1
The Ground Water Cleanup Controversy
At hazardous waste sites nationwide, industries and government agencies are spending millions of dollars trying to clean up contaminated ground water. These cleanups are required by federal and state laws passed in the last two decades—mostly in response to public concern that drinking contaminated ground water may cause cancer or other illnesses. The laws require that, in most instances, the contaminated ground water be restored to a condition that meets state and federal drinking water standards.
Recently, some have begun to question current approaches to ground water cleanup. Evidence suggests that restoring contaminated ground water to drinking water standards poses considerable technical challenges that may sometimes be insurmountable. For example, at one New Jersey site, a computer manufacturing company spent $10 million removing toxic solvents from ground water, but not long after the cleanup system was shut down the solvent concentrations in some locations returned to levels higher than before cleanup began (see Box 1-1). This company's effort and others like it have raised concern about whether the amount spent to clean up ground water is proportionate to the benefits society receives. Businesses and government agencies paying for the cleanups are calling for reconsideration of whether returning all contaminated ground water to drinking water standards is a realistic goal. At the same time, public interest groups are advocating maximum protection of the public's right to a safe water supply, both in places where the ground
BOX 1-1 GROUND WATER CLEANUP IN SOUTH BRUNSWICK TOWNSHIP, NEW JERSEY: SYMBOL OF A BROADER PROBLEM In 1977, toxic solvents were discovered in one of the three main wells supplying drinking water to South Brunswick Township in New Jersey. The attempt to remove this contamination has become a symbol of the broader problems with ground water cleanup nationwide. Government investigators traced the contamination to a nearby computer manufacturing facility. The facility owner agreed to clean up the site and installed a series of pumps to extract the polluted water and eliminate the contaminants. Over the next six years, the company spent $10 million pumping out water and treating it. At the end of this period, the water in the well appeared to meet drinking water standards. However, within three years of when the company shut down the treatment system, contaminant levels in some areas near the site rose above the drinking water standards again. In one location, the contaminant concentration was twice as high as before cleanup began. Technical experts called to the site traced the return of contamination to solvents that had migrated underground and lodged in subsurface geologic formations. These solvents were dissolving slowly into the clean ground water flowing around them. The initial pumping had removed most of the dissolved solvents but had not removed the undissolved solvents, which were recontaminating the clean ground water. As a result, the site owners had to resume pumping to prevent the contamination from spreading. In addition, they installed a million-dollar treatment system at the wellhead to provide clean drinking water. This case is often cited as an example of the failure of conventional ground water cleanup technologies. However, as will be explained in Chapter 3, scientists have now recognized that a primary reason for the return of contamination at this site was the failure to install a containment system around the undissolved solvents. While the water within the boundaries of a containment system would not have met potable standards, the containment system would have isolated the solvents and prevented recontamination of most of the ground water. (For technical details about this site, see Box 3-3 in Chapter 3.) REFERENCES: Stipp, 1991; EPA. 1989. |
water is currently used for drinking and in places where it might be used in the future.
This report provides a comprehensive evaluation of the technical and policy dilemmas surrounding current ground water cleanup efforts. It assesses whether conventional and innovative cleanup technologies are capable of restoring contaminated ground water to drinking water quality. It reviews physical and chemical factors that impede cleanup regardless of the technology chosen. It discusses factors other than technical feasibility—human health, ecology, and costs—that are critical components in the ground water cleanup debate. And it provides advice on
how to set policies for ground water cleanup that reflect the capabilities of technology while still protecting human health and the environment.
The report was prepared by the Committee on Ground Water Cleanup Alternatives, appointed by the National Research Council. The National Research Council appointed the committee to prepare this report because of widespread concern in the scientific community that the technical complexities of ground water cleanup, which are becoming increasingly apparent, may call for changes in ground water cleanup policies. The committee consisted of 19 experts in ground water cleanup technology, policy, and law representing a balance of viewpoints; members came from industry, government, environmental groups, academia, and con-suiting firms. The committee met nine times over a two-year period to review technical information and deliberate policy issues. The committee called upon the wider technical community to provide data related to the capabilities of ground water cleanup technologies. In addition, the committee invited people with a stake in ground water cleanup—citizens whose lives have been affected by contamination and industries that have invested large sums in cleanup—to present their viewpoints at committee meetings.
