5
Characterizing Sites For Ground Water Cleanup

The previous chapters have discussed how the complexities of the subsurface environment and the types of contaminants present in the subsurface limit the ability of existing technologies to clean up contaminated ground water. However, the poor performance of ground water cleanup systems at many sites to date is not just a function of site complexity and technical limitations; it is also a result of insufficient or inaccurate characterization of the problem prior to cleanup, which has resulted in a flawed design of the cleanup system. For example, at 18 of the 77 sites listed in Appendix A, the remedial actions have failed to contain the ground water contamination. It is likely that the failure of containment at many of these sites is due to inaccurate characterization of the horizontal and vertical extent of contamination. The lack of proper characterization has often occurred even after huge sums of money have been spent and considerable time has elapsed in characterizing the site.

Proper site characterization is an essential component of designing a ground water cleanup system, but at the same time perfection in site characterization is unattainable. Given the complexity of the subsurface, the enormous range of scales that must be taken into account, and the inability to view the subsurface directly, developing a precise image of the contaminated environment is impossible. In fact, attempting to achieve perfection in site characterization can be counterproductive because excessive drilling for the purpose of sampling would destroy the structure that it aims to portray. Thus, site characterization must achieve



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 193
Alternatives for Ground Water Cleanup 5 Characterizing Sites For Ground Water Cleanup The previous chapters have discussed how the complexities of the subsurface environment and the types of contaminants present in the subsurface limit the ability of existing technologies to clean up contaminated ground water. However, the poor performance of ground water cleanup systems at many sites to date is not just a function of site complexity and technical limitations; it is also a result of insufficient or inaccurate characterization of the problem prior to cleanup, which has resulted in a flawed design of the cleanup system. For example, at 18 of the 77 sites listed in Appendix A, the remedial actions have failed to contain the ground water contamination. It is likely that the failure of containment at many of these sites is due to inaccurate characterization of the horizontal and vertical extent of contamination. The lack of proper characterization has often occurred even after huge sums of money have been spent and considerable time has elapsed in characterizing the site. Proper site characterization is an essential component of designing a ground water cleanup system, but at the same time perfection in site characterization is unattainable. Given the complexity of the subsurface, the enormous range of scales that must be taken into account, and the inability to view the subsurface directly, developing a precise image of the contaminated environment is impossible. In fact, attempting to achieve perfection in site characterization can be counterproductive because excessive drilling for the purpose of sampling would destroy the structure that it aims to portray. Thus, site characterization must achieve

OCR for page 193
Alternatives for Ground Water Cleanup a balance between providing enough information to design an effective remediation system and recognizing that significant uncertainties about the site will remain. This chapter discusses the site characterization process and provides the Committee on Ground Water Cleanup Alternatives' assessment of what level of site characterization is adequate. Site characterization is an extensive subject and is covered only briefly in this chapter. For detailed information about the mechanics of site characterization, readers should refer to the guidance documents referenced at the end of this chapter (EPA, 1988a, 1988b, 1989, 1990, 1993a; Mercer et al., 1990; NJDEPE, 1992; Cohen and Mercer, 1993). GOAL OF CHARACTERIZATION Characterization of sites with ground water contamination is conducted to determine the extent of contamination and to select and design a remedy. Site characterization studies should be planned with these purposes in mind. Site characterization studies need not be designed to develop as complete a picture as possible of the subsurface environment. Highly detailed characterization studies may be appropriate at research sites, but they are not needed at most sites. The committee has observed that the characterization process often drags out and produces substantial quantities of the wrong kinds of data because the main goals of the characterization process are forgotten. Various Environmental Protection Agency (EPA) documents accurately convey the purpose of site characterization. For example, the regulation implementing the Superfund law, which is known as the National Contingency Plan, describes the purpose of site characterization in the following language: The purpose of the remedial investigation/feasibility study (RI/FS) is to assess site conditions and evaluate alternatives to the extent necessary to select a remedy. The EPA's guidance document on conducting remedial investigations and feasibility studies (EPA, 1988a) states: The objective of the RI/FS process is not the unobtainable goal of removing all uncertainty, but rather to gather information sufficient to support an informed risk management decision regarding which remedy appears to be most appropriate for a given site. The appropriate level of analysis to meet this objective can only be reached through constant strategic thinking and careful planning concerning the essential data needed to reach a remedy selection decision [emphasis added]. The EPA's memorandum outlining policies for addressing sites with

