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.
Overview One of the most controversial and difficult decisions facing public policymakers and regulatory agencies responsible for the remediation of contamination at hazardous waste sites is the defi- nition of cleanup levels for environmental media found to contain toxic or hazardous materials. Since the passage of the Comprehen- sive Environmental Response, Compensation, and Liability Act (CERCLA) in 1980, this issue has demanded an ever-increasing level of effort by all participating parties, including affected com- munity groups, environmental organizations, generators, remedi- ation contractors, environmental lawyers, and regulatory agencies at all levels. The stakes in this debate are high. Human health is at risk. The cost of the remedial measures required to achieve background levels or other conservative cleanup levels for a given contaminant could exceed the industry's, and eventually the public's, capacity to fund. In some cases, no existing technologies can achieve these Tow levels. On the other hand, setting less stringent cleanup levels based on practicality or cost-effectiveness could result in solutions that cause unanticipated harm to human health and lead to costly legal battles between governmental agencies and impacted or inter- ested parties most notably various environmental organizations and citizens groups. Given the contentious nature of this issue and the critical role played by water science and technology, the Water Science
2 HAZARDOUS WASTE SITE MANAGEMENT and Technology Board (WSTB) selected as the topic for its third colloquium the question of setting water quality goals at hazardous waste sites. The colloquium's objective was to evaluate whether the scientific, technical, and regulatory methods currently used for setting cleanup levels are adequate, and, if they are not, to suggest areas for improvements. Within the limits of the available time and resources for WSTB colloquia, this was an ambitious undertaking. It required, first, that the current procedures used throughout the United States be discu~ed critically, and second, that the scientific bases for the decision process be presented and critically assessed. These requirements formed the basis of the format used in the 1 2-day meeting. Three speakers were asked to address current approaches used by regulatory agencies and the U.S. Department of De- fense (DOD) in addressing water quality cleanup levels. The views of impacted parties regarding the adequacy of these ap- proaches were also presented by representatives from water utility, industry, and environmental groups. Finally, two speakers were asked to assess the adequacy of two key scientific areas that play a major role in specifying cleanup levels, namely, ground water modeling and risk assessment. Each issue paper was followed by a formal critique, and workshops were held to assess the scien- tific and technical bases used in the standard-setting process (the workshops addressed engineering, hydrogeology, risk assessment, and regulatory strategies). The recent passage of the Superfund Amendments and Reau- thorization Act (SARA) in November 1986 has increased the im- portance of this colloquium. Whereas the number of sites known or expected to contaminate ground water grows rapidly, the num- ber of sites at which hazards have been eliminated remains low. Of the approximately 23,000 potential sites listed by the U.S. Environmental Protection Agency (EPA), approximately 900 are currently (1987) listed or proposed for listing on the National Pri- ority List (NPL). According to J. Winston Porter, EPA's assistant administrator for solid waste and emergency response, speaking at a recent conference in Washington, D.C. (November 1986), de- cisions on remedial action, and, implicitly, decisions on acceptable cleanup levels in affected media have been made at approximately 130 of these sites. In addition to NPL sites, there are nearly 7,000 treatment, storage, and disposal facilities regulated under the Re- source Conservation and Recovery Act (RCRA), 911 operating
OVERVIEW 3 DOD facilities, more than 15,000 presumably nonhazardous land- fi~Is, and perhaps hundreds of thousands of leaking underground storage tanks from which contamination of soil, ground water, or surface waters may have occurred or may be about to occur. Thus, the backlog of sites at which remediation will be re- quired is large, and the selection of cleanup levels will be a major focus of federal, state, and local regulatory agencies for many years to come. The impact of this process on the ultimate costs of hazardous waste site remediation is uncertain. As of 1986, EPA estimated that remedial actions at an NPL site cost an average of $8.6 million. With the passage of SARA, these costs are expected to increase. Section 121 of that law stressed the importance of selecting permanent remedies to the maximum extent practica- ble. The cleanup levels specified for ground water at Superfund sites must be based on applicable or relevant and appropriate requirements (ARARs) provided by other federal environmental statutes such as the Clean Water Act and the Safe Drinking Water Act. Of particular significance was the requirement that maximum contaminant level goals, which are designated by EPA's Office of Drinking Water (ODW), must be achieved by any remedial action, provided such goals are technically feasible when cost is taken into consideration. Based on a memorandum of understanding between ODW and EPA's Office of Emergency' and Remedial Response, it now appears that maximum contaminant levels will be the ARARs used for site remediation. Prior to the passage of SARA, the resolution of conflicts over the most appropriate remedial action at Superfund sites was han- dIed on a case-by-case basis. State regulatory agencies have fol- lowed a similar strategy. Implicit in the decisions reached on ap- propriate remediation programs was the setting of target cleanup levels for affected environmental media. Prior to SARA the var- ious forms of legal settlements between regulatory agencies and responsible parties reached at Superfund sites illustrated the di- versity of methods used and the lack of consistent guidelines. The types of agreements included the following: . cash "buy-outs," in which responsible parties agreed to pay a certain amount to regulatory agencies for the relief of future liability, without regard to specific cleanup levels; . agreements to conduct specific remediation activities, with- out designated cleanup levels;