HISTORY OF GROUND WATER CLEANUP
Ground water contamination is relatively new on the nation's list of recognized environmental problems. Early environmental legislation focused on the more obvious problems: air and surface water pollution. For example, in Pittsburgh and St. Louis, air pollution was once so severe that drivers had to use headlights in the middle of the day during the winter. Along the Cuyahoga River near Cleveland, pollution was so extreme that the river caught fire—once in 1936, twice in the 1950s, and again in 1969. It was easy for the public to recognize the need to clean up surface water and air, and Congress enacted legislation to protect these resources as early as the 1940s and 1950s.1 Ground water, however, was long believed to be naturally protected by the layers between the earth's surface and the water table, which people believed would filter out contaminants. The problem of ground water contamination did not receive widespread public recognition until the 1970s, when contamination episodes began receiving notice in the popular press. In the most publicized of these incidents—known as Love Canal—President Carter declared an emergency in Niagara Falls, New York, because of health concerns linked to ground water contamination, and many homes were evacuated.
Legislation
In 1980, prompted by the Love Canal incident, Congress for the first time made ground water cleanup a high national priority with the passage of the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA), commonly known as the Superfund act.2 CERCLA established a $1.6 billion federal fund (which has since grown to $15 billion), the Superfund, to pay for cleaning up abandoned hazardous waste sites (EPA, 1990; Guerrero, 1991). CERCLA also provided authority for the Environmental Protection Agency (EPA) to sue parties responsible for the contamination to recover cleanup costs; these groups have since become known as "potentially responsible parties."
In 1984, Congress broadened the nation's ground water cleanup program by amending the Resource Conservation and Recovery Act (RCRA) to require cleanup of contamination at active facilities that treat, store, or dispose of hazardous waste. To continue handling wastes, operators of active RCRA sites must agree to clean up existing pollution. RCRA also covers cleanup of contamination from leaking underground storage tanks containing petroleum products and other organic liquids.
Since the passage of CERCLA and the 1984 RCRA amendments, virtually all states have enacted laws granting them authority to require cleanup of sites with contaminated ground water (EPA, 1994). CERCLA and RCRA have strongly influenced the state laws, although some state laws are more stringent than the federal versions.
Early Research
Despite the limited public awareness of ground water contamination prior to 1980, some scientists had begun studying the problem several generations before CERCLA's passage. The earliest ground water contaminant recognized by scientists was human sewage (for a historical perspective, see Mallman and Mack, 1961). In 1854, a London doctor linked a cholera epidemic to contamination of drinking water supplies—including a neighborhood water well—with sewage. In Switzerland in 1872, a typhoid epidemic was traced to sewage contamination in a river that recharged a town's ground water supply. In 1909, two German researchers ran a series of controlled tests to investigate bacterial migration underground and established that bacteria could travel with ground water from one well to another.
As chemical use increased after World War II, isolated reports of chemical contamination of ground water appeared. In 1947, for example, hexavalent chromium from electroplating wastes was discovered in a Michigan ground water supply after homeowners complained that their
water had turned yellow (Deutsch, 1961). Relatively common after the war were complaints of foaming ground water—from contamination with the surfactant alkyl benzene sulfonate that had leaked from septic systems. Recognizing the increasing potential for chemical contamination of ground water, the American Water Works Association created a task force of scientists, the Task Group on Underground Waste Disposal and Control, to study the problem in the early 1950s. The task group's first report, issued in 1952, documented that very few states were aware of the potential for ground water contamination (Middleton and Walton, 1961).