OCR for page 193
Alternatives for Ground Water Cleanup dense nonaqueous-phase liquid (DNAPL) contaminants (Clay, 1992) reflects a similar intention: The degree of effort expended in locating DNAPL accumulations should be based on the degree of characterization necessary for remedy selection. Locating DNAPL in small stratigraphic or structural discontinuities is generally not possible. However, efforts should be made to identify subsurface geologic environments where DNAPL accumulations may be present. Given the purpose of site characterization as expressed in these documents and the limited number of currently available remedial alternatives for contaminated ground water, it follows that site characterization studies can focus on collecting the data required to choose among a limited number of alternatives. PLUME VERSUS SOURCE Conceptually, the contaminated ground water environment consists of two distinct parts, as explained in Chapter 2: (1) the plume of dissolved contaminants and (2) contaminant source areas. Contaminant source areas include not only typical near-surface sources such as leaking drums, process wastes, and sludges, but also deep subsurface sources such as residual nonaqueous-phase liquids (NAPLs), pools of NAPLs, and metals that have precipitated in mineral phases having low solubility. The prospects for ground water cleanup are much different for the plume of dissolved contaminants than they are for the source areas. Based on this observation, it is clear that site characterization studies should be designed to define early the parts of the site that can be considered source areas and the parts that can be considered as the dissolved plume, because the potential remedial options are significantly different for the two parts. Figure 5-1 illustrates hypothetically how the source and dissolved plume areas might appear for a site contaminated with trichloroethylene (TCE). As shown in the upper part of the figure, the source zone covers a small area relative to the plume of dissolved contamination, which extends far beyond the source zone. As shown in the lower figure, the source zone consists of not only the sludges and drums that were the original source of the TCE, but also accumulations of TCE as a nonaqueous phase along the contaminant flow path. An example of a site with a large subsurface NAPL source is the King of Prussia site discussed in Box 3-5. At that site, the chemical 1,2,3-trichloropropane has migrated both horizontally and vertically large distances from the source area in fractured rock. Trichloropropane at residual sat-

OCR for page 193
Alternatives for Ground Water Cleanup FIGURE 5-1 Source and dissolved plume areas for a hypothetical site contaminated with TCE. Source: After Cherry et al., 1992. uration on fracture surfaces remains as a long-term source of dissolved contaminants. An example of a site with a mineral phase acting as a long-term source of ground water contamination is the United Chrome site in Oregon (listed in Appendix A), where chromium is the dominant contami-

