inclusion of social and economic factors. Sufficient attention to these attributes is necessary for user confidence in model outputs. The mathematical models used by EPA, DOE, and DOD as aids to setting priorities differ widely. The three agencies' mathematical models examined by the committee have forms traceable to these agencies individual mandates and to the complexity and number of the sites they manage. EPA needed an early screening model to eliminate the vast majority of nominated sites from further consideration. The major question was what sites should be selected for derailed investigation. DOD needed a screening model for use at a later decision stage when funds for remediation would become inadequate for the need. The question here was when should a given site be remediated. DOE needed a more comprehensive model that could address its fewer but larger facilities, each with many complex sites. The major question was how funds available for remediation of a given site should be distributed among its many contaminated sites in order to optimize reduction of risk to humans and the environment. The three models have similarities in the environmental pathways they consider, but they differ in the stages within the overall priority-setting process at which they are applied. They differ in the types of input information that they use, in the environmental media they consider for transport of contaminants and exposures, in the relative importance they give to human health and the envi-

OCR for page 267
Ranking Hazardous-Waste Sites for Remedial Action ronment, and in how they handle social and economic aspects. They also differ greatly in the weighting given to the different exposure pathways, as well as the selection of the factors used to weight the influence of the environmental and toxicological data. Further, the definition of site is not consistently used within any agency, or between different agencies. For example, EPA's HRS has been used to list the Rocky Mountain Arsenal—an entire DOD installation—on the NPL, and also to list Basin F—one small part of the Rocky Mountain Arsenal—on the NPL. Obviously, the risk from many sites represented by the whole Arsenal poses a greater risk than any single site by itself. Also obviously, the cost for cleanup of the entire Arsenal is greater than the cost for any one of its sites. This example suggests why ranking of sites by the risk posed, when remediation costs are not considered, can be quite misleading. The different models are applied at different steps in the priority-setting process and, as such, have different data and resource needs and provide estimates of site hazard with different levels of accuracy and precision. This might lead to inconsistent rankings between the models, a problem that should be expected and cannot in itself be avoided. However, the relationship between pathway subscores in the three models often differ substantially, as do the weightings given to the different exposure pathways and the selection of factors used to weight the influence of the environmental and toxicological data and the value of natural resources. The different weightings provided often reflect differences in value judgments, but the process by which these value judgments were obtained is not often clear. Because the model results have an effect on the expenditure of vast amounts of public funds, a more consistent and better documented approach to obtaining and using value-laden weighting factors would seem appropriate. State ranking systems for waste sites follow one of several ap-

OCR for page 267
Ranking Hazardous-Waste Sites for Remedial Action proaches: a quantitative approach similar to the EPA HRS, other explicit numeric systems leading to a site-sepcific score, or tend to differentiate all sites into a small number of categories of priority based mainly upon narrative descriptions of site characteristics. For many of the states considered, there is evidence of very thoughtful development of site ranking models. However, how the relationships between the model parameters were developed and what strategies were used for combining parameters is not always clear and often not documented, thus the models tend to lack credibility. For this reason, similar questions of appropriateness of the logic for combining various scores within the ranking methods applies to states' approaches as well as to approaches of federal agencies. States not using formal ranking models often tend to develop less data-intensive methods that rely on the judgment of professionals in the state agencies to integrate information into site rankings. A more detailed evaluation, beyond the scope of this study, would be needed to evaluate the utility of such approaches relative to the mathematical modeling methods used by the federal government and other states. IMPROVING THE MODELS Ranking models can and should play a greater role in the priority-setting process than is currently the case. Models can be important tools in a priority-setting process because they can integrate a wide variety of important technical, social, and economic factors. The committee believes that with achievable upgrades in certain aspects, these models could play a

OCR for page 267
Ranking Hazardous-Waste Sites for Remedial Action more important role in the overall priority-setting process, pointing toward a more equitable distribution of funds and a wiser and more financially sound national effort towards site remediation. Some of these aspects are summarized next. Documentation and Clarity An important part of any model development is documentation that permits reviewers to understand why the models are structured the way they are and the process by which coefficients that reflect value judgments have been derived. In addition, although models may be technically sophisticated, their core elements should, to the maximum extent feasible, be intuitively as clear as possible to technical and nontechnical audiences. These aspects are important to increase the understanding of the model or process and the acceptance of the results produced. The output from a ranking model should provide information in addition to the overall score itself so that one can understand why a high or low score was obtained. The additional information would include individual environmental pathway scores, whether site contaminants pose acute or chronic risks, and how the model's value-weights affect the overall score. Public Involvement The process of developing a model (or any major component of the model) should be as open as possible, involving both stake-holders and the technical community. Value preferences should be explicit in the models, and coefficients reflecting these preferences should be developed with the affected parties in an open and well-defined process. The process of applying the model to a given

OCR for page 267
Ranking Hazardous-Waste Sites for Remedial Action site (or to a large installation such as a military base or a DOE facility) should be similarly open, so that there is the greatest understanding of the results of the model. Validation The development and introduction of any important decision-aiding model, such as those under discussion here, should include an explicit process for validating the components of the model and the overall model itself. Flexibility should be provided for revising the components of the model to reflect new knowledge. Adequate validation to objective criteria might require the development of a collection of test sites that have agreed-upon priority rankings, resulting from a comprehensive evaluation, against which to compare the results of model output. The purpose is to test model results and to build user confidence in model outcomes. Although it is not validation in the strict sense, comparing the performance of one ranking model with performances of other ranking models is a useful exercise (see Chapter 8). The approach should be used on a regular basis in the future to compare the performance of newly developed or modified models against the output of established ones. A set of reference sites should be established for use in developing input data for a wide range of hazardous potential (high to low). Using a set of about 12 dissimilar sites would help ensure that the models are compared on the basis of a broad range of site characteristics. The approach would be useful for checking whether revised algorithms are performing as expected or whether the models can discriminate, in a numerical sense, among various degrees of potential site hazards in the range of high to low.

OCR for page 267
Ranking Hazardous-Waste Sites for Remedial Action Explicit Consideration of Socioeconomic Effects Consideration of danger to public welfare by EPA is required under CERCLA, and part of the Agency for Toxic Substances and Disease Registry's mandate in SARA is to consider negative effects (of waste-site contaminants) on quality of life. A comprehensive site evaluation model should include explicit considerations not only of human health and the environment, but also of socioeconomic impacts on the surrounding community. Such considerations are probably always part of the priority-setting process, but they generally are not made explicitly, and so are not open to public scrutiny and evaluation. If this important element in setting priorities can be given a common explicit basis, then greater confidence in the overall process will be achieved. Methodologies that allow the incorporation of rigorous socioeconomic impact assessments directly into models for ranking hazardous sites are currently available. Omission of an explicit treatment of these socioeconomic components in hazard ranking models can lead to a biased priority-setting agenda.