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.
2 Establishing and Meeting Ground Water Protection Goals in the Superfund Program EDWIN F. BARTH III, WILLIAM HANSON, AND ELIZABETH A. SHAW Decisions on contaminated ground water at uncontrolled haz- ardous waste sites are complicated because of complex fate and transport patterns. The process being developed will guide reme- dial project managers and other decisionmakers concerned with ground water remedial actions at Superfund sites so that a consis- tent ground water evaluation and decision approach is applied to all such sites. APPLICABLE OR RELEVANT Al APP1tOPR~TE REQUIREMENTS Under the National Contingency Plan (NCP) (Federal Regis- ter, 1985), remedial actions at Superfund sites shall meet or exceed all applicable or relevant and appropriate federal requirements and consider other pertinent federal criteria, advisories, and guidances and state standards. Federal requirements that may be applicably, relevant, or appropriate to Superfund ground water actions are in- cluded in the Resource Conservation and Recovery Act (RCRA) Subpart F regulations. Determinations of ground water protection This paper was first presented at the 7th National Conference on Management of Uncontrolled Hazardous Waste Sites, Washington, D.C., December 1-3, 1986. It was prepared prior to the Superfund Amendments and Reauthorization Act. The concept presented in this paper may be modified following the promulgation of the revised National Contingency Plan. 22
GROUND WATER PROTECTION GOALS 23 levels under both RCRA (as alternate concentration levels) and ~me,, ~, Superfund may be based on a site-specific risk assessment. The Safe Drinking Water Act and the Clean Water Act resulted in the development of maximum concentration levels (MCEs), maximum concentration level goals (MC[Gs), health advisories, and water quality criteria for the protection of pub- lic health, all of which are evaluated for ground water protection levels in the Superfund program. EPA's ground water protection strategy (U.S. EPA, Office of Ground-Water Protection, 1984) is an important component of Superfund's ground water approach. The strategy says that ground water should be protected differ- entially based on characteristics of vulnerability, use, and value. Special ground water (CIass I) is highly vulnerable to contami- nation because of the hydrological characteristics of the areas in which it occurs. It is characterized by either of the following: the ground water is irreplaceable, in that no reasonable alternative source of drinking water is available to substantial populations; or the ground water is ecologically vital, providing the base flow for a particularly sensitive ecological system that, if polluted, would destroy a unique habitat. Current-use ground water (Class ITA) and potential-use ground water (Class IlB) that are sources of drinking water (or have other beneficial uses) include all non-CIass ~ ground water that is currently used or is potentially available for drinking water or other beneficial use. Ground water not considered to be a potential source of drink- ing water and of limited beneficial use (Class IlI) is nonusable ground water that is highly saline that is, with total dissolved solids (TDS) levels over 10,000 milligrams per liter (high) or that is otherwise contaminated beyond levels that allow cleanup using methods reasonably employed in public water treatment systems. This condition must not be the result of a single waste site but rather the result of a wide range of sources. Class Ill is further separated by the degree of interconnection with adjacent water. Class IlIA is highly to moderately interconnected and is thus most relevant to Superfund. Class ITIB ground water has a Tow de- gree of interconnection and typically occurs at greater depths. As will be explained in this paper, the Superfund program will use these ground water characteristics in the evaluation of alternative response actions.