OCR for page 193
Alternatives for Ground Water Cleanup nant (Palmer, 1992). At that site, most of the chromium in the subsurface is present in the mineral barium chromate, which is sparingly soluble. Ground water in contact with this mineral has an equilibrium chromium concentration of only, about 10 mg/liter, and as a result a very large number of pore volumes of water must be pumped through the contaminated zone to remove all of the barium chromate, which acts as a long-term source of dissolved chromium. MANAGEMENT OPTIONS The site characterization process is greatly simplified by the observation that there are not many remedial management options for contaminated ground water. The options, illustrated in Figure 5-2, are essentially (1) contain the contaminated ground water, (2) restore the contaminated ground water, or (3) do nothing if the contaminated ground water poses no risk and/or natural attenuation will be sufficient to resolve the problem. Characterization studies should determine early on which general remedial approach is appropriate for each subarea of the site. In many cases, the appropriate remedy for source areas will be containment, and the appropriate remedial goal for the plume of dissolved contaminants will be active restoration. Natural attenuation may be acceptable for some zones where it can be shown that natural biological, chemical, and physical processes will lower contaminant concentrations to cleanup goals before the contaminants reach receptors; it may also be acceptable for ground water that discharges into surface water bodies where it can be demonstrated that the ground water will not be used and the resulting concentrations in the surface water body will be below applicable standards. Once the goal of remediation is established, site characterization studies should be designed to collect the data required to design a technically effective and cost-effective remedial action to obtain the desired end result (whether containment, active restoration, or natural attenuation). For some sites, determining the appropriate general remedial approach requires only a limited amount of data. For example, consider a site where residual polychlorinated biphenyls (PCBs) are found below the water table in fractured bedrock as the result of the past migration of a DNAPL containing PCBs. In most environments, migration of PCBs is very limited because of their strong tendency to sorb to solid materials in the aquifer, with the PCBs traveling at as little as one-hundredth the rate of the ground water. As a result, in most situations such a PCB plume consists of the area containing the DNAPL (generally present at residual saturation) and a very small fringe of dissolved PCBs. Because the plume of dissolved PCBs is very small, almost the entire plume can be considered as the source area. The appropriate remedial measure at such a site

OCR for page 193
Alternatives for Ground Water Cleanup FIGURE 5-2 Options for managing sites with contaminated ground water. In the option shown at the top, a pumping well is installed to prevent enlargement of the plume of dissolved contaminants but not to clean up the plume. In the middle option, a system is installed to clean up the plume, while a containment system is installed around the contaminant source area. In the bottom option, site conditions are such that the plume, which has no significant remaining source areas, attenuates without human intervention. SOURCE: Adapted from Cherry et al., 1992. will likely be containment because of the inability to remove the PCBs with existing technologies as a result of their extremely low solubility. DATA REQUIREMENTS The level of characterization required to select a remedial action is that which is sufficient to determine the extent of contamination, the res-

OCR for page 193
Alternatives for Ground Water Cleanup toration potential of each subarea of the site, and potential receptors of the contaminated water. In simplest terms the required data include information to define or estimate the horizontal and vertical extent of ground water contamination, information to estimate the locations of contaminant source areas, information to describe the hydrogeologic setting, and information to estimate the site's restoration potential. Hollow-stem auger used to obtain subsurface samples and install monitoring wells. Courtesy of U.S. Environmental Protection Agency, R. S. Kerr Environmental Research Laboratory.

OCR for page 193
Alternatives for Ground Water Cleanup FIGURE 5-3 Categories of data required to select and design a ground water remedial action (adapted from EPA, 1993b). These data are sufficient to formulate a conceptual model of the site that will be necessary for analyzing such fundamental questions about the contaminated ground water system as Where is the contamination and what chemical compounds are present? How can further migration of the contamination be stopped or reduced to acceptable levels? What receptors are at risk, and can that risk be characterized? Is restoration of the ground water possible with existing technologies, and if so, what are the technical options for remediation? Figure 5-3 illustrates the basic elements of a site conceptual model.

