17 |
A Framework for Ecological Risk Assessment |
| Lawrence W. Barnthouse | |
| Oak Ridge National Laboratory, USA |
In 1983, the US National Research Council (NRC) defined a set of general principles for human health risk assessment that are now widely used as the basis for managing risks of chemicals to human health (National Research Council, 1983). The NRC's report described risk assessment as a procedure for linking scientific information about potentially hazardous substances to the decision-making process through which human exposures to these substances are regulated. The report attempted to provide a clear distinction between the roles of scientists (risk assessors) and decision makers (risk managers) in the assessment process, and to define a general procedure to identify the kinds of scientific information appropriate for quantifying human health risks of toxic chemical exposures. Although most of the report deals only with chemicals, the committee clearly intended for its major conclusions to apply to non-chemical risks as well.
Within the past several years, interest has grown in defining an analogous set of principles to assess ecological risks of toxic chemicals and other stresses. Despite the wide disparities between different kinds of ecological assessments (e.g., between the regulation of pollutant discharges and the management of fisheries), a unifying set of principles for ecological risk assessment is now emerging (USEPA, 1992; Suter, 1993a).
In this chapter, a framework for ecological risk assessment is drawn from several recent sources, and its implications for the future of chemical risk management are discussed. In addition, risk assessment will be related to the new paradigm for global environmental management implied by the term "sustainable development."
The NRC (1983, p. 18) defined human health risk assessment as "...the characterization of the potential adverse health effects of human exposures to environmental hazards. The US Environmental Protection Agency (USEPA, 1992) recently proposed a similar definition of ecological risk assessment: "...a process that evaluates the likelihood that adverse ecological effects may occur or are occurring as a result of exposure to one or more stressors." Even more generally, ecological risk assessment can be defined by restating the 1983 NRC definition in ecological terms: the characterization of the adverse ecological effects of environmental exposures to hazards imposed by human activities.
The term "adverse ecological effects" includes all biological and non-biological environmental changes that society perceives as undesirable. The term "hazard" include both unintentional hazards such as pollution and soil erosion and deliberate management activities such as forestry and fishing which are often hazardous either to the managed resource itself or to other components of the environment.
The definitions of USEPA and the modified one of NRC emphasize that risk assessment is a decision-making process, not a computational technique. Because this perspective allows for extensive reliance on qualitative information and expert judgement, it is consistent with the current state-of-the-art in ecological science, which is much more qualitative than quantitative. More important, these definitions emphasize that the objective of risk assessment is not to provide scientific truth but to promote sound environmental decisions.
In addition to a definition, the NRC (1983) described a framework for human health risk assessment consisting of four components: hazard identification, dose-response assessment, exposure assessment, and risk characterization. The purposes of the framework were to (1) provide for more detailed definitions of the scientific components of risk assessments, (2) define the relationship of risk assessment to risk management, and (3) facilitate the development of uniform technical guidelines. An analogous framework for ecological risk assessment could be used to evaluate the consistency and adequacy of individual assessments, to compare assessments for related environmental problems, to identify explicitly the connections between risk assessment and risk management, and to identify environmental research topics and data needs common to many ecological risk assessment problems. The USEPA (1992) described ecological risk assessment as a three-part process consisting of problem formulation, analysis, and risk characterization. The term "analysis" was further subdivided into characterization of exposure and that of ecological effects.
A more general framework that integrates human health with ecological concerns is depicted in Figure 17.1. The relationship among the four components is a hybrid between the arrangements proposed by the NRC (1983) and USEPA (1992). This framework, and the definitions provided below, have now been formally proposed by the NRC's Committee on Risk Assessment Methodology (NRC, 1993).
Hazard identification may be broadly defined to be the determination of whether a particular hazardous agent is associated with health or ecological effects of sufficient importance to warrant further scientific study or immediate management action. The purpose of hazard identification is to determine whether a "hazard" exists; and, if one is identified, to determine the kinds of additional scientific information required to assess the degree of risk present and evaluate the alternative risk management actions. Typical kinds of information used for hazard identification include quantitative structure-activity relationships (QSARS), short-term toxicity tests, and reviews of existing information about the characteristics of potentially affected ecosystems.
