Click for next page ( 4


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 3
Overview and Recommendations TH E PROB LE M The reliable assessment of hazards or risks arising from grounclwater contamination problems and the design of efficient and effective techniques to mitigate them require the capability to predict the behavior of chemical contaminants in flowing groundwater. Reliable and quantitative predictions of contaminant movement can be made only if we uncterstand the processes controlling transport, hydrodynamic dispersion, and chem- ical, physical, and biological reactions that affect soluble concentrations in the ground. This report acIdresses the status of scientific understanding of transport of contaminants in grounc3water in terms of hydrologic theory (Chapters 2 and 3) and case studies of occurrences of grounc~water contamination (Chapters 6-121. The widespread use of chemical products, coupled with the disposal of large volumes of waste materials, poses the potential for widely distributed grounc~water contamina- tion. New instances of groundwater contamination are continually being recognized. Hazardous chemicals, e.g., pesticides, herbicides, ant] solvents, are used ubiquitously in everyday life. These and a host of other chemicals are in wiclespreac! use in urban, inclustrial, and agricultural settings. Whether intentionally clisposec] of, accidentally spilled, or applied to the ground for agricultural reasons, some of these chemicals can eventually reach the groundwater and contaminate it. Because of the volumes of toxic wastes and because of their stability in groundwater, such contamination can pose a serious threat to public health. Grounc~water is the subsurface transporting agent for dissolved chemicals including contaminants. Materials dissolvecl from the wastes may be transported from the burial or disposal site by groundwater flow, with the result that the quality of water from wells is recluced by the contaminated groundwater. In acIdition, natural discharges of an aquifer, such as at springs and seeps, can return a contaminant to the surface. Because of the slow rates of grounc~water movement ant] natural flushing of aquifers, when areas are contaminated, they commonly remain so for decades or longer. The major geo- physical inputs to the problems of waste disposal and grounc~water contamination deal with the chemistry and rates and directions of contaminant transport. 3

OCR for page 3
Overview and Recommendations Totals The challenges are (1) to prevent the introduction of contaminants in an aquifer, (2) to predict their movement if they are introduced, and (3) to remove them, to the extent possible, to protect the biosphere effectively. The largest potential source of contamination of grounc~water is the disposal of solid ant] liquid! wastes. During the past several decades, legislation reflecting environmental concerns has attempted to restrict air and surface-water pollution; this has resulted in increased disposal of wastes in the subsurface. In 1980, the U.S. Environmental Pro- tection Agency (EPA) estimated that there were 200,000 landfills and dumps receiving 150 million tons per year of municipal solid wastes ant! 240 million tons per year of industrial solid wastes. In addition to landfills, 176,000 surface impoundments receive 10 trillion gallons per year of liquid] industrial wastes. The results of the EPA surface impoundment assessment are shown in Table 1. Not all wastes are hazardous. However, the EPA (1980) estimated that 142,000 tons of hazardous wastes are generated daily in the United States approximately 60 million tons annually. These wastes are generated at more than 750,000 sites. To some extent many of the larger industrial sites hacI, in the past, lancifilIs on their property, which receiver] much of their plants' waste. The EPA (1980) estimated that 50,000 such sites have been usecI, at some time, for the disposal of hazardous wastes. Of these, 1200 to 2000 are thought to pose threats to the environment. Waste disposal is not the only source of groundwater contamination. As additional sources, the EPA (1980) also lists septic systems, agriculture, acciclental leaks and spills, mining, highway de-icing, artificial recharge, underground injection, and saltwater encroachment. Current estimates of the extent of grounc~water contamination suggest that 0.5 to 2.0 percent of the groundwater in the conterminous Uniter] States may be contaminated (see Chapter 1~. Although this might not seem to indicate a large problem, much of the contamination occurs in areas of heaviest reliance on groundwater. Several grounct-burial disposal practices and repositories have been in use for many years and have been shown to cause minimal or no grounc~water contamination. How- ever, some waste-clisposal practices have resulted in irreversible contamination of groundwater. The examples of Love Canal, the EPA list of more than 400 contaminated sites, and others indicate the magnitude of the resulting grounc~water contamination problems with possible health hazards and other deleterious effects. Restoration of contaminates] aquifers can be extremely expensive. This may involve TABLE 1 Preliminary Findings of the EPA Surface Impoundment Assessmenta 10,819 19, Il6 14,677 7,100 24,527 1,500 77,739 25,749 36,179 19,167 24,451 64,951 5,745 176,242 Category Industrial Municipal Agricultural Mining Oil/gas brine pits Other Sites Located Impoundments Located Sites Assessed 8,193 10,675 6,597 1,448 3,304 327 30,544 The Surface Impoundment Assessment also released analysis based on data from the assessed industrial sites. Almost 70 percent of the industrial impoundments are unlined. Only 5 percent are known to be monitored for groundwater quality. About one third of the impoundments contain liquid wastes with potentially hazardous constituents. Analysis of sites for the chemical and allied products industry reveals similar findings with the exception that over 60 percent of the sites may contain liquid wastes with potentially hazardous constituents. 4