OCR for page 193
Alternatives for Ground Water Cleanup Horizontal And Vertical Extent Of Contamination Defining the areas of the subsurface that contain contaminants at concentrations in excess of health-based levels is the obvious first step in developing an appropriate management strategy for the contaminated ground water. Nevertheless, at many sites where remediation is ongoing, the horizontal and vertical extent of ground water contamination has not been defined. The main reasons for this are simple: (1) in complex hydrogeologic systems, defining the extent of contamination is expensive and time consuming, and (2) zones of ground water contamination seldom respect property boundaries, sometimes creating difficulties in obtaining permission to collect samples. Defining the extent of contamination in complex hydrogeologic environments generally requires the drilling of a large number of boreholes from which samples of ground water are collected either directly or from monitoring wells installed in the boreholes. Many nonintrusive or minimally intrusive techniques are available for estimating ground water quality, as are many analysis techniques for estimating the extent of contamination from existing data on the hydrogeologic system, but none of these techniques is a reliable indicator of ground water contamination. Nondisruptive characterization methods—such as ground-penetrating radar, electroconductivity measurements, and microwave radiation—all require that the contaminant alter a physical property of the ground water to a degree detectable by the geophysical technique. Unfortunately, the relatively low contaminant concentrations typically found in ground water are difficult to measure with these nondisruptive methods. Similarly, minimally intrusive techniques, such as soil vapor analyzers, can be useful for indicating the location of plumes near the water table, but they are not useful for indicating deep areas of contamination. Even when used for indicating contamination in the water table zones, these methods are susceptible to a large number of both false positives and false negatives because they have detection limits of 1 to 10 µg/liter for many common organic chemicals, which is the range of health-based cleanup levels for many contaminants. The data collected from wells operated as part of a pump-and-treat system generally are not useful for defining the horizontal and vertical extent of contamination because these data only provide information on average conditions in the vicinity of the well, not precise concentrations for a given location. Therefore, in defining the extent of contamination, there are no true short cuts to collecting and analyzing ground water samples. In most situations, ground water samples are most efficiently collected from monitoring wells, and many—and often deep—monitoring wells are required to define the hor-

OCR for page 193
Alternatives for Ground Water Cleanup izontal and vertical extent of contamination. In some situations, the characterization process can be expedited by collecting samples directly from boreholes, but the usefulness of these techniques is limited because it is difficult to check sample reproducibility. What constitutes adequate vertical definition of a ground water problem is a frequently asked question in site characterization studies. At many sites, especially those with DNAPLs in fractured rock, contamination can extend to depths of 300 meters and more. Adequate vertical definition for purposes of designing a remedy to protect human health and the environment does not necessarily require installation of monitoring wells below the zone of contamination. However, it does require an understanding of the flow direction in the deep zones and the flow paths emanating from the deep zones. If an existing monitoring system shows contamination in the deepest wells but also shows the existence of strong upward hydraulic gradients, then it is probably not necessary to install wells below the zone of contamination. On the other hand, if downward hydraulic gradients exist and there is no information on potential contaminant migration pathways, then most likely deeper monitoring points need to be installed. Sampling requires entering property to collect soil samples, drill or sample wells, and perform other tests. As a result, the investigator must obtain permission from the landowner or have a legal right to enter for this purpose, and this can sometimes create problems. Federal and state environmental statutes typically grant government agencies and their delegates the right to enter private property to take samples at reasonable times for reasonable purposes related to the implementation of the statute. Typically, government agencies notify property owners and enter into voluntary access agreements that indemnify the property for any injury caused by sampling (Clay, 1991). However, in addition to compensation for damage caused by the sampling itself, in some cases the landowner may have a potential claim for property damage or personal injury relating to the ground water contamination if contamination is detected. The landowner may raise these claims when access is requested and may even attempt to condition access on resolution of such claims. In other cases, landowners may be reluctant to allow sampling on their property because they may fear being held liable for any contamination found. Locations of Contaminant Source Areas Successful site characterization requires identifying the locations of potential sources of contaminants so that they can be removed or contained. As illustrated by some of the case studies in Chapter 3, failure to

OCR for page 193
Alternatives for Ground Water Cleanup Monitoring the water level in a well during an aquifer pump test. Courtesy of U.S. Environmental Protection Agency, R. S. Kerr Environmental Research Laboratory. remove or contain these sources will result in continuing dissolution of contaminants when the remediation stops, which will preclude aquifer restoration. Identification of shallow sources of dissolved contaminants and NAPL sources where the NAPL is less dense than water (i.e., LNAPLs), is relatively straightforward and is discussed in the guidance documents referenced at the beginning of this chapter. Identification of DNAPL sources and precipitated metals is not straightforward, although current EPA policy reflects the need to identify DNAPL sources (Clay, 1992; EPA, 1993b). DNAPL sources are difficult to identify because they are denser than water, can migrate below the water table (often to great depths), and generally move along relatively discrete pathways. (In contrast, LNAPLs generally spread out on the water table in a relatively continuous plume that is easy to locate.) Ideally, DNAPL presence can be identified by visual examination of subsurface samples, but this is often not the case. The conventional model of subsurface DNAPLs (as illustrated in Figure 5-1) shows the DNAPLs pooling on low-permeability layers. At sites where most of the DNAPL is present in pools, the DNAPL can be observed in fluid samples taken from wells penetrating the DNAPL zone.