24 HAZARDOUS WASTE SITE MANAGEA1:13NT DEVELOPMENT OF GROW WATER ALTERNATES In general, source control measures should facilitate the achievement of long-term remediation objectives and goals for ground water. EPA's guidance document for feasibility studies under the Comprehensive Environmental Response, Compensa- tion, and Liability Act (CERCLA) (U.S. EPA, 1985a) calls for the development, screening, and detailed evaluation of alternatives proposed for remedial actions. For ground water contamination problems, this process involves the development of a limited num- ber of remediation alternatives to be presented to the decision- maker. The performance goal of each ground water alternative should be expressed in terms of a cleanup concentration (in the ground water) and a time period for the restoration for all locations in the area of attainment. Performance goals in terms of ground wa- ter concentrations may be available as MCEs, proposed MCEs, or more stringent state standards. If these are not available, concen- trations may be derived from health-based criteria such as excess unit carcinogenic risk (UCR) or referenced dose values. Other potentially approved standards include health advisories or water quality criteria or both. Health-based criteria may also be de- veloped if no standards, advisories, or criteria are available. (The reader is referred to the Superfund Public Health Evaluation Man- ual [U.S. EPA, 1985b] for information on developing health-based criteria.) Restoration time periods may range from very rapid (1 to 5 years) to relatively extended (perhaps several decades). If ground water with the characteristics of Class ~ or Class I] is contaminated with known or suspected carcinogens, the program suggests the development of a limited number of ground water protection goals be developed that vary between 10-4 UCR and 10-7 UCR and vary between restoration time periods. A point- of-departure alternative for initial decision evaluation should be developed at a 10-6 UCR with a limited restoration time period. For noncarcinogens, alternatives should be developed that meet chronic or acute threshold levels in varying restoration periods. In situations in which the plume is not in close proximity to a receiving body of water, plume containment measures (e.g., gradient control) should also be evaluated, which will eventually result in a 10-4 UCR and 10-7 UCR for carcinogen levels in the ground water. Other alternatives (a limited number, possibly
GROUND WATER PROTECTION GOALS 10-3 Oh In O ¢~ 10-5 Z ~ O. I_ ° ,$ _ - C] LU LL C: 10-4 10-6 10-7 25 + _ Point of Departure _ + + Minimum Alternatives Recommended for Detailed Evaluation \ Natural Attenuatlon/Containment \ \ - o YEARS Time to Achieve Cleanup Target (no scale) Note: Noncarcinogens will have a threshold level that is not variable. FIGURE 2-1 Suggested alternatives to be developed for ground water contaminated with carcinogens. two or three) should also be developed around the point of de- parture. (Figure 2-1 presents a conceptual risk/restoration time plot of these suggested alternatives for carcinogens contaminating ground water with the characteristics of Class ~ or Class IT.) The alternatives will then be evaluated to compare the trade-offs be- tween the cleanup level, the time it takes to achieve the level, and the cost of the action. DECISION ANALYSIS The decision as to which remedial action alternative to select and implement depends on many factors. Those factors relating to the concentration level for carcinogens in the ground water include other health risks borne by the affected population and population sensitivities. For example, at the Reilly Tar Superfund site (U.S. EPA, 1984), the population had been exposed to contaminated ground
26 HAZARDOUS WASTE SITE MANAGEMENT water for an indeterminable period of time, which influenced the decision to use a "more protective" lower concentration level. Simi- larly, a more protective lower concentration level may be evaluated if the exposed population is unusually sensitive to the contami- nants. Acute and chronic levels for noncarcinogens are threshold values and therefore are not influenced by these two factors. Factors that influence the restoration time period for ground water contaminated with carcinogens and noncarcinogens are as follows: feasibility of providing an alternative water supply; current use of ground water; potential need for ground water; electiveness and reliability of institutional controls; and ability to monitor and control the movement of contami- nants in ground water. The existence of other drinking water sources of sufficient quality and yield, sources that are readily available and that may be used as an alternative water supply, reduces the importance of rapid restoration of the contaminated ground water. On the other hand, where a demand for drinking water from ground water is likely in the future and other potential sources are not sufficient, those remedies that achieve more rapid restoration should be fa- vored. The effectiveness and reliability of institutional controls to prevent the use of contaminated ground water for drinking water purposes should also be evaluated. If these controls clearly are not effective, rapid restoration may be necessary. In some circumstances, complex flow patterns increase the potential for unanticipated migration pathways and may reduce the effectiveness of remedial action. In these situations, remedial actions that will rapidly restore ground water, such as extensive source control and high-rate pumping, should be emphasized. Other factors that should be considered in determining the appropriate ground water protection goals for carcinogens and noncarcinogens include limiting the extent of the contamination, its impact on environmental receptors, the technical practicability of implementing the alternative, and the alternative's cost. Limited increases in concentration may be evaluated if the expanded area of ground water contamination is relatively small,