OCR for page 3
Overview and Recommendations drilling many wells and pumping vast quantities of groundwater. The pumped water can often be treated, to reduce the concentration of contaminants, and reinfected into the aquifer. For example, the current estimated costs for contaminant containment efforts at the Rocky Mountain Arsenal, Colorado, are about $100 million; the cost estimates for "total" decontamination of the Arsenal range from $800 million to $1 billion (U. S. Army, 1982~. The question of cost/benefit trade-offs to society in these cases needs to be examined carefully; some sites may prove to be so expensive to restore that they may have to be designated as permanently contaminated. The subsurface can be used for waste repositories. However, such repositories should be selected, designed, and engineered in terms of the hydrology, geology, hydrogeo- chemistry, and microbiology of a particular site and the characteristics of the specific wastes. Care must be taken to provide isolation of potentially toxic substances from the biosphere for long periods of time. The necessary isolation time depends on the toxicity of the particular contaminant and on the extent to which the substances will be degraded or diluted to below toxic levels over time with a high degree of certainty. For example, current estimates concerning high-level radioactive wastes suggest that a period of ~ .. r ., ~ . , 1 . ~ ~ . ~ ~ _ ~ V ~ ~~ ~ Isolation from tne o~ospnere of 1000 to lO,OOO years will allow decay of many of the hazardous radionuclides to safe levels (NRC, 19831. However, certain nondegradable nanny toxic substances along warn some of the radioactive wastes with very long-lived radionuclides may require permanent isolation. If the subsurface is to be utilized as a repository of wastes, both increased scientific knowledge and improved engineering techniques must be brought to bear on the problem of groundwater contamination. The development of better methods for man- aging and disposing of today's waste and for rectifying yesterday's mistakes will be necessary. The scientific problem is one of understanding the physical-chemical-bio- logical system sufficiently well to be able to predict the movement and fate of contam- inants. This challenge is not met easily, but with concentrated effort it should be possible to isolate toxic substances in the ground in certain types of environments in a way that will pose no hazards. There seems to be a general, although inaccurate, impression that little is known about the occurrence of groundwater and that the science of groundwater hydrology is in its infancy. This stems in part from the fact that groundwater occurs in the subsurface, out of sight, and can only generally be investigated indirectly, e. g., through observations made in boreholes. Problems of flow through porous media have been investigated since the work of Henry Darcy in 1856. Flow and transport in the subsurface continue to be studied in petroleum engineering, groundwater hydrology, soil science, and, to some extent, chemical engineering. While the applications of the scientific knowledge often differ, the physical and chemical processes that apply are the same. These disciplines com- plement one another, and to a large extent technology has been exchanged between them. During the course of this study, it became apparent that the unsolved technical problems dealing with groundwater contamination pale beside the institutional issues (introduced in Chapters 13 and 14) and the public's perception of risk associated with waste disposal. Past economic considerations may be interrelated with the problems associated with groundwater contamination. The disposal of wastes in the ground has often been viewed as attractive because of the slow movement of groundwater relative to the rapid contaminant transport in surface waters or the atmosphere. Therefore, the potential cleanup costs were delayed costs that did not constitute part of the then current operational expenses. In addition, costs that might be incurred decades later took account of the present and future value of the dollar. This leads to the question: Is the present worth of such far-or future costs truly so small or negligible that it makes no "economic" sense to invest now to prevent those future damages? Although we do 5

OCR for page 3
Overview and Recommendations not have an answer to this question, we should consider that the physical reasons responsible for slow groundwater transport are the same reasons that make the cleanup and mitigation of its contamination more difficult and expensive. Because hazardous chemicals are so widely used, their disposal in a safe and elective manner is especially difficult. The Resource Conservation and Recovery Act (RCRA) legislated a "cradle to grave" manifest control system for hazardous substances. The system is intended to track the whereabouts of hazardous materials until they are used or disposed of. In most past practice, the records kept on the nature and volume of wastes disposed of in a particular site have been incomplete. This has creates! one of the problems in analyzing the movement of contaminants in that the source of pollutants . l~ 1 1 T ~ . .1 _ 1 . 1 . 1 . 1 1 . 1 ~ ~~ . ~ . . 1 . 1 is usually poorly known. Until the system legislated by the H(,HA is edectively imple- mented, the whereabouts of hazardous materials will remain uncertain. Environmental legislation the Safe Drinking Water Act, the Resource Conservation and Recovery Act, and the Comprehensive Environmental Response Compensation and Liability Act (commonly referred to as Superfund) further restrict disposal of wastes in or on the solid earth. Taken collectively, the environmental legislation, which has restricted the use of the air and surface water for disposal of wastes, has now limited the use of the land for waste disposal. Clearly a modern society must either dispose of the enormous quantities of residual products (wastes) that it generates or in some way reprocess those residuals. While restrictions placed by legislation protect the environ- ment, one can question whether we have moved very far toward developing a rational strategy of waste disposal that provides long-term protection to society and the envi- ronment. A strategy of disposal or reprocessing, or both, must be forthcoming if we are to act responsibly toward future generations. GROUNDWATER: THE RESOURCE Groundwater is a heavily used resource; its magnitude has been described in the U.S. Water Resources Council Bulletin 16 (1980~: The Nation's groundwater resource is enormous it is our largest freshwater source in terms of volume in storage. Beneath the conterminous United States lie some 65 quadrillion gallons or 200 billion acre-feet of groundwater within a few thousand feet of the land surface, part of which is replenishable upon use. However, it should be recognized that not all of this water is practicably recoverable. Withdrawals amounted to about 83 billion gallons a day in 1975. This withdrawal is approximately 20 percent of the total withdrawal use of water in the Nation excluding hydroelectric use; it constitutes only a fraction of the groundwater development possible. Groundwater constitutes a significant source of water in almost all of the United States, particularly in the West, where it is heavily used for irrigation. California alone accounts for 23 percent of national groundwater use, and California and Texas together account for 37 percent. SOURCES OF GROUNDWATER CONTAMINATION Almost every major industrial and agricultural site has in the past disposed of its wastes on site, often in an inconspicuous location on the property. Every municipality has had to dispose of its waste at selected locations within its proximity. Accidental spills of toxic chemicals have also occurred, often without particular attention to or concern for the consequences some practices of cleaning a toxic spill involve flushing it with water until it.disappears into the ground. Past waste-disposal practices and clearing with spills have not always considered the potential for groundwater contamination. 6