OCR for page 193
Alternatives for Ground Water Cleanup However, direct observation of DNAPLs is extremely rare. Field experience indicates that in many cases, the majority of the DNAPL is present not in pools, but as residual saturation (small globules trapped in the porous materials of the aquifer). In these situations, the free-phase DNAPL originally present in the aquifer migrated until it reached residual saturation. In general, DNAPL at residual saturation is not observable in subsurface samples. When the DNAPL at residual saturation ages—as a result of slow mass transfer into the soil microstructure, dissolution, and biodegradation—detection becomes even more difficult, yet the aged residual can still provide a significant source of contamination. When subsurface DNAPLs are present at residual saturation, indirect means are generally necessary to determine their presence. Determination of chemical concentrations in water is the best indirect method (Cohen and Mercer, 1993). Concentrations near the solubility of the DNAPL compound are positive confirmation that DNAPLs are present. However, generally DNAPL compounds are detected at less than 10 percent of their aqueous solubility limit in ground water due to the effects of nonuniform ground water flow, variable DNAPL distribution, mixing of ground water in sampling wells, and reduced effective solubility of individual compounds in a multiliquid NAPL mixture (Feenstra et al., 1991). Field work has shown that contaminant concentrations of greater than 1 percent of the aqueous solubility limit are highly suggestive of NAPL presence. Concentrations of less than 1 percent, however, do not preclude the presence of NAPLs (Cohen and Mercer, 1993). The 1 percent rule is frequently used in site characterization studies, and, given the available characterization technologies, it is the best means for estimating the extent of DNAPLs in the subsurface. At many sites where DNAPL presence is suspected, investigators are reluctant to define the vertical extent of contamination because they fear remobilizing the DNAPLs and inducing new downward migration. A recent EPA memorandum (Clay, 1992) states, ''Drilling through DNAPL zones into deeper stratigraphic units should be avoided.'' Potential remobilization of DNAPLs is a concern, but it does not mean that drilling should be avoided. Field experience indicates that remobilizing DNAPLs by drilling is extremely uncommon, possibly because in many cases much of the DNAPL may be present at residual saturation rather than in pools. Avoiding defining the vertical extent of the DNAPL zones at these sites will in many cases make it impossible to design an appropriate remedy. At many sites, a lack of vertical defining will give the illusion that the DNAPL zone is much larger than it actually is. At present, locating source areas consisting of metals that have precipitated as mineral phases is even more difficult than locating source areas containing DNAFLs. The experience with these types of sites is

OCR for page 193
Alternatives for Ground Water Cleanup Collection of a ground water sample from a well. Courtesy of Rice University, Department of Environmental Science and Engineering. very limited compared to sites with DNAPLs, and therefore no rules of thumb have been developed. From a practical perspective, it is almost impossible to collect data prior to pumping and treating that would allow determination of whether a mineral phase is acting as a long-term contaminant source. At the United Chrome site listed in Appendix A, the conclusion that a mineral phase was controlling the cleanup rate was a hypothesis that was developed and tested after data from a pump-and-treat system indicated that concentrations were not changing with time. From a theoretical perspective, describing the processes controlling min-