Though scientists recognized the ground water contamination problem long before the general public and government agencies, until recently only a few researchers were studying ground water cleanup technologies. At a 1961 conference on ground water contamination sponsored by the U.S. Public Health Service, one speaker remarked that because of the lack of trained ground water scientists, "there are few fields in which our ignorance is so profound" (McKee, 1961). There has been a dramatic
increase in the number of ground water scientists since CERCLA's passage in 1980. Nevertheless, many of the same problems that researchers identified at the 1961 conference as key to understanding ground water contamination remain to be answered today—questions such as how geology, soil type, contaminant chemistry, and microbial activity affect the spread of contamination (McKee, 1961; McCarty, 1990).
Sources of Contamination
The primary type of ground water contamination of concern in the United States today is contamination from hazardous chemicals. The use of such chemicals is ubiquitous: substances found in contaminated ground water are used in everything from lumber treating to electronics manufacturing, fuels, food production, and agricultural chemical synthesis. When used, stored, or disposed of on land, these chemicals may eventually migrate to the ground water below.
Common causes of ground water contamination are accidental spills; intentional dumping; and leaks in storage tanks, industrial waste pits, and municipal or industrial landfills. In addition, significant quantities of contaminants may be released through routine activities such as washing of engines and rinsing of tanks. Standard application of agricultural chemicals is also a source of ground water contamination. For example, the EPA estimates that about 1 percent of all drinking water wells in the United States exceed a health-based limit for pesticides (EPA, 1992b). Although pesticide application is a potentially important source of contamination, this report focuses on the point sources of contamination found at hazardous waste sites and other sites where hazardous chemicals have leaked or spilled into the environment. Because point sources affect only a limited area, they present a more manageable problem than contamination of large areas of land with agricultural chemicals, which might far exceed the limits of cleanup technologies. Table 1-1 ranks chemicals found at hazardous waste sites in order of prevalence and gives common sources for these chemicals.
MAGNITUDE OF THE PROBLEM
Because of the widespread use and disposal of hazardous chemicals on land, the ground water contamination problem is potentially very large. However, estimates of the total number of contaminated sites have varied. In general, existing estimates have included seven categories of sites:
TABLE 1-1 The 25 Most Frequently Detected Ground Water Contaminants at Hazardous Waste Sites
-
closed or abandoned hazardous waste sites requiring cleanup under CERCLA;
-
active hazardous waste treatment, storage, and disposal facilities requiring cleanup under RCRA;
-
facilities with leaking underground storage tanks (used for storing gasoline and other fuels, as well as various chemicals used in manufacturing);
-
sites managed by the Department of Energy (DOE) (contaminated with the byproducts of nuclear weapons production);
-
sites managed by the Department of Defense (DOD) (contaminated with a variety of substances, including remnants of conventional weapons manufacturing and fuels used in the nation's defense fleet);
-
federal facilities other than those managed by DOD and DOE (such as abandoned mining sites owned by the Forest Service, grain storage facilities operated by the Commodity Credit Corporation, and research laboratories managed by a variety of federal agencies); and
-
sites managed under state laws similar to CERCLA and RCRA.
Table 1-2 shows estimates of the number of sites in each of these categories as compiled from three different sources. As the table shows, the total number of sites where ground water may be contaminated is likely to be in the range of 300,000 to 400,000. However, it is extremely impor
TABLE 1-2 Number of Hazardous Waste Sites Where Ground Water May Be Contaminated
tant to recognize that the magnitude of the contamination problem varies widely at these sites. Ground water contamination from a single leaking underground storage tank at a gas station affects a relatively small area and, as discussed in Chapter 3, is relatively easy to clean up. On the other hand, contamination at CERCLA sites and at major DOE installations may be widespread and very difficult to clean up. The differences between these types of sites are illustrated by the costs of cleaning them up. According to recent EPA data, the average cost of cleaning up a leaking underground storage tank is $100,000,3 while the average cost of cleaning up a Superfund site is $27 million (EPA, 1993). By far the bulk of the sites listed in Table 1-2 are contaminated from leaking underground storage tanks. The larger sites posing the greatest hazard to public health and the environment represent a relatively small portion of the total potential number of sites.