OCR for page 3
Overview and Recommendations FIGURE 1 Schematic representation of contaminant plumes possibly associated with various types of waste disposal. Groundwater contamination (see Figure 1) may be localized or spread over a large area, depending on the nature and source of the pollutant and on the nature of the groundwater system. A problem of growing concern is the cumulative impact of con- tamination of a regional aquifer from nonpoint sources (i.e., those that lack a well- clefined single point of origin), such as those created by intensive use of fertilizers, herbicides, and pesticides. In addition, small point sources such as numerous do- mestic septic tanks or small accidental spills from both agricultural and industrial sources threaten the quality of regional aquifers. The situation on Long Island, New York, illustrates the impact of widely distributed small point sources of pollution on an aquifer system where approximately 3 million residents rely on wells as their sole source of water supply. Domestic wastewater seeping from thousands of septic systems and leach- ates from landfills and industrial waste-disposal sites have contaminated the shallow groundwater in many parts of Long Island, as described in Chapter 9. Research is neecled to evaluate the groundwater resources threatened in this manner in order to understand how contamination from such disuse sources is attenuated by dispersion ant] chemical reactions in the groundwater systems. Septic tanks are a frequently used method for disposal of sewage. Where they are properly sited, such as in sparsely populated areas and in soils with good drainage above the water table, septic tanks generally pose little or no hazard. All too frequently, however, they are installed with drain fields that are too small and intersect nearby groundwater supply wells. In such situations, sewage often contaminates wells in the area or moves to the land surface, or both. Even where septic systems are well drained, they may eventually pollute the groundwater. The EPA (1980) found that about a third 7

OCR for page 3
Overview and Recommendations of all septic tank installations are not operating properly and that the consequent pol- lution both above and below ground is substantial. The solution to grounc~water con- tamination from septic systems, beyond better engineered on-site facilities or improved maintenance, may lie in better land-use control and in effective regulations for septic tank installation. Sewage-disposal activities are introducing viruses into a variety of groundwater sources; but the persistence and movement of these viruses has only recently become the subject of scientific inquiry, and the extent to which such viruses in groundwater pose a hazard to public health is still largely unknown. Recent technological advances have made it possible to recover minerals that occur in low concentrations in the Earth's crust through in situ leaching. (Uranium is com- mercially mined by this method.) Research on methods of recovering disseminated copper has led to similar techniques being applied at several southwestern U. S. copper mines. By combining injection and production wells in a manner similar to their use in oil reservoirs, groundwater flow through the ore body is controlled. Chemical reagents are usually introduced that make the mineral of interest more soluble and thereby more economically recoverable by the moving fluids. In Texas, uranium is being mined by in situ leaching in rocks that have a potential for use as a freshwater supply. Care must be taken to flush out or neutralize the uranium-leaching solutions, which, if they re- mained, would contaminate the aquifer (see Chapter 12~. The removal of the injected! chemical reagents following mining can be made more complicated if the reagents are dispersed in the groundwater. Dispersion requires circulation of larger quantities of fluid than originally envisages! to accomplish adequate cleanup. Only in a small number of mining operations has such cleanup been under- taken. None has completed the stage of grounc~water restoration, although a few pilot operations are in progress. DISPOSAL PRACTICES A number of practices have resulted in migration of waste products to the subsurface (see Figure 14. Among these are the following: landfills and dumps, evaporation ponds or lagoons, septic systems (addressed earlier), coal and mineral tailings piles (see NRC, 1981a, 1981b), and deep burial and deep injection. Each ofthese disposal methodologies presents technical and hydrogeologic problems and the potential for groundwater con- tamination. Landfills Landfills are probably the most widely used means of solid-waste disposal. Several chapters in this report (Chapters 4, 7, 8, and 10) specifically discuss landfills and some of their associated groundwater problems. The problems are often related to the hy- drogeologic setting of a specific landfill. In our efforts to restore contaminated aquifers, we need to consider this hydrogeologic setting carefully. The potential for groundwater contamination from landfills was recognized several decades ago. One of the more common measures taken to avoid groundwater contam- ination is to locate landfills in areas composed of rock or soil of low permeability so that water does not percolate through the landfill into the underlying aquifer. However, recent studies of landfills situated in low-permeability rocks indicate that in humid areas many of the landfill trenches in which the wastes are buried become filled with water from rain and snow. This surface water seeps downward through the landfill cover, fills the trenches, and eventually overflowsthe so-called "bathtub" effect. In these in- 8

OCR for page 3
Overview and Recommendations stances the contamination is not a groundwater problem, which would result from water in the trenches seeping downward to underlying aquifers, but rather a surface-water problem. One approach taken to avoid the surface-water problem has been to cover the trenches with a low-permeability material. However, the wastes when placed in the landfill are often poorly compacted and decay with time; the wastes in the trench become naturally compacted and can result in breaching the cover, which allows the trench to become saturated, eventually overflowing the bathtub. This effect is difficult to overcome. A possible solution to the bathtub effect is to design the landfill with a controlled leak. This requires draining the landfill from below either naturally with an underlying aquifer or with installed drains. In the case of an underlying aquifer, the groundwater flow would need to be large enough to dilute the contaminants below toxic levels. Alternatively, the drainage from the landfill can be collected and processed through water-treatment plants. Careful hydrologic and engineering analyses are needed to design systems that will prevent the occurrence of unacceptable levels of both ground- water and surface-water contamination. Evaporation Ponds and Lagoons Evaporation ponds and lagoons are used widely for disposal of wastes dissolves! in water. The technique allows the water to evaporate, leaving behind a concentrated solid residue, which can be disposed of more readily than the liquid wastes. However, such poncis and lagoons often leak and contaminate the underlying groundwater. Attempts to rectify the problem include lining the pond to reduce the leakage. But it is difficult to add an impermeable lining that will function for prolonged periods (usually a decade or longer). Few, if any, economically feasible materials currently available can resist corrosion from a broad range of chemicals and still maintain physical strength great enough to resist mechanical failure. A number of substances such as plastic linings and layers of low-permeability clays have been used with some success. As discussed in Chapter 6, evaporation ponds have been used extensively at the Rocky Mountain Arsenal for waste disposal. Initially these ponds were unlined and resulted in widespread con- tamination of the groundwater. However, even a lined pond at the Rocky Mountain Arsenal was found to have leaked after more than 20 years of use. Deep Injection or Burial Deep burial of toxic wastes has several advantages over shallow burial or surface-storage systems. The most important advantage is that contaminants that may become dissolved in groundwater will not migrate directly to the land surface. The increaser! length of the groundwater flow path allows time for decomposition of unstable chemical com- pouncis, the decay of radionuclides, and the dilution of toxic materials by dispersion. Deep burial also affords protection against the possibility that hazardous materials will be exposed at the surface through slow processes of erosion. The depth of burial is generally related to rock permeability; the deeper the burial, the lower the permeability. On the basis of hydrogeologic criteria alone, many different rock types would seemingly provide safe repositories at depths greater than 100 m. Deep repositories in most locations will eventually fill with water. However, if zones of significant groundwater circulation are avoided, repositories at depths of more than 300 m in granitic rocks would be likely to take several hundred years to fill with water once they are closed. In a well-placed repository, several thousand years may be needed for a simple piston-flow displacement of all the water in the flooded repository (see Chapter 5~. 9