GROUND WATER PROTECTION GOALS 27 the period of degradation is short, and the ultimate discharge of the plume has no significant effect on surface waters. The technical practicability of each alternative must also be evaluated in light of the contaminant characteristics and applica- ble hydrogeological conditions, which may not allow the effective implementation of the alternative to clean up the ground wa- ter. Environmental receptors should be taken into account when ~v~l~.t.in~r t.hP ~.nnronri~.t.~ ~.l~an,~n ~on~.entration levels and time period. o A- -or- -rid Finally, under the NCP, response actions must be cost-e~ec- tive. Therefore, a careful evaluation of capital outlays and the operation and maintenance costs associated with each alternative must be considered and compared to those of each of the other alternatives. Ground water remediation time frames may be ex- tended if the agency decides that the costs to meet performance goals in 1 to 5 years are extraordinarily high and as long as insti- tutional controls will be effective for the additional period. Figure 2-2 presents general ground water goal areas associated with the ground water characteristics on the risk/restoration plot for carcinogens. The clecisionmaker should first evaluate the point- of-departure remedy and then move to other general areas on the plot as influenced by the ground water's characteristics. The reader should be cautioned that the general areas delineated on the plot are not rigid. FIEXI ISLE DECISION PROCESS Complex fate and transport mechanisms of contaminated ground water often make it difficult to predict accurately the performance of ground water remedial action. Therefore, the re- medial process must be flexible, allowing changes in the remedy based on the performance of several years of operation. To illustrate this principle, Figure 2-3 presents three possible situations that may occur after several years of a ground wa- ter response action. In the first scenario (Case 3A), the target concentration will be reached within the desired time period. In the second scenario (Case 3B), the target concentration will be reached somewhat later than the desired time period. In the final scenario (Case 3C), the target concentration will not be reached in a foreseeable time period. A performance feedback concept has been incorporated into
28 10-3 J 1 0~4 ^ TIC ~ 0 5 Z ci O _ _ UJ G 10~6 10-7 _ HAZARDOUS WASTE SITE MANAGEMENT Mlnimum AKomaUvos Rocommen~bd for Detailed Evaluation 1 \ General Area for Polnt of Departure CLASS 1 I Generals I Arm l for I 1 CLASS 1 | IIA water l Ground | Ground I CharacterlsUcs CharacterlsUCS | Characterl~UCs l ~I ~ \ \ ~ ~ \ \ \ \ \ \ General Area for CLASS IIB \ Ground \ \ i, - Natural Anonua~don/Contalnmont o YEARS Time to Achieve Cleanup Target (no scale) Notes: Ground water with charactenst~s of Class I aquifers is expected to be restored most rapidly. Ground water with Class IIB characteristics may be restored more slowly because of the reduced potential for imrned~te exposure. In general the remedial alternatives developed are "bounded" by the point of departure alternative and the natural anenuat~n/cont~nment rernedm at the 10 and 10 risk levee. FIGURE 2-2 Performance range for ground water remedial alternatives. the decision process so that in situations in which the performance goal will not be met (e.g., in Case 3B and Case 3C) the decisions may be reevaluated based on actual experience. If the remedial action is not meeting expectations, the decisionmaker should de- cide the extent to which further or different action is necessary and appropriate to protect human health and the environment. Figure 2-4 illustrates this evaluation process. Should it be de- termined that it is not practicable to restore the ground water to the initial cleanup goal level, an exception to the NCP could be demonstrated, based on extraordinary costs or the technical impracticability of meeting applicable or relevant and appropriate federal requirements.
GROUND WATER PROTECTION GOALS REMAINING GROUND WATER CONTAMI NANT CONCENTRATION REMAINING G ROUND WATE R CONTAMI NANT CONCENTRATION REMAINING G ROUND WATE R CONTAM I NANT CONCENTRATION LEGEND Remedial Action Performance Goal Time of Performance Evatua~,on 29 Actu al Performance . ~ L Pred icted Performance __ i ~ DURATION Of REMEDIAL ACTION Case 3A Ground Water Goal will be achieved _ ~ DURATION OF REMEDIAL ACTION Case 3B Ground Water Goal will be achieved in longer time frame DURATION OF REMEDIAL ACTION Case 3C Ground Water Goal will not be achieved over long period of time FIGURE 2-3 Possible restoration scenarios when evaluating performance data.
30 REMEDY A COMPLETE / Have \ YES ~ Objectives ma\ Been Met HAZARD O US WASTE SITE MANAGEMENT [ Select Remedy . _ ~ Implement Remedy/ Monitor Performance > ~YES _ ~ _ ~ NO ~3 / Revise targets \ YES NO By · | Continue Operation/Mon~toring ~I ~ NO _ I Unorade Remedv | ~ _ , I FIGURE 2-4 Flexible decision process for ground water remedial actions. REFERENCES FederalRegs~ter. 1985. November 20. U.S. Environmental Protection Agency (EPA). 1984. Record of Decision- Reilly Tar Site. Washington, D.C. U.S. EPA. 1985a. Guidance on Remedial Investigations Under CERCLA. Prepared for Hazard Waste Engineering Research Laboratory, Office of Emergency and Remedial Response, and Office of Waste Programs Enforcement. Washington, D.C., and Cincinnati, Ohio. June. U.S. EPA. 1985b. Superfund Public Health Evaluation Manual. Draft Office of Emergency and Remedial Response. OERR. Washington D.C. December. U.S. EPA, Office of Ground-Water Protection. 1984. Ground-Water Protec- tion Strategy. Washington, D.C. August.