In part because of the wide variation in contaminated sites and because the total number of sites is uncertain, estimating the total national costs of cleaning up contaminated ground water is extremely difficult. One recent, widely publicized report concluded that over the next 30 years, the nation as a whole will spend $480 billion to $1 trillion, with a ''best guess'' of $750 billion, cleaning up the types of sites listed in Table 1-2 (Russell et al., 1991). With 90 million households in the nation (Industrial Economics, Inc., 1991), this represents a cost of $8,000 per household. Another recent report concluded that by the year 2000, the nation will be spending nearly $24 billion per year complying with requirements for hazardous waste and underground storage tank cleanup under RCRA and site cleanups under CERCLA (Carlin et al., 1992, p. 38). Some contest the accuracy of such cost estimates because of the high level of uncer-
tainty associated with the magnitude of the contamination problem and the large number of assumptions underlying the estimates. Nevertheless, the potential enormity of the costs has fueled the debate about whether the benefits the nation will receive from ground water cleanup at hazardous waste sites justify the costs.
CAPABILITIES OF CLEANUP TECHNOLOGIES
The rate at which new contaminated sites have been discovered has far exceeded the evolution of cleanup technologies (McCarty, 1990). Almost all ground water cleanup systems currently installed and planned involve variations of a technology called "pump and treat." Pump-and-treat systems operate by pumping ground water to the surface, removing the contaminants, and then either recharging the treated water into the ground or discharging it to a surface water body or municipal sewage plant (see Box 1-2).
Once ground water has been pumped to the surface, contaminants can be removed to very low levels with established technologies used to treat drinking water and wastewater. However, pumping out the water does not guarantee that all of the contaminants have been removed from beneath the site. Contaminant removal is limited by the behavior of contaminants in the subsurface a function of contaminant characteristics, site geology, and extraction system design.
When CERCLA was passed in 1980, the nation had very little experience with ground water cleanup. The details of how hydrogeology and contaminant chemistry might affect pump-and-treat systems were being investigated by a few forward-looking scientists but were not widely known in the regulatory community. As a result, the ground water cleanup efforts of the 1980s were a series of large, relatively uncontrolled experiments in whether existing technology was capable of overcoming natural physical and chemical factors that retain contaminants in the subsurface.
The first widely recognized evaluation of how the early pump-and-treat systems performed was released by the EPA in 1989 (EPA, 1989). The EPA studied 19 sites, expanding the number to 24 in an updated study published in 1992 (EPA, 1992a). Both the original 1989 report and the 1992 update found that while pump-and-treat systems may remove significant amounts of contaminant mass and prevent contaminants from spreading, most systems have so far failed to reach cleanup goals. At many of the sites studied, the contaminant concentration decreased rapidly when the pumps were first turned on, but then it leveled off and progressed toward cleanup goals much more slowly than the designers originally predicted. A 1991 study by researchers from Oak Ridge Na-
BOX 1-2 THE MANY VARIETIES OF PUMP-AND-TREAT SYSTEMS The conventional pump-and-treat system, shown in Figure 1-1, consists of a series of wells that pump contaminated water to the surface. A surface treatment unit removes the contaminants. The treated water is disposed of through wells or sprinklers that recharge the water into the ground or through pipes linking the surface treatment unit to a lake, ocean, stream, or wastewater treatment plant. At some sites, engineers have experimented with enhancements to augment the performance of conventional pump-and-treat systems. One type of enhancement improves the efficiency of contaminant extraction by injecting surfactants, steam, or other substances underground. A second enhancement uses air to volatilize contaminants. A third enhancement involves injection of substances that help transform the contaminants in place—chemical compounds that oxidize the contaminants, for example, or oxygen that encourages microorganisms to degrade the contaminants. In addition to their use for ground water cleanup, pump-and-treat systems may be used to clean contaminated soil between the earth's surface and the water table (a region called the vadose zone). Vadose zone pump-and-treat systems involve flushing the soil with water (to dissolve contaminants or remove them by hydraulic force) or air (to volatilize the contaminants). This committee has included all these enhancements under the general term "pump and treat" because all involve the pumping of fluids (either water or air). More importantly, the committee wishes to emphasize that all these enhancements are subject to the same major limiting factors that affect conventional pump-and-treat systems. As discussed later in this report, no known enhancement can completely compensate for the complexities of geology and contaminant chemistry that slow cleanup efforts. |
tional Laboratory reached similar conclusions (Doty and Travis, 1991). Around the same time as the first EPA study, reports of the limitations of pump-and-treat systems began to receive wide notice in technical publications (see, for example, Mackay and Cherry, 1989; Travis and Doty, 1990).