OCR for page 3
Overview and Recommendations Although deep injection or burial is a suitable method for the disposal of toxic wastes, a number of site-specific scientific and engineering problems need to be considered. For example, a method proposed for the disposal of high-level radioactive wastes is a deep repository (approximate 1 km) in crystalline rocks (see NRC, 1983~. Unfortunately, most crystalline rock masses are fractured to some extent. This poses problems for hydrologists because flow and transport in fractured rocks are not well understood. While this is an area of active research, currently there is no consensus within the hydrogeologic community on the theory of how to treat flow and transport in fractured rocks. As a result, predictions of groundwater flow and transport for several thousand years in fractured rocks are difficult to provide with a high degree of confidence. Each borehole or shaft to a repository or disposal horizon represents a short circuit along which wastes could flow back to shallow levels and interact with the biosphere. The engineering problems in effectively plugging such holes with seals that will isolate the repository for periods of several thousand years are not simple. While deep disposal is undoubtedly one of the preferred methods of waste disposal for highly toxic wastes, its most serious disadvantage is its cost. Deep repositories will be expensive to construct and difficult to monitor; and if errors are maple or unexpected flaws are uncovered, they will be costly to correct (e. g., removing the waste and placing it in a better location or a better designed repository). Multiple Barriers For deep repositories the approach currently favored from the vantage point of risk analysis by both the military and the U. S. Department of Energy involves the concept of multiple barriers. A number of engineered barriers would include: (1) waste forms that are not readily soluble, (2) canisters that would isolate the wastes for long periods, and (3) backfilling the repositories with materials that are highly sorptive and of low permeability. The engineered barriers may then be coupled with a number of natural barriers including: (1) storage in low-permeability media; (2) siting in areas where the natural groundwater flow moves contaminants away from the biosphere, at least in the immediate vicinity of the repository, if not regionally; and (3) siting in rocks containing naturally occurring minerals (e.g., zeolites) that tend to sorb the contaminants or in highly porous rocks (e.g., luffs) in which diffusion into the matrix can be a retardant during fracture flow. At Oak Ridge National Laboratory, in a use of the concept of multiple barriers, radioactive wastes have been mixed with cement and implaced by hydraulic fracturing techniques in low-permeability shale. In the arid and semiarid portions of the western United States, thick unsaturated zones often overlie a deep water table. Studies of waste disposal, especially in the more arid areas of Nevada, suggest that there is limited or no transport of wastes through the ground in this environment. The natural movement of moisture through the un- saturated zone appears, at least from preliminary scattered observations, to be too small to transport significant amounts of wastes. Numerous closed hydrologic basins exist within Nevada, Utah, and parts of adjacent states. These basins have thick unsaturated zones in some places up to 600 m thick and internal drainage, i.e., there is no discharge to the sea. Such closed basins in arid areas have an obvious advantage for waste-disposal purposes; even if the contaminants move, they will remain within the basin. For long-term containment of wastes, climatic changes, such as were associated with the advances of Pleistocene ice sheets, should be considered in the prediction of waste movement. Sufficient work has been done at the Nevada Test Site to indicate that the water table did not change significantly during the more recent Pleistocene ice advances. The finding suggests that areas may be found where possible changes in climate will not significantly affect the rates of groundwater transport. Furthermore, the sorptive 10

OCR for page 3
Overview and Recommendations properties of most alluvium constitute a natural barrier to contaminant migration even if significant changes in groundwater flow were to occur (see Chapter 114. EFFECTIVE DISPOSAL SITES To some extent our past practice of waste disposal can be characterizes] as a strategy of "out of sight, out of mind." A number of disposal sites have proven to be effective in containment and reasonably safe. In some instances, careful thought and planning in site selection and disposal practices have led to safe containment of highly toxic wastes. However, siting must be done on a case-by-case basis (see, e.g., Chapter 4) and must depend on the nature of the wastes. Unfortunately for scientific purposes, not many potential groundwater contamination problems have a history of investigation and a data base to document the situation. Some of the best-studied examples of field! sites with the potential for groundwater contamination involve major nuclear test facilities. Investigations at some of these sites date back to the early 1950s, when studies were initiated by the Atomic Energy Com- mission. Geologists and hydrologists were consulted to help un~lerstand and monitor the potential migration of nuclear wastes. At several of these facilities, wastes have been implaced within the ground with a minimum of problems. For example, most radionuclides from an underground nuclear explosion at the Ne- vada Test Site have experienced little or no migration in the grounc~water from the region of release. This is because the Nevada Test Site, which is located in the Basin and Range province in an area of internal drainage, has appropriate geologic and hy- drologic features and little precipitation, factors that minimize the migration of the radionuclides. The rocks, primarily volcanic luffs and derived tu~aceous alluvium, con- tain large quantities of zeolites, which act as a natural exchange medium and tend to adsorb many of the contaminants, as suggested in Chapter 11. The high connected porosity of the luffs further retards migration by diffusive processes. This characteristic is of particular importance in retarding the migration of nonsorbing species through fracture flow. In addition, if wastes were to enter the natural groundwater flow system, they would be transporter] over long horizontal (listances for long periods of time, which would allow for radioactive decay of many of the more toxic radionuclides before they could be transported back to the near surface and the biosphere. Another example of a site with appropriate hydrogeologic features resulting in effec- tive containment is the Idaho National Engineering Laboratory (INEL, formerly the U.S. National Reactor Test Station, NETS), which also involves the disposal of radio- active wastes. At INEL various wastes, including tritium (3H), strontium-90 (90Sr), and cesium-137 (l37Cs), have been discharged into the underlying Snake River Basalt Aqui- fer. Tritium has been transported conservatively (i. e., without chemical reactions other than radioactive decay); the dispersion is such that tritium concentrations are below hazardous levels once the contaminants have been transporter] several kilometers down- gradient from the point of injection. Both 90Sr and 137CS are adsorbed and remain tied to the aquifer skeleton within approximately 2 km of the injection well. None of the wastes poses significant hazards except in the immediate vicinity of the disposal site. Other such suitable disposal sites can be found in a variety of environments through careful and innovative exploration and research. However, the necessary site-specific geologic and hydrologic data are frequently inadequate for a reliable evaluation to be made. As the Nevada Test Site illustrates, the subsurface can provide a natural migration barrier. If done with care, toxic wastes can be isolated from the biosphere for periods so long that they can be measured in terms of geologic time. The criteria are relatively simple: the total waste system containers and repository as well as geologic, geo- chemical, and hydrologic setting should isolate the wastes from the biosphere for a 11