GRO UND WATER PRO TECTION GOALS 31 PROVOCATEUR'S COMMENTS Joe} H~rerhborn The preceding paper does not give me a lot of opportunity to be critical in a sense because it is a general framework, with which ~ find myself in agreement. It is a technically rational framework. One point ~ find myself in particular agreement with, which was not stressed in the presentation but is in the paper, is the use of classification systems, particularly for aquifers. As Superfund grows (we are talking about thousands of sites), it becomes necessary to move away, in my opinion, from a logic that says that every site is unique. Although every site may be unique, just in the same way that every person is unique, that does not mean that you cannot use classification systems to help manage a very complex and large number of sites. So ~ applaud the use of a classification system in this case, for aquifers. There are a couple of issues that have not been fleshed out in the presentation. First, it is going to be increasingly necessary to deal with multiple exposures by the government or whatever au- thority is dealing with cleanups and performing risk assessments. People have to pay more attention to the exposures that citizens are getting from various sources. In other words, if you do a risk assessment and you say, here is the exposure from a particular contaminated water supply, you cannot neglect the fact that the same population is being exposed to the same, similar, or different chemicals from other roots of exposure, including other cleanups. We have seen situations in which half a dozen cleanups are going on, stuff is going into a river that is becoming the drinking water supply downstream, and none of this has been factored into the risk assessment. ~ feel that this is a fundamental fallacy and limi- tation of what we see going on in risk assessment. A lot of levels of exposure that might be deemed acceptable on a case-by-case basis are certainly not acceptable in a cumulative sense. A very interesting point in this paper is the framework, which is something, again, ~ agree with. We have talked about it for years, and it does not get much attention in the "How clean is clean?" issue. It is a completely different logic that is predicated on the idea that the starting point should be the issue of the future
32 HAZARDOUS WASTE SITE MANAGEMENT use of a natural resource. What will be the future use of a piece of land or a body of water? It is the future use in particular that will determine exposures, and from there you can deal with risk. If we talk about the future use of a natural resource as the primary factor, then there are a lot of policy implications because in the United States the use of water and land is fundamentally a local and state decision, not a federal one. ~ think we have created a monster with the Superfund program that is, a great deal of federal authority (because the federal government provides a lot of the money); yet if we apply this logic of dealing with the future use of natural resources, you would have to shift decisionmaking to a local level. Now, ~ personally think that is desirable; shift the burden to the people who have to live with that resource to decide how clean is clean and to deal with the institutional problems of ensuring future use. We have a long history of dealing with deeds, deed restrictions, and things like that. Historically, Love Canal was an example not of a failure of industry or technology but a failure of institutional control on the future use of land. If you look at chemical codes, the authorities, the limitations on the use of that land, you will find that it was government, local government authorities, who in fact simply forgot about what was told to them and went ahead and used the natural resource in an inappropriate way-and that was the real cause, historically, of Love Canal. ~ think we have to deal more with this issue of the future use of natural resource and to think of it as a departure point in the assessments of cleanup goals. It happens to be what is done in the United Kingdom, and ~ find that their cleanup program is much more cost-effective and sensible than ours. Superfund, in my view, is probably going to be an incredible waste of money because we have a great deal of underreaction and a great deal of overreaction. It is a system that is not optimized in any engineering sense. One other fact that is buried in the framework discussed in the paper is the real problem: it sounds fine, in terms of generalities, to talk about cleanup goals driving what will happen, but in fact, what really happens in the system is that technology and money dictate how much cleanup is done. ~ do not think we ought to ignore the fact that there are opportunities in this framework presented by EPA to have people decide on the level of cleanup simply on the basis of what available technology can do or what bureaucrats have decided is an appropriate amount of money for a site. My interpretation of how Superfund really
GROUND WATER PROTECTION GOALS 33 works is that somebody decides this site is worth about $10 million; now, go out and tell me how to spend $10 minion, and that is the amount of cleanup you get. ~ am not so sure that this is going to change in the more eloquent framework presented here. Finally, my last point is that any framework in any technical methodology ought to be sensitive to implementation issues. Just as giving a loaded gun to a child does not make any sense, giv- ing risk assessment to people without adequate information and adequately trained people to use it is also folly. Creating a Su- perfund program at an $8 billion spending level without adequate information, adequate technology, and adequately trained people is another folly, analogous to giving a loaded gun to a child. ~ wish we would talk more about how we can implement these frame- works, how we can implement the use of risk assessment; and if we cannot implement it now or in 5 to 10 years, are you willing to talk about making a commitment to delay action until we get the information or the technology or the trained people? If you want to wait, then what is the interim strategy and what are the interim framework and methodologies to be used? For the most part, what see are long-term methodologies and a lack of implementation ability now.