4 HAZARDOUS WASTE SITE MANAGEMENT ~ the specification of "requisite" remedial technologies to eliminate hazards to human health and the environment; . open-ended commitments to do "whatever is necessary" to protect human health; and ~ agreements to remediate until specified cleanup levels are achieved. Only in the latter case have environmental criteria or standards been used explicitly to select the remedial alternative. This lack of a consistent method for setting target cleanup levels was a primary concern of EPA and impacted parties, and it led to the explicit language on this issue in Section 121 of SARA. Section 121 was an important and controversial attempt to produce uniformity in the remediation process according to ex- plicit criteria and standards. But has Section 121 provided explicit and unequivocal guidance to EPA and other regulatory agencies for setting cleanup levels at Superfund hazardous waste sites? A review of the papers presented at this colloquium would suggest that it has not. In addition, many non-NPL sites throughout the nation also require the setting of cleanup levels, and EPA protocol as outlined in Section 121 of SARA may not always be followed or be applicable to all sites. Three major issues emerged from the colloquium regarding cleanup level setting. First, the point of compliance at which ARARs shouic! be applied must be resolved. Impacted parties (wa- ter utilities, environmental groups) generally support compliance at the edge of the waste management unit or site of release, while generators argue for a point of compliance at property boundaries or at the point of impact (e.g., a downgradient water well). Second, the appropriate level of risk and the acceptable target levels must be selected. As expected, impacted parties support very conservative risk management decisions, with explicit support for cleanup levels corresponding to at least the one-in-a-million incremental cancer risk level for known or suspected human and animal carcinogens. Private industry and other generators (e.g., DOD), on the other hand, stress a comparative risk approach and argue that 1 in 100,000 (~0-5) or 1 in 10,000 (10-4) is a more practical and cost-effective target level that still adequately protects human health and the environment. Finally, the colloquium participants raised the issue of the adequacy of our current data base for making both risk analyses
OVERVIEW 5 and risk management decisions. There is considerable uncertainty that has not yet been quantified in all of the scientific techniques required for a quantitative resolution of the cleanup level dilemma. Exposure assessment using current models of contaminant trans- port in affected media (primarily unsaturated or saturated soil) is constrained by the lack of data on the fate of contaminants. The effectiveness of many remedial technologies to achieve very Tow leveb of contamination in soils or ground water is poorly un- derstood. In addition, the toxicologic data base and the methods used to estimate chronic risks at low levels of human exposure to contaminants are highly uncertain. Despite these shortcomings the risk management process implied in the debate over cleanup levels will continue to rely on informed judgments based on the existing scientific and technological data base. The colloquium provided a useful overview and a critique of the strategies that might be used to proceed in this uncertain process. OVE1IVIEW OF PAPERS As Richard Dowd, the keynote speaker, clearly indicated, set- ting cleanup levels at hazardous waste sites reflects many of the same issues that have confronted regulators for the past 15 years. Decisions must be made in spite of severe limitations on the accu- racy of predictions regarding the impacts of these decisions. Regulators must choose safety factors to account for this level of uncertainty. In contrast to the regulatory issues of the 1970s, however, the focus of regulation has shifted to the control of toxic materials with unverifiable potential long-term chronic ejects. This fact places a greater burden on the science of quan- titative risk assessment, which relies in turn on the accuracy of predictions regarding the fate of recalcitrant or poorly degradable organic contaminants in the subsurface environment a highly complex environment, when compared to surface waters or the lower atmosphere. Drawing on his experience with the implementation of the Clean Air Act, Dowd stressed the importance of developing meth- ods for setting standards at hazardous waste sites that use the best available scientific knowledge, that can be verified by current means of measurement, and that use safety factors or logical con- servative assumptions. He urged the commitment to a rational