OCR for page 3
Overview and Recommendations very long period. In consideration of organic wastes, acIditional criteria, such as organic reactions in and the microbiology of grounc~water, need to be considered in the whole system. GROUNDWATER HYDROLOGY Conservative Transport Elective utilization of the subsurface as a repository for wastes depends on information as to how the wastes are transported. Groundwater is the transporting agent, and it is generally agreed that we know a great deal about the flow of liquids through porous media. (In this case we define a porous medium as a typical aquifer or reservoir ma- terial sand, gravel, sandstone.) However, the problem of contaminant transport is somewhat more complex than the problem of flow. With conservative contaminant transport (transport without chemical reactions) a physical mixing occurs that is generally referred to as hydrodynamic dispersion. Dis- persion is caused by both microscopic and macroscopic variations in the fluid velocity. The magnitude of dispersion is much larger under natural conditions in the field than in the laboratory commonly 1000 to lO,OOO times larger. The larger dispersion ob- served in the field is caused by the large-scale variations in fluid] velocity produced by geologic deposits that commonly have a continuous spectrum of microscale to large regional-scale variations in permeability (the permeability of natural materials varies over 15 orders of magnitucle). The matter of how best to characterize hydrodynamic dispersion is currently the subject of scientific debate; the status of our knowledge of the physical process is reviewed in Chapter 2. As either tracers or contaminants are transported away from a source, the magnitude of the dispersion increases. This is reasonable because the large-scale mixing depends on the heterogeneity characterizing the geology of the site. As the contaminant moves it becomes more affected by the geologic heterogeneity, resulting in increased disper- sion. Thus it becomes difficult to determine the dispersivity, the parameter most widely utilized to characterize dispersion, and tracer tests tend to become impractical. How- ever, a number of field sites where contaminants have moved have now been studied. Investigations suggest that the dispersivity generally ranges from 10 to 100 m (the larger the dispersivity, the larger the (lispersive mixing) once the contaminants have migrated several hundred meters or more. Perhaps the most difficult and as yet unsolved problems in groundwater flow and transport are associated with fractured rock. Most crystalline rocks have been found to be fractured to great depth. Recent interest in utilizing crystalline rocks for waste repositories, particularly nuclear repositories, has focused on flow and transport through this type of rock. Two schools of thought on the flow in fractured rock have evolved. One treats the flow in individual fractures and then aggregates these to form a flow system. The second deals with the fractured rock mass statistically. The difficulties in describing individual fractures, especially in the subsurface, have tended to persuade a growing number of investigators to treat the large-scale flow and transport problem in a statistical fashion. To understand and make reliable predictions of contaminant movement, it is necessary to know more about the relationships among primary geologic features, such as how porosity and permeability are related to geologic environments of deposition and how the number of fractures and their orientation and spacing are related to tectonic stresses. Not enough is known to allow quantitative predictions of the occurrence and nature of fractures in rocks. 12

OCR for page 3
Overvieu) and Recommendations Transport with Reactions Both inorganic and organic chemical reactions may occur during the transport process. The analyses of the problem of the reactions and their effect on transport are currently approached from two different directions: (1) studies of overall chemical equilibria and (2) transport analyses involving kinetics (in effect an irreversible thermodynamic ap- proach), as pointed out in Chapter 3. In the equilibrium approach one assumes that the reactions are sufficiently fast that chemical equilibrium generally is quickly achieved. In this approach the details of the transport process are not taken into account specifically. One need only apply the constraints of classical equilibrium geochemistry to understand the chemistry of the system. Knowing what chemical reactions to consider is not a trivial problem; it is of paramount importance. Much of the current work on chemical contamination has been approached from the framework of equilibrium geochemistry. The transport approach relaxes the equilibrium assumptions. On the other hand, the mathematics is much more complex, and one must have some understanding of the kinetics of the reactions of concern. The geochemistry of inorganic chemical constituents is better understood than that of organic constituents. Fortunately, under many conditions existing in nature, many of the inorganic contaminants are not particularly mobile. In the words of one member of the panel: "Mother Nature has been generally kind to us by impeding the transport of inorganic contaminants." ORGANIC COMPOUNDS Organic contaminants in groundwater are a major problem. The use of synthetic organic chemicals has grown at a phenomenal rate during this century; the growth has been especially rapid since World War II. A variety of organic compounds has gained wide- spread use in industry; agricultural pesticides and herbicides have been utilizer] exten- sively in recent years. For many of the organic compounds now in use, little is known about their toxicity. Some are believed to be hazardous at very low concentrations tens or hundreds of parts per billion. In the past groundwater was not routinely tested for low levels of organic compounds; however, with the recognition that low-level organic contamination is a potential health hazard and with advances in the technology of analytical chemistry, increased instances of organic contamination have been identified. The chemical processes involved with organic contaminant transport are reviewed in detail in Chapter 3. As examples, Chapter 6, dealing with problems at the Rocky Mountain Arsenal, and Chapter 10, dealing with a landfill in Delaware, highlight some of the problems with organic contaminants. Organic solvents, which are volatile in the atmosphere, move into unsaturated soils, often as a result of spills, and remain there for long periods commonly tens of years. In an unsaturated soil, equilibrium is established between the liquid and the vapor phases of the contaminants. Each succeeding rain upsets the equilibrium and in the process carries some of the contaminant to the water table. Often the organic solvent remains as a source of groundwater contamination for Tong periods. Because some of these solvents are toxic at the parts per billion level, they pose particularly hazardous problems. Recent research has established that many, if not most, of the chemical reactions involving organic compounds, in both the saturated and unsaturated zones, are con- trolled by microorganisms. Contrary to earlier studies, recent investigations have es- tablished high levels of microorganisms both above and below the water table; Wilson en cl McNabb (1983) summarized several sets of recent data, which are given in Table 13