OCR for page 193
Alternatives for Ground Water Cleanup eral phases is much more difficult than describing the processes controlling NAPLs because of the strong influence of ground water chemistry on the behavior of the mineral phase. Hydrogeologic Setting In the first phase of site characterization, a broad description of the hydrogeologic setting is required to estimate the site's restoration potential and to identify potential ground water receptors. The basic information required at this stage is a description of the stratigraphy, thickness, and continuity of the aquifers and the extent of preferential pathways such as fractures, buried stream beds, and solution cavities. In later phases of site characterization, data are required to determine the rates and direction of ground water flow, an understanding of which is necessary to design a cleanup system and to define the contaminated area. The subsurface properties required to determine rates and direction of ground water flow, which are discussed in detail in Chapter 2, include the following elements: the hydraulic properties of the contaminated region (e.g., hydraulic conductivity, porosity, and storage coefficient distributions); system stresses (e.g., recharge, discharge, pumping rates, and hydraulic gradient); and system geometry (e.g., physical and hydraulic boundaries, stratigraphy, lithology, structural geology, and heterogeneities). A particularly difficult problem in characterizing ground water flow is heterogeneity in aquifer properties, especially permeability. Fractured rock settings comprise an extreme example. According to EPA guidance (EPA, 1991), a complete description of a contaminated fractured rock system would include the following data: (1) the dimensions of the system; (2) individual fracture lengths, aperture widths, locations, and orientations; (3) the hydraulic head throughout the system; (4) the porosity and permeability of the rock matrix; (5) the sources of water and contaminants; (6) the nature and concentrations of the contaminants throughout the system; and (7) the chemical interactions between the contaminants and the rock matrix. Presently, collection of such detailed information is neither technically possible nor economically feasible at the scale of most contaminated sites. In addition, this detailed information is generally not needed to design a pump-and-treat system to contain the contaminated ground water, which is likely the only realistic objective for extensively contaminated fractured rock environments. Restoration Potential Traditionally, site characterization studies have been conducted with

OCR for page 193
Alternatives for Ground Water Cleanup Excavating a trench during installation of piping for a ground water monitoring system. Courtesy of U.S. Environmental Protection Agency, R. S. Kerr Environmental Research Laboratory. the implicit understanding that sufficient information would be collected to design a pump-and-treat system capable of restoring the aquifer to certain standards, usually health based. As this report has documented, however, at many sites complete restoration to health-based standards is not possible with present technologies. Thus, an explicit step in the site characterization process should be to estimate the restoration potential of each subarea of the contaminated site. Table 3-2, in Chapter 3, shows a method for estimating restoration potential developed by the committee. In this table, the relative ease of cleanup is a function of the hydrogeologic setting and the properties of the contaminants. At one extreme are sites in category 1: these sites have mobile dissolved contaminants and homogeneous geology; they can generally be restored to health-based standards. At the other extreme are sites in category 4: these sites have NAPLs and complex geologic conditions; restoration of such areas to health-based standards is highly unlikely with present technologies. In the middle are sites in categories 2 and 3: these sites may have sorbing contaminants and LNAPLs present under varying geologic conditions; restoration of some of these sites may be possible, but restoration of others will not. For most sites in categories 2 and 3, the only way to accurately estimate the restoration potential will be to evaluate the performance of an interim or pilot-scale remediation system. It may be necessary to operate such an interim system for long

OCR for page 193
Alternatives for Ground Water Cleanup periods of time in order to estimate restoration potential, because removing several aquifer volumes of water from the area of interest may be necessary to fully understand the characteristics of the site. As noted in Chapter 2, cleaning up any contaminated site is theoretically possible, provided the cleanup can continue for an infinitely long time period into the future. Therefore, evaluation of restoration potential often reduces to an evaluation of whether or not restoration can be achieved in a reasonable time period. Chapter 3 describes techniques for estimating cleanup time, using the data discussed above. However, as discussed in Chapter 3, these techniques generally produce estimates that have a large uncertainty associated with them and often have a bias toward underestimating cleanup times. As a result, usually the only way to accurately estimate the cleanup time is to evaluate the performance of an interim or pilot-scale remediation system. In considering a site's restoration potential, the importance of distinguishing between dissolved plume areas and contaminant source areas cannot be overemphasized. Often, dissolved plume areas will have a high restoration potential, while source areas will not. STAGES OF SITE CHARACTERIZATION AND MANAGEMENT Almost all ground water remedies will be long-term actions, requiring much more than five years to complete. In fact, many ground water cleanups will last for decades, if not centuries. Because of the complexity of the subsurface environment and the long time frames required for ground water cleanup, site characterization must not be viewed as a onetime event—or even a several-time event—that ends when the final remedial investigation report is submitted. Hydrogeologic and water quality data will be collected throughout the life of a ground water cleanup system, and these data should be reviewed regularly to determine if the cleanup system is meeting its objectives and, if not, whether adjustments are needed. Site characterization consists of several iterative stages, starting with investigation of the site's history and continuing throughout remediation. Figure 5-4 illustrates these stages. In the initial stage, the investigator reviews background information about the site and previously collected data to determine the likely original sources of contamination and what additional information is required. In the intermediate stages, the investigator collects sufficient information to define the volume of contaminated ground water and to design and implement a containment system. Also in the intermediate stages, the investigator defines potential long-term subsurface sources of contamination and collects sufficient