6 HAZARDOUS WASTE SITE MANAGEMENT standard-setting methodology, anchored in reality and good sci- ence. This is without question a desirable goal; yet the application of these principles in the current environment of intense public concern over real or perceived hazards will be difficult. Current Methods Haling S. Brown's paper, which reviewed the approaches used by five governmental agencies for setting cleanup levels, suggested that Dowd's concerns are being addressed in several instances. Brown provided a succinct and useful summary of methods pro- posed or actually used by EPA, the U.S. Army, the California Department of Health Services (DHS), the Washington State De- partment of Ecology (WDOE), and the New Jersey Department of Environmental Protection (N]DEP). With the exception of the U.S. Army method, each of these approaches is currently used (DHS, WDOE, N3DEP) or will be used (EPA) to establish tar- get cleanup levels at both NPL and non-NPL sites. Generally, these methods are rational; that is, they use procedures that are based on our current scientific understanding of contaminant fates and human exposures, they attempt to provide safety factors that explicitly address the areas of limited knowledge, and they are linked to regulatory requirements established under other federal environmental statutes. In essence, the methods are attempts to define a risk assess- ment procedure for selecting cleanup levels, with the common goal of protecting human health and the environment. Although the methods differ in their details, which are clearly described in Brown's paper, the unambiguous application of any of the methods is inhibited by a number of common problem areas: . are lacking; media-specific numerical criteria for many contaminants numerical criteria (e.g., standards, water quality criteria) developed for other federal statutes may not be applicable to hazardous waste sites; . the extrapolation of toxicologic data obtained from animal experiments is highly uncertain and not verifiable; . multiple contaminants are treated as having additive ef- fects rather than multiplicative ejects; and
OVERVIEW . . 7 the assumption of equilibrium partitioning between media Is very conservative and usually does not match reality. In addition to these shortcomings, it should be noted that none of the methods addresses the issue of quality control of the techniques (primarily solute transport models) used to estimate contaminant movement. Large uncertainties must be expected, de- pending on the type of mode! used, the complexities of the media being modeled, and the amount and quality of the data available. Furthermore, as discussed by the colloquium participants, the un- ambiguous definition of the contaminant concentration is lacking, given the three-dimensional nature of the subsurface environment and the expected concentration fluctuations with time. Despite the passage of SARA, continuing disagreements among the op- posing parties in these issues can be expected, along with a high potential for litigation. Although the writers of SARA had hoped to define an objec- tive and uniform method for setting ground water cleanup levels, as discussed in Linda Greer's paper, the overview of the decision process presented in the paper written by Edwin Barth, William Hanson, and Elizabeth Shaw of EPA suggests that a case-by-case approach will still be used in reaching a record of decision for the remedial action at each site. Whereas target cleanup levels must be selected from applicable or relevant requirements, the final cle- termination of the cleanup levels rests on the "appropriateness" of the requirement, taking into account such imprecise concepts as fund balancing, technical feasibility, and cost-effectiveness. Barth and coauthors stress the importance of evaluating alternatives that meet a range of ground water protection goals, within various time periods. In addition, because of uncertainties in the efficacy of the technologies used for ground water cleanup, the authors recommend flexibility in the decision process to permit alternative remediation strategies, should monitoring indicate that the se- lected remedy will not achieve the selected goals. This "reopener" recommendation, however, although logical on technical grounds, could further complicate and delay the decisionmaking process for the selection of a remedy. The method developed by the DHS, as described by David Len and Paul HadIey, encompasses many elements of the proposed EPA method, but it takes a more conservative position regard- ing the range of alternatives that should be evaluated during the
8 HAZARDOUS WASTE SITE MANAGEMENT decision process. Any proposed remedial alternative must, at a minimum, meet the specified target goal, designated as an applied action level (AARON), for each contaminant or mixture of contami- nants in each medium of concern. Although no mention is made of technical feasibility or cost-effectiveness, the DHS decision tree method ultimately rests on EPA's feasibility study process as de- scribed in the agency's guidance documents. There appears to be no relief from the difficult value judgments required by the risk management process. Views of Impacted Parties Based on other provisions of SARA that stress the involvement of citizens and community groups in the decision process, the success or failure of any approach to specify acceptable target cleanup levels will depend heavily on the opinions of impacted parties. The remarks by Ronald Esau, representing an impacted water utility with 60 percent of its water supply source at risk from accidental industrial releases of organic contaminants, and Linda Greer, from the Environmental Defense Fund, an influential national environmental organization, indicate that support for cleanup levels less stringent than background or a low risk level of 10-6 appears unlikely from these groups. Such a