OCR for page 3
Overview and Recommendations 2. The amount of biomass identified is much larger that that which normally occurs in rivers and streams. The organisms identified are almost exclusively some form of bac- teria. The relationship between the concentration of the pollutant and its susceptibility for biotransformation is complex. Wilson and McNabb (1983) stated: . . . compounds that usually are considered degradable may not be transformed by the subsurface micro- organisms if the compound is present at low concentrations. Similarly, compounds present at high concen- tration may only be partially degraded when oxygen is entirely depleted and can only be degraded further after dispersion or other physical processes mix the contaminated water with oxygenated water. [Table 3, this volume] presents the authors' opinions concerning the prospects for biotransformation of several important classes of organic pollutants in groundwater. These predictions are based on a cautious extrapolation from the behavior of these compounds in other natural systems and on our admittedly limited experience with their behavior in the subsurface.... The control exerted by microorganisms on the chemistry of various organic contam- inants is a particularly important area for increased future research. We have only begun to appreciate fully the importance of bacteria in controlling the chemical changes taking place. Only a limited number of situations and organic compounds has been investigated. With further study, it may be possible to control the degradation of particular contam- inants in groun~lwater through the in situ introduction of certain cultures of microor- ganisms. Many of the organic residuals are sufficiently stable in nature to remain intact in- clefinitely during groundwater transport. Future methods of coping with organic resicI- uals may involve incinerating them at sufficiently high temperatures in order to break them down chemically. (Some toxic organics may not be totally broken down to nontoxic components using current incineration practices.) It may well be that the toxic organic compounds that cannot be incinerated or treated to make them less hazardous on a practical basis will have to be disposed of by deep burial, a methocl now proposed for radioactive wastes. MATHEMATICAL ANALYSIS One method of studying the transport of contaminants in groundwater is by mathe- matically simulating a field situation in which a well-clefined contaminant plume has moved through time. This requires a historical set of data that defines the plume and documents the movement through time. Only a limited number of field sites exists TABLE 2 Number (millions per gram of dry material) of Organisms in the Subsurface Environmenta Depth to Water Just above Just below Site Table (m) Subsoil Water Table Water Table Lula, Okla. 3.6 February 1981 6.8 3.4 6.8 June 1981 9.8 3.7 3.4 Fort Polk, La. Borehole 6B 6.0 3.4 1.3 3.0 Borehole 7 5.0 7.0 1.3 9.8 Conroe, Tex. 6.0 0.5 0.3 0.6 Long Island, N.Y. 6.0 36 3.0 170 Pickett, Okla. 5.0 5.2 From Wilson and McNabb (1983) with permission of the American Geophysical Union. 14

OCR for page 3
Overview and Recommendations TABLE 3 Prospect of Biotransformation of Selected Organic Pollutants in Water-Table Aquifersa Aerobic Water, Concentration of Pollutant (1lg/L) Anaeroblc Class of Compounds 100 10 Water Halogenated aliphatic hydrocarbons Trichloroethylene None None Possible'' Tetrachloroethylene None None Possible'' 1, 1, 1-Trichloroethane None None Possible'' Carbon tetrachloride None None Possible'' Chloroform None None Possible'' Methylene chloride Possible Improbable Possible 1, 2-Dichloroethane Possible Improbable Possible Brominated methanes Improbable Improbable Probable Chlorobenzenes Chlorobenzene Probable Possible None 1, 2-Dichlorobenzene Probable Possible None 1, 4-Dichlorobenzene Probable Possible None 1, 3-Dichlorobenzene Improbable Improbable None Alkylbenzenes Benzene Probable Possible None Toluene Probable Possible None Dimethylbenzenes Probable Possible None Styrene Probable Possible None Phenol and alykl phenols Probable Probable ProbableC Chlorophenols Probable Possible Possible Aliphatic hydrocarbons Probable Possible None Polynuclear aromatic hydrocarbons Two and three rings Possible Possible None Four or more rings Improbable Improbable None aFrom Wilson and McNabb (1983) with permission of the American Geophysical Union. bPossible but probably incomplete. CProbable but at high concentration. where enough historical data have been collected to define clearly the movement of a contaminant plume. The Idaho National Engineering Laboratory (INEL) is a well- studiecl example where there are three major classes of contaminant transport: con- servative (i.e., chloride), first-order radioactive decay (i.e., tritium), and sorptive (i.e., strontium). Other case studies describing specific studies of contaminant transport and conditions are described in Chapters 6-12. One result will, perhaps, indicate the capability of the current mathematical models. Using values of dispersivity determined from examining the distribution of chIorides in the waste plume at INEL, the movement of 3H was simulated. Figure 2 is a com- parison of the tritium plume as mapped from field data to the model-generated plume. In general, the match is quite goocI. Recently Lewis and Goldstein (1982) compared the distribution of contaminants at INEL in October 1982 with that predictecl by Robertson ancl Barraclough (1973~. Lewis and Goldstein concluded: "The model simulations made in the early 1970s of the waste chloride and tritium plumes were approximately correct, but differed in detail, as expected, because of several conservative assumptions that tended towar(l 'worst case' situations. " ~ . Our general experience with mathematical model analysis is such that we can state several points with some confidence: 1. Usually the analysis of flow and transport gives reasonably good results for con- taminants in which no chemical reactions (other than radioactive decay) occur. This provides some confidence in the general theory of both flow and transport. 15