THE POTENTIAL CONFLICT BETWEEN TECHNOLOGY AND POLICY
The results of the studies of the late 1980s and early 1990s led many people to question whether the risk reduction that pump-and-treat systems achieve is worth their cost. Some have interpreted the studies to mean that cleaning. up ground water to health-based levels is impossible. Others are more optimistic and contend that the data are inadequate to support such an extreme conclusion, given the relatively small number
of systems studied and the relatively short amount of time these systems have been operating. These people point out that the 24 systems evaluated in the 1992 EPA study had an average operating time of less than seven years.
That the ability of technology to restore contaminated ground water was in question was not known beyond a limited group of scientists when Congress enacted the major ground water cleanup laws. As a result, current ground water cleanup policies rest on the assumption that restoring contaminated ground water is technically straightforward. Regulations under CERCLA, RCRA, and their equivalents at the state level require the establishment of strict numerical concentration goals for cleanup. Under CERCLA and RCRA, cleanup goals are generally set at drinking water standards—known as "maximum contaminant levels." Under state programs, goals are sometimes stricter—such as concentrations equal to natural background levels in uncontaminated areas. In setting ground water cleanup goals, government regulators have only rarely considered whether existing technology is capable of meeting these goals.
As a result of the increasing publicity of frustrated cleanup efforts such as that described in Box 1-1, many people have criticized the existing approach to setting ground water cleanup goals. These critics contend that the setting of goals that may not be technically achievable establishes unrealistic public expectations and misuses financial resources. On the other hand, supporters of existing goals maintain that strict goals are necessary to induce the maximum level of cleanup possible and to provide the maximum level of public health protection. Supporters of existing goals also argue that these goals help encourage the development of improved technologies.
Thus, a potential for conflict has arisen between existing policies for ground water cleanup and the capabilities of existing technologies. On one hand is the desire to eliminate the health risks of ground water contamination, as mandated by law. On the other hand are reports that technology has so far been unable to reach health-based cleanup goals at many sites, despite substantial effort. Over the last few years, policy-makers in the EPA and other branches of government have recognized this potential conflict and have been reevaluating the current approach to ground water cleanup in light of new knowledge about technical limitations. This report provides information to guide these policymakers in assessing whether the nation's current ground water cleanup programs adequately reflect the capabilities of existing and emerging technologies.
NOTES
REFERENCES
Carlin, A., P. F. Scodari, and D. H. Garner. 1992. Environmental investments: the cost of cleaning up. Environment 34(2):12-20, 38-44.
Deutsch, M. 1961. Incidents of chromium contamination of ground water in Michigan. Pp. 98-104 in Ground Water Contamination: Proceedings of the 1961 Symposium, Cincinnati, Ohio, April 5-7. PB-214895. Springfield, Va.: National Technical Information Service.