OCR for page 193
Alternatives for Ground Water Cleanup FIGURE 5-4 Iterative stages of site characterization (adapted from EPA, 1993b).

OCR for page 193
Alternatives for Ground Water Cleanup information to design and implement a remedial action to contain and/or remove these sources. In later stages, the investigator evaluates the restoration potential of each subarea of the site and collects sufficient information to design a system that will meet that potential. As the stages of site characterization proceed and as knowledge of the site becomes more complex and complete, the conceptual model may be represented in part by a mathematical model to provide a tool for analyzing remedial alternatives and for planning additional data collection. Containment systems can generally be designed with the data gathered in the early stages of site investigation. However, in many situations (represented by categories 2 and 3 in Table 3-2) it is difficult, if not impossible, to collect the data required to determine the site's restoration potential in the early stages of investigation. In such situations, the data required for determining restoration potential and for designing a system to meet that potential can best be collected by stressing the system by pumping from contaminated locations and observing the response. The data collected from operating a containment system can provide one such indicator of the ground water system's response to stresses. Ground water systems are dynamic and complex, and as a result some uncertainty will always remain regarding the performance of a remedial action. The data collection program should not be overdesigned in an attempt to eliminate all uncertainties, as this will be impossible. Rather, remedial designs should be robust, should include monitoring systems to detect failure, and should include contingency plans to make changes in remedial measures if failure occurs. Mistakes are inevitable, and thus it is essential that the remediation plan include a mechanism to detect and correct failure to control the contamination. RESEARCH NEEDS Site characterization is often a slow process, in part because of the long time required to conduct a thorough analysis with existing subsurface sampling technologies. Installation of remediation systems in the past has been delayed because of the limits on the ability to collect and analyze site data quickly; where containment systems were not installed, the contamination problem worsened with the passage of time while site characterization proceeded. Remediation efforts have also suffered from the inability of site characterization to accurately predict early on the restoration potential of the site and to quantify and locate subsurface sources of contamination. Research to improve the speed of site characterization and its ability to locate sources of contamination would greatly improve the ability to carry out effective ground water cleanups. At the same time, however, one must realize that even with improvements in

OCR for page 193
Alternatives for Ground Water Cleanup site characterization technology, a complete picture of the subsurface will be impossible to obtain, and in many cases operation of a remediation system (whether pilot or full scale) will be the only way to accurately determine the site's restoration potential. Research to address the following questions would help advance the site characterization process: How can the extent of ground water contamination be assessed rapidly? How can the location and mass of contaminants present in source areas, especially those with NAPLs, be more accurately quantified? How can the characterization process be improved to provide more reliable and timely estimates of the restoration potential of contaminated aquifers? What measures—technical and institutional—are needed to ensure the setting of realistic cleanup objectives and the optimal design of a cleanup system based on site characterization data? What measures—technical and institutional—are needed to ensure that knowledge about the site gained during the course of remediation will be considered in reassessing cleanup objectives and the design of the treatment system during long-term cleanup projects? CONCLUSIONS Based on a review of the processes for characterizing contaminated sites in order to design ground water cleanup systems, the committee reached the following conclusions: The goal of site characterization studies should be the collection of sufficient data to select and design a remedial action. Site characterization studies need not be designed to develop as complete a picture as possible of the subsurface environment. These types of characterization studies may be appropriate at research sites, but they are not needed at most sites. The subsurface environment is complex, and as a result, the committee has observed that the characterization process often drags out because the main goals of the characterization process are forgotten. Site characterization studies should define early the parts of the site that can be considered source areas and the parts that can be considered dissolved plume areas. Restoration of dissolved plume areas may be possible at many sites, while restoration of subsurface source areas may not be possible. At the same time, however, the restoration potential of the dissolved plume areas cannot be realized until the source areas are contained. At many sites, operation of prototype systems will be necessary