posture can also be expected from other water utilities. Their real fear is that the distribution to consumers of contami- nated ground water that meets all ARARs might lead to toxic tort cases, with large costs to the consumers rather than to the parties responsible for the contamination. Linda Greer presented the environmental community's posi- tion on the selection of ARARs for ground water cleanup (using existing detection limits) and on the definition of the point of com- pliance. Although this position is a desirable goal for impacted parties, according to many colloquium participants, technical fea- sibility and fund balancing issues make these goals unattainable. One party directly impacted by this decision process is in- dustry. As Thomas HelIman pointed out, remediation costs could escalate dramatically if overly conservative standards are imposed at hazardous waste sites. The cost for remediation of 1,800 poten- tial NPL sites could increase to $81 billion, compared to the $8.5 billion cost currently estimated by EPA. HelIman indicated that the number of sites to be remediated
OVERVIEW 9 is likely to decrease if costs escalate, and excessive delays are also to be expected. In addition, he raised the specter of unforeseen but clearly negative consequences for the economic viability of industry. Whether or not this is a substantive issue would require more extensive analysis. The Status of the Tethnical Formation Base Risk management decisions associated with setting cleanup levels at hazardous waste sites rest on three broad scientific and technical disciplines: hydrogeology, risk assessment, and remedial engineering. The adequacy of our current knowledge in these areas was the focus of the final two papers in the colloquium and the subsequent workshops, although the time available was insufficient to address thoroughly the complex scientific and technical issues involved in evaluating the adequacy of existing methods for setting cleanup levels. In each broad area, models are available to assess the fate (hydrogeology, engineering) and potential impact (risk assessment) of many hazardous chemicals on humans and the environment, but there are widely divergent levels of certainty in the projected results. Although the three broad areas overlap divergent scientific dis- ciplines, the deficiencies identified in the papers by James Davidson and P. S. C. Rao, by Robert Tardiff, and in the work groups, were remarkably similar. Among these deficiencies or shortcomings are the following: ~ models (solute transport, risk extrapolation, performance of remedial technologies) do not account for all of the processes affecting the fate and impact of the contaminants; the models lack accuracy when confronted with a high degree of heterogeneity (complex hydrogeology, multiple contami- nants, two-phase flow, variable susceptibility in populations); . data requirements to ensure high levels of confidence in the accuracy of predicted results are prohibitively expensive and sometimes, in the case of risk assessment, impossible to obtain (we cannot test a million mice, much less test the chronic impacts of contaminants on humans); and all analytical methodologies suffer from a lack of knowI- edge on the fundamental processes underlying observed phenom- ena (biotic and abiotic fate of organic contaminants, biology of carcinogenesis, contaminant adsorption on soil).
10 HAZARDOUS WASTE SITE MANAGEMENT These deficiencies produce levels of uncertainty that have not been well defined to date in the remediation process. Despite these uncertainties, however, all participants clearly preferred the use of the existing technical information base in the decision process rather than a reliance on arbitrary or poorly documented decisions. SUMMARY The process of setting cleanup levels at hazardous waste sites poses new challenges to the regulators, the regulatory community, and the impacted parties, when compared to the implementation of other environmental statutes. Although all participants in the debate over the details of the risk assessment/risk management process desire an objective, tractable method for setting cleanup levels, numerous obstacles make this goal difficult to achieve. All of the participants in the colloquium were in agreement on the importance of using the best available scientific and technical knowledge in the decision process. Although the techniques em- ployed in hydrogeology, risk assessment, and remedial engineering are deficient in many aspects, they provide essential bases for in- formed decisionmaking in the face of great uncertainty. Indeed, the approaches proposed by EPA and other regulatory agencies can- not avoid the difficult value judgments that are implicitly required when large areas of uncertainty exist in the technical information base. It was apparent from the discussions held during the collo- quium that there has been progress toward a unified approach to setting cleanup levels at hazardous waste sites. Further advances in our understanding of the basic mechanisms of contaminant fate and the impact of contaminants on humans and the environment are urgently needed, however. Technologists must continue to strive to educate the public about the potential hazards at sites and the risks associated with alternative remediation strategies. Perhaps the continued debate on the Chow clean is clean?" issue will lead to a successful balancing of the conflicting demands of affected parties and result in environmental protection at a cost commensurate with the reduction in actual risks. It is the hope of all of the colloquium participants that this document will be a useful benchmark in that debate. MICHAEL KAVANAUGH, Chairman
lssne F~ers and Pr~ocateUr~ Comments