Doty, C. B., and C. C. Travis. 1991. The Effectiveness of Groundwater Pumping as a Restoration Technology. Oak Ridge, Tenn.: Department of Energy, Oak Ridge National Laboratory, Health and Safety Research Division, Risk Analysis Section.
EPA (Environmental Protection Agency). 1989. Evaluation of Ground-Water Extraction Remedies: Volumes 1 and 2. Washington, D.C.: EPA, Office of Emergency and Remedial Response.
EPA. 1990. Superfund Fact Sheet: Who Pays for Superfund? Washington, D.C.: EPA.
EPA. 1992a. Evaluation of Ground-Water Extraction Remedies: Phase II, Volume I—Summary Report. Publication 9355.4-05. Washington, D.C.: EPA, Office of Emergency and Remedial Response.
EPA. 1992b. National Pesticide Survey: Update and Summary of Phase II Results. EPA 570/9-91-020. Washington, D.C.: EPA, Office of Pesticide Programs.
EPA. 1993. Cleaning Up the Nation's Waste Sites: Markets and Technology Trends. EPA 542-R-92-012. Washington, D.C.: EPA, Office of Solid Waste and Emergency Response.
EPA. 1994. An Analysis of State Superfund Programs: 50-State Study—1993 Update. Washington, D.C.: EPA, Office of Emergency and Remedial Response.
Guerrero, P. F. 1991. Superfund: issues that need to be addressed before the program's next reauthorization. Statement before the Subcommittee on Investigations and Oversight, Committee on Public Works and Transportation, U.S. House of Representatives, October 29, 1991. GAO/T-RCED-92-15. Washington, D.C.: General Accounting Office.
Industrial Economics, Inc. 1991. Initial Assessment of the Scope of the Contaminated Media Problem. Washington, D.C.: EPA, Office of Policy Analysis and Contaminated Media Cluster.
Mackay, D. M., and J. A. Cherry. 1989. Groundwater contamination: pump-and-treat remediation. Environ. Sci. Technol. 23(6):630-636.
Mailman, W. L., and W. N. Mack. 1961. Biological contamination of ground water. Pp. 35-43 in Ground Water Contamination: Proceedings of the 1961 Symposium, Cincinnati,
Ohio, April 5-7, 1961. PB-214895. Springfield, Va.: National Technical Information Service.
McCarty, P. L. 1990. Scientific limits to remediation of contaminated soils and ground water. Pp. 38-53 in Ground Water and Soil Contamination Remediation: Toward Compatible Science, Policy, and Public Perception. Washington, D.C.:National Academy Press.
McKee, J. E. 1961. Research needs in ground water contamination. Pp. 205-212 in Ground Water Contamination: Proceedings of the 1961 Symposium, Cincinnati, Ohio, April 5-7, 1961. PB-214895. Springfield, Va.: National Technical Information Service.
Mercer, J. W., D. C. Skipp, and D. Giffin. 1990. Basics of pump-and-treat ground-water remediation technology. EPA/600/8-90/003. Ada, Okla.: EPA.
Middleton, F. M., and F. Walton. 1961. Organic chemical contamination of ground water. Pp. 50-56 in Ground Water Contamination: Proceedings of the 1961 Symposium, Cincinnati, Ohio, April 5-7, 1961. PB-214895. Springfield, Va.: National Technical Information Service.
Office of Technology Assessment. 1989. Coming Clean: Superfund Problems Can Be Solved. PB90-142209. Springfield, Va.: National Technical Information Service.
Russell, M., E. W. Colglazier, and M. R. English. 1991. Hazardous Waste Remediation: The Task Ahead. Knoxville: University of Tennessee, Waste Management Research and Education Institute.
Stipp, D. 1991. Throwing good money at bad water yields scant improvement. Wall Street Journal. May 15, 1991. A1.
Travis, C. C., and C. B. Dory. 1990. Can contaminated aquifers at Superfund sites be remediated? Environ. Sci. Technol. 24(10):1464-1466.