OCR for page 193
Alternatives for Ground Water Cleanup to obtain the data required to design and predict performance of full-scale remedies. Because of the complexity of many contaminated sites, predicting how the ground water system will respond to operation of the remediation system may not be possible without observing the effects of a prototype system. Improved methods are needed for assessing the scope and distribution of contamination. Existing site characterization techniques cannot always accurately locate subsurface sources of contamination and are often costly and slow. REFERENCES Cherry, J. A., S. Feenstra, and D. M. Mackay. 1992. Developing rational goals for in situ remedial technologies. Paper presented at Subsurface Restoration Conference, Dallas, June 21-24, 1992. Clay, D. R. 1991. Policy Toward Owners of Residential Property at Superfund Sites. Directive 9834.6. July 3,1991. Washington, D.C.: EPA. Clay, D. R. 1992. Considerations in Ground-Water Remediation at Superfund Sites and RCRA Facilities—Update. Directive 9283.1-06. May 27, 1992. Washington, D.C.: EPA. Cohen, R. M., and J. W. Mercer. 1993. DNAPL Site Evaluation. Boca Raton, Fla.: C. K. Smoley. EPA (Environmental Protection Agency). 1988a. Guidance for Conducting Remedial Investigations and Feasibility Studies Under CERCLA. Directive 9355.3-01. EPA/540/ G-89/004. Washington, D.C.: EPA. EPA. 1988b. Guidance on Remedial Actions for Contaminated Ground Water at Super-fund Sites. Directive 9283.1-2. EPA/540/G-88/003. Washington, D.C.: EPA. EPA. 1989. RCRA Facility Investigation (RFI) Interim Final Guidance. Directive 9502.00-6D. EPA/530/SW-89-031. Washington, D.C.: EPA. EPA. 1990. Handbook—Ground Water, Volume I: Ground Water and Contamination. EPA/625/6-90/016a. Washington, D.C.: EPA. EPA. 1991. Handbook—Ground Water, Volume II: Methodology. EPA/625/690/016b. Washington, D.C.: EPA. EPA. 1993a. RCRA Ground-Water Monitoring: Draft Technical Guidance. EPA/530-R-93-001. Washington, D.C.: EPA. EPA. 1993b. Guidance for Evaluating the Technical Impracticability of Ground-Water Restoration. Directive 9234.2-25. Washington, D.C.: EPA. Feenstra, S., D. M. Mackay, and J. A. Cherry. 1991. A method for assessing residual NAPL based on organic chemical concentrations in soil samples. Groundwater Monit. Rev. 11(2):128-136. Mercer, J. W., D. C. Skipp, and D. Giffin. 1990. Basics of Pump-and-Treat Ground-Water Remediation Technology. EPA/600/8-90/003. Washington, D.C.: EPA. NJDEPE (New Jersey Department of Environmental Protection and Energy). 1992. Field Sampling Procedures Manual. Trenton, N.J.: NJDEPE. Palmer, C. 1992. Limitations to pump-and-treat remediation of a chromium-contaminated site. Abstract presented at National Ground Water Association Conference: Aquifer Restoration—Pump-and-Treat and the Alternatives, Las Vegas, September 30-October 2, 1992.