OCR for page 3
Overview and Recommendations GROUND MILES O 1 2 81 I T I 1- . ~ o 1 2 3 KILOMETERS / / / DISPOSAL i/: ~ /, ! ~ J\ ,/- ~ ~ Bow ~ ~ SPOSAL WE ~ L ~,''1 ~ / if \ \? j l\ m? , ~ EXPLANATION EQUAL CHLORIDE CONCENTRATION IN mg/ ~ FOR 1968-69 50 WELL SAMPLES ~ 50 DIGITAL MODEL FIGURE 2 Comparison of waste chloride plumes in the Snake River Plain aquifer (at INEL) measured during 1968-1969 with that calculated using a mathematical model. 2. Simple chemical reactions such as sorption are also simulated in our analysis reasonably well. 3. Hydrodynamic dispersion as a mixing process is important. 4. The current state of the art is such that predictions of conservative contaminant movement are possible with a reasonable degree of confidence, especially for the short term periods of a decade or perhaps several decades. Mathematical analysis of a relatively complex transport process is well within our current capabilities, provided sufficient data are available for model calibration. This does not imply that we understand all phenomena and that further research is not necessary; the necessary research is discussed below. RESTORATION OF CONTAMINATED AQUIFERS The "restorability" of a contaminated area of an aquifer is highly dependent on its geologic, geochemical, and hydrologic properties and on the chemical and physical 16

OCR for page 3
Overview and Recommendations properties of the contaminant. In many cases, restoration is so expensive that cleanup is not considered to be economically feasible. Nevertheless, in response to public or governmental demands for positive action in cases where groundwater contamination threatens public health, aquifer cleanup programs are being required, such as at the Rocky Mountain Arsenal and more recently in "Silicon Valley" near San lose in California where organic solvents have been found in the groundwater. General management options currently available for restoring water quality in aquifers include the following: (1) eliminate the source of contamination but allow restoration to proceed only through natural flushing, dilution, and geochemical or biological re- actions; (2) accelerate removal of contaminants through withdrawal wells, drains, or trenches; (3) accelerate flushing with artificial recharge; (4) install "impermeable" bar- riers to contain a contaminated area; (5) induce in situ chemical or biological reactions that would neutralize or immobilize the contaminant; and (6) excavate and remove the contaminated part of the aquifer. The selection of the best approach for a particular situation requires the ability to predict changes in flow and chemical concentration in the aquifer for each possible management alternative. This in turn requires both ade- quate field data to describe the aquifer systems and the development of accurate sim- ulation models to define the groundwater flow system, the pollutant-transport mech- anism, and the nature and rates of chemical, physical, and biological reactions. The problem of cleaning up an aquifer is typified by the problems at the Rocky Mountain Arsenal near Denver, Colorado. Waste products from a chemical plant that produced chemical warfare agents and, later, insecticides leaked into a shallow aquifer. Over time these contaminants were transported toward the South Platte River by groundwater. The U. S. Department of Defense is now attempting to contain the escape of further contaminants from the Arsenal. A cleanup technique has been devised in which contaminated groundwater is pumped frown the aquifer upgradient from an en- gineered groundwater barrier; the water is treated to reduce the concentration of the contaminants to safe levels; ant] the treated water is recharged to the aquifer down- graclient from the barrier. The efforts at the Arsenal are described in Chapter 6, and the associated costs are described by the U.S. Army (1982~. In almost all situations, prevention of groundwater contamination is clearly much cheaper than restoration. The cost of cleanup may be so great that society may simply have to designate certain aquifers as permanently contaminated and unsuitable for further use other than for waste disposal. Virtually all the grounc~water in the Earth's crust will cycle back to the biosphere; however, the time for the cycle to be completer] varies enormouslyfrom days to tens of thousands of years or even longer. In some instances, its movement is so slow that, once contaminated, groundwater will remain so more or less permanently. CONCLUSIONS AND RECOMMENDATIONS 1. Groundwater, which is ubiquitous in the crust of the Earth, serves not only as a widely distributed source of water but as a host and transporting agent for contaminants. Scientific understanding of the chemistry and transport of contaminants in groundwater is inadequate to predict the fate of contaminants reliably. Research is neecled on the effects of chemical reactions on transport and dispersion of contaminants by groundwater and the quantification of flow in fractured media. Continued research can reduce the uncertainties associated with predicting the fate of contaminants in the subsurface. The results of such research should allow for safer disposal methods, at reduced costs. A number of the more important areas for continued scientific research are listed below. 17