Figure 17.1. A risk assessment framework intended to facilitate integration of human health and ecological risks of toxic chemicals and other environmental stresses (from NRC, 1983, and USEPA, 1992)
Exposure-response assessment may be defined as the determination of the relation between the magnitude of exposure and the probability of occurrence of the effects in question. In the case of toxic chemicals, information included in an exposure-response assessment could include detailed toxicological information (e.g., chronic toxicity, mode of action, sensitivities), mesocosm or field test data, field surveys to compare exposed and unexposed sites, and population or ecosystem modelling. The responses addressed in ecological risk assessments include both direct effects of exposure and the much broader indirect effects, such as secondary poisoning of raptors due to accumulation of pesticide residues.
Exposure assessment may be defined as the determination of the extent of exposure to the hazardous agent in question. As applied to toxic contaminants, this component includes measurement or prediction of the movement, fate, and partitioning of chemicals in the environment. The term can also be legitimately applied to non-chemical stresses, including both physical stresses (such as habitat change and UV radiation) and biological stresses (such as species introductions).
In actual assessments, exposure assessment and exposure-response assessment occur roughly in parallel, and must be closely linked. The arrangement of those components in Figure 17.1, within a single box divided in half by a "permeable membrane," is intended to emphasize the ties between them.
Risk characterization may be defined as the description of the nature and often the magnitude of risk, including attendant uncertainty, expressed in terms comprehensible to decision makers and the public. The purpose of risk characterization is to integrate information obtained from all of the other components and to communicate it to decision-makers in a form comprehensible to non-specialists and relevant to the decision being made. Contents should include (1) a description of the nature and often the magnitude of risk to ecological resources, and (2) qualitative and quantitative characterization of uncertainty. Because the purpose of risk assessment is to support decision-making, communication with decision- makers is a critical aspect of risk characterization.
In addition to the four basic components, Figure 17.1 depicts the relationship between risk assessment and environmental management. Policies and regulations determine the scope and content of risk assessments; results of risk assessments provide the scientific foundation for decisions. The risk assessment process should not, however, end when a regulatory decision is made. Follow-up in the form of monitoring (where measurable effects have been predicted), validation studies, and basic research are needed to improve the data and models available to technical risk assessors whenever the same or a similar problem is encountered in the future.
An additional concept relevant to this chapter is the distinction between assessment endpoints and measurement endpoints (Suter and Barnthouse, 1993). Assessment endpoints define the adverse effects to be avoided or the biological resources to be protected. Measurement endpoints are those variables measured in laboratory or field studies. Assessment endpoints are most often defined in terms of population or ecosystem properties such as risks of population extinction, reduction in productivity, or altered species composition. Assessment endpoints should be defined at the beginning of an assessment, as part of hazard identification, in order to guide the acquisition of exposure and effects data. Findings presented to decision-makers as part of risk characterization should be expressed in terms of assessment endpoints. Assessment endpoints are rarely directly measurable, because exposure to the hazardous agent has not yet occurred, because the scale and complexity of the system of interest (e.g., a bird population distributed over thousands of km2) precludes direct measurement, or because the precision of the available measurement techniques is insufficient to detect changes before they become irreversible.
Measurement endpoints may include toxicity tests, field surveys of contaminant distributions, or widely-spaced samples of population or ecosystem characteristics. Most of the technical literature on ecological effects of toxic chemicals deals with measurement rather than assessment endpoints. Over the last decade, great advances have been made in our understanding of modes of action of toxic chemicals, biochemical markers of toxicant exposure, the relative sensitivities of different kinds of organisms, and the changes in populations and ecosystems that can be caused by exposure to toxic chemicals. Such studies, however well performed, are usually insufficient as ecological risk assessments, because the scales of time and space that are amenable to rigorous scientific investigation are usually much shorter and smaller than the scales of interest in environmental management (Suter and Barnthouse, 1993). Additional extrapolation using expert judgement and mathematical models is necessary to relate these measurements to assessment endpoints that can be effectively communicated to decision-makers and that can support informed decisions.
To illustrate the application of the framework to ecological assessment problems, distinguishing between two major classes of risk assessments is useful: predictive assessments and retrospective assessments. The objective of a predictive assessment is to estimate the potential consequences of an action prior to taking that action. Assessments performed prior to the manufacture of toxic chemicals or the registration of pesticides are typically predictive assessments.
The objective of a retrospective assessment is to determine the causes and consequences of an event that has already occurred. Assessments of the need for environmental restoration of contaminated soil, water, and sediment are among the most common type of ecological risk assessment in the United States. Fisheries management and protection of endangered species also frequently involve retrospective risk assessments.
Predictive ecological risk assessments both in the United States and in the OECD are highly standardized, and are based principally on laboratory-derived toxicity test data and simplified environmental transport models. Hazard identification in these assessments consists of (1) documentation of basic physicochemical characteristics of the material being regulated (e.g., chemical structure, solubility, octanol-water partitioning coefficient), (2) summarization of available information on toxicity, and (3) description of likely environmental pathways, exposed species, and maximal exposures. The objective of the hazard identification step in these assessments is to quickly and efficiently classify chemicals as being clearly harmless, clearly hazardous, or potentially hazardous. Regulatory agencies such as the US Environmental Protection Agency, which must assess several thousand new chemicals every year, rely on standardized hazard identification protocols to facilitate approval of chemicals posing little or no risk (Bascietto et al. , 1990). The OECD (1984) has promulgated a similar scheme to test chemicals in the European community. Chemical manufacturers utilize similar protocols (e.g., Kimerle et al., 1983) to identify potentially hazardous chemicals during the development process, before major financial commitments are made.
If a potential hazard is identified, then more rigorous exposure and exposure-response studies may be performed. Aquatic exposure assessments would typically involve modelling of the transport and fate of the chemical in question in generalized streams, ponds, lakes, or estuaries. Many models of the transport and fate of chemicals have been developed and several excellent reviews have been published (Jørgensen, 1984; Cohen, 1986; OECD, 1989; MacKay and Patterson, 1993). Exposure assessment models for terrestrial ecosystems are much less developed. Most often, field tests are used to provide direct measurements of the persistence of chemicals in soil and uptake by vegetation.
Many methods are now available to characterize exposure-response relationships for use in predictive ecological risk assessments. Kendall (1992) recently summarized the approaches used to characterize effects of pesticides on birds. These range from biochemical studies of the relationships between doses and enzyme activities (e.g., cholinesterase) to field tests in which birds are allowed to forage on fields to which pesticides have been applied using realistic application regimes. The USEPA uses a standardized tiered testing scheme in which increasingly realistic tests are performed at different levels (Urban and Cook, 1986).
Most existing predictive ecological risk assessment protocols characterize risks by comparing estimated exposure concentrations to a test-derived effects criterion such as a "no-observed-effects level," or a "lowest-observed-effects level." Depending on the kind of test performed, the criterion may include factors of 10 to 100 uncertainty or safety factors designed to ensure that risks are not underestimated. This procedure is acknowledged as being overly simplistic (Bascietto et al., 1990). Methods to replace the uncertainty factors with empirical estimates of uncertainties inherent in extrapolating test data between species and test types have been developed (Suter et al., 1983; Sloof et al., 1986; Suter et al., 1987; Kooijman, 1987; Volmer et al., 1988; van Straalen and Denneman, 1989; Barnthouse et al., 1990) but not widely implemented. Empirical and theoretical models have also been used to characterize ecological risks in terms of effects on exposed populations and ecosystems (Bartell et al., 1992; Barnthouse et al., 1990; Barnthouse, 1993), but these methods are not now used in regulatory practice.
Retrospective ecological risk assessments are more diverse and more difficult to characterize than are predictive assessments. Suter (1993b) distinguished three types of retrospective assessments: source-driven, exposure-driven, and effects-driven. A source-driven assessment is initiated by the observation of a source of environmental contamination such as an oil spill or a discharge pipe. An exposure-driven assessment is initiated by the observation of contaminated environmental media or biota. An effects-driven assessment is initiated by the observation of adverse changes in ecosystems, populations, or individual organisms. The two common elements in all of these assessments are (1) the possibility (at least in principle) of basing the assessment on actual field data and (2) the diagnosis of the sources of observed contamination or the causes of observed effects usually a major goal.
Assessments of the ecological risks of tributyl tin (Huggett et al., 1992) provide an excellent example of retrospective ecological risk assessment. Tributyltin (TBT) is a chemical biocide often used as an antifouling agent in boat paints (Blunden and Chapman, 1982). Concentrations as low as 1 mg/litre are lethal to larvae of some species, and non-lethal effects have been observed at concentrations as low as 0.002 mg/litre (Huggett et al., 1992). Adverse changes in a variety of marine invertebrate species, including commercially valuable shellfish, were observed in Europe in the early 1980s (Alzieu, 1986; Abel et al., 1986). The changes observed included both declines in abundance of populations and morphological changes in the surviving organisms. Laboratory experiments with the oyster Crassostrea gigas showed that concentrations of TBT similar to concentrations found in coastal waters following the introduction of TBT-based antifouling paints caused the same patterns of abnormal shell growth observed in wild oyster populations. Snails in the vicinity of a marina on the York River, Virginia (US), were shown to have an abnormally high incidence of imposex (expression of male characteristics by female organisms), an effect previously observed under laboratory conditions in females of the European oyster Ostrea edulis (Huggett et al., 1992). Regulatory agencies in Europe and the US have concluded that the diagnostic link between TBT exposures and adverse effects on shellfish are sufficient to demonstrate a significant ecological risk. Although the US Environmental Protection Agency has not issued regulations for TBT, the US Congress and several states have restricted the types of boats to which TBT-based paints may be applied and the leaching characteristics of the paints that may be used.
TBT is somewhat unusual in that a single chemical was responsible for observed adverse ecological changes and that easily diagnosed morphological changes could be associated with TBT exposures. Suter and Loar (1992) described a much more common situation: assessment of ecological risks associated with a complex mixture of contaminants derived from a large federal facility containing multiple contaminant sources. The objectives of this assessment are to determine whether remedial actions are needed to reduce exposures of organisms to contaminants derived from the US Department of Energy Oak Ridge Reservation, to identify the specific chemicals and sources requiring remediation, and to monitor recovery of on- and off-site ecosystems during the remediation process (expected to last for several decades).
A combination of field and laboratory studies have been employed. The field studies include characterization of contaminant distributions in water, sediment, and biota below known point sources, in the streams draining the reservation, and in the Clinch River adjacent to the Oak Ridge Reservation. Several biological responses are also being measured. Population and community (fish and invertebrates) characteristics are being monitored and correlated with exposure concentrations. Physiological and biochemical responses (i.e., biomarkers) known to be linked to contaminant exposures are also being measured (Shugart et al., 1992).
Ultimately, these diverse types of data will be used to characterize the ecological risks of the current levels of environmental contamination on the Oak Ridge Reservation and to evaluate the relative ecological costs and benefits of potential restoration alternatives.
No discussion of environmental risk assessment written presently can be considered complete without a consideration of how the concept of risk assessment relates to the concept of environmental sustainability. The World Commission on Environment and Development (1987) defined sustainable development as the form of development or progress that "... meets the needs of the present without compromising the ability of future generations to meet their own needs." At the "Earth Summit" in Rio De Janeiro, delegates from 178 nations signed a declaration embracing sustainable development as an environmental management goal. The concept of sustainable development implies that the earth's ecosystems should be managed in such a way that they persist in more or less their present form, and continue to provide material sustenance, services, and aesthetic enjoyment to human societies. This goal has profound implications for environmental management and risk assessment. In sustainability-based environmental management, the focus is on maintaining or improving the quality of the environment, not on restricting discharges or requiring particular waste treatment technologies. Because the environment rather than the chemical or the technology is the focus of management, the management of chemicals must be integrated with management of other stresses on ecosystems. Decisions about remedial actions to restore ecosystems affected by chemical contamination have to account for other existing stresses such as erosion, eutrophication, and direct human exploitation. Most important, sustainable environmental management implies that the welfare of ecosystems is intimately tied to the welfare of human societies. Environmental management must consider both the human and non-human implications of decisions, and more often than not decisions that promote one will promote the other as well.
In many ways the call for sustainable development is reinforcing a trend in environmental management that began well over a decade ago. Most of our major ecosystems have already been substantially altered from their pristine state, sometimes unintentionally, but more often through deliberate actions intended to benefit man. Restoration rather than protection is the appropriate management goal. Attempts to restore these systems must recognize that most have been altered as greatly by erosion, physical alteration, and harvesting as they have by chemical pollution. Water quality management in the United States has been moving away from regulation based on the toxicity of the effluents being discharged to regulation based on measurements of the biotic structures of the ecosystems receiving the effluents (Bascietto et al., 1990).
On a higher level, the USEPA's Science Advisory Board recommended that the Agency adopt an integrated approach to environmental management by using comparative risk assessments to prioritize its research and regulatory programmes. A national Environmental Monitoring and Assessment Program was instituted to measure the state of the major US ecosystems and provide agency decision-makers with information that can be used to allocate programme resources (Hunsaker et al., 1990). Industry is also taking an active role, with many large companies developing their own pollution reduction programmes and attempting to assess the environmental impacts of