OCR for page 3
Overview and Recommendations Continued investigation is necessary of the geochemistry of contaminant reactions in both unsaturated and saturated subsurface environments. Certain of the organic compounds appear to be especially persistent in groundwater and hazardous at low concentrations. Many, if not most, organic reactions are biologically controlled; at pre- sent the role of microorganisms is poorly understood. Only a limited number of situations and organic compounds has been investigated. Certain biological reactions might be enhanced and thereby accelerate biodegradation of contaminants. Increased understanding of the dispersion process in groundwater transport is important in predicting contaminant transport. Further research is needed into coupling chemical reactions into the transport formulation of the processes. In order for this formulation to be used, the rates of reactions must be known better than at present. Our understanding of groundwater flow and transport in fractured rock is ele- mentary. There is still considerable discussion within the scientific community of the appropriate theory. Given an adequate theory, it may be possible to develop methods for measuring the appropriate parameters in the field that characterize flow and transport in fractured rocks. However, field characterization of fractured rocks is certain to be costly. _ . . . . . . 2. Even though scientific information is incomplete on the fate of contaminants in grounc3water, the panel concludes that most wastes can safely be disposed of in the subsurface if repositories are selected, designed, and engineered on the basis of the nature of the wastes and adequate knowledge of the hyclrology, geology, anal hydro- geochemistry of the particular site. There s1?ouic! be a more thorough searchfor disposal sites that can be used safely to isolate toxic wastes from the biospherefor long periods. If a policy is adopted to establish disposal sites in the vicinity of the facility generating the wastes, sites must be located that provide the necessary multiple barriers to con- taminant migration. For example, the bathtub effect continues to cause troublesome problems at landfills. The solution may have to include precompaction of wastes, the construction of a good seal over the landfill, and possibly the incorporation of a controlled leak. This will involve geologic, hydrologic, and geochemical research as well as research into engineering of the disposal site and the waste form. Deep geologic disposal is now being considered for high-level nuclear wastes. It may also have to be considered for toxic, not readily treatable, wastes. A continued search for deep waste repositories throughout the country is needed. 3. If adequate scientific understanding of the fate of contaminants in groundwater existed and suitable disposal sites and methods were identified, certain aspects of the groundwater contamination problem from the disposal of wastes would still persist. The hundreds of millions of tons of wastes generated each year could overwhelm our ability to find suitable sites. In addition, the wide variety of wastes generated from individual municipalities and industries has often, for convenience, been disposed of in the same repository; this practice complicates the prediction of the fate of these mixed wastes and the design and engineering of the repository system. Therefore, we recommend that Q strategy be developed that provides for the segre- gation, treatment, anct disposal of wastes according to their hazards and their chemical affinities. In some instances recycling residual by-products may be the most cost- effective means of treating potential wastes. Various contaminants are hazardous at different levels of concentration and behave quite differently when in the ground. Some are persistent anal quite mobile; others are readily degraded and easily adsorbed. Depending on their persistence and mobility in a particular environment, cli~erent wastes should be treated and disposed of differently. Wastes can be separated into at least three different classes, each class to be disposed of in a different manner: 18

OCR for page 3
Overview and Recommendations I. Relatively immobile or nontoxic wastes that can be safely disposed of in the ground. II. Wastes that can be chemically or biologically treatable or recycled. For example, some of the more toxic organic wastes can be incinerated at temperatures high enough to destroy the compounds. III. Particularly toxic wastes that are not readily treatable, such as some of the high- level radioactive wastes, or are persistent and mobile in groundwater. The strategy for waste disposal should require different treatment ant] disposal for each class of waste. For those that are relatively immobile and degrade readily (Class I above), conventional burial in a well-designed landfill is an appropriate disposal method. Wastes that are not readily treatable and are highly toxic, persistent, and mobile in grounc~water (Class III wastes) would need special disposal measures (e.g., multiple- barrier schemes) such as those now being proposed for high-level nuclear wastes. The by-products of treatable wastes (Class II wastes) coulc! conceivably be persistent and highly toxic and require special consideration. Treatment such as high-temperature incineration of certain toxic organics could greatly reduce the volume of wastes to be disposed of in the ground. Such a strategy would certainly involve additional costs in segregating wastes. How- ever, the cost of cleaning up a poorly designed system of disposal is usually much larger than the cost of carefully designed clisposal. 4. The public must understand that use of the products of technology carries with it the responsibility for safe disposal of that technology's wastes. A pervasive difficulty with waste disposal is that indivicluals do not want the wastes, especially toxic wastes, disposed of in their neighborhood. However, this perception has been an important consideration in providing effective waste management. In recognition of the need to dispose of vast quantities of waste, governmental and industrial organizations need to agree whether various classes of wastes should be disposed of locally or in regionally designated repositories. Wastes, especially those produced in large volumes, will probably have to be disposed of reasonably close to the location where they are generated. The quantities of many municipal and industrial wastes are such that transportation over long distances is costly. Furthermore, in the case of highly toxic wastes, transportation by rail or roadway involves risks that cannot be ignored. The nation may need to move toward a policy of requiring states to dispose of wastes within the state in which they are generated or within regions determined by interstate compacts. In Chapter 13 it is pointed out that there are efficiencies to be gained from assigning the disposal decisions to the lowest administrative level at which appropriate and reasonable decisions can be made. In addition, in Chapter 14 it is argued that institutions must be established to examine trade-offs and to determine acceptable waste-disposal solutions that will minimize risks to society. REFERENCES Lewis, B. D., and F. J. Goldstein (1982). Evaluation of a Predictive Ground-Water Solute-Transport Model of the Idaho National Engineering Laboratory, Idaho, U.S. Geol. Surv. Water-Resour. Invest. 82-25, 71 PP NRC (1981a). Disposal of Excess Spoils from Coal Mining and the Surface Mining Control and Reclamation Act of 1977, Board on Mineral and Energy Resources, National Research Council, National Academy Press, Washington, D.C., 207 pp. NRC (1981b). Coal Mining and Ground-Water Resources, Board on Mineral and Energy Resources, National Research Council, National Academy Press, Washington, D.C., 197 pp. NRC (1983). A Study of the Isolation System for Geologic Disposal of Radioactive Wastes, Board on Radioactive Waste Management, National Research Council, National Academy Press, Washington, D.C., 345 pp. 19

OCR for page 3
Overview and Recommendations Robertson, J. B., and J. T. Barraclough (1973). Radioactive and chemical waste transport in groundwater at the National Reactor Testing Station, Idaho: 20-year case history, in Proceedings of the International Symposium on Underground Waste Management and Artificial Recharge, Am. Assoc. of Petrol. Geologists, Tulsa, Okla., pp. 291-322. U. S. Army (1982). Draft. Selection of a Contamination Control Strategy for RMA, Rocky Mountain Arsenal Contaminant Control Program Management Team, submitted to the State of Colorado and the U.S. Environmental Protection Agency. U.S. Environmental Protection Agency (1980). Groundwater Protection, Washington, D.C., 36 pp. U.S. Water Resources Council (1980). U.S. Water Resources Council Bulletin 16, Washington, D.C. Wilson, J. T., and J. F. McNabb (1983). Biological transformation of organic pollutants in groundwater, EOS: Trans. Am. Geo phys. Union 64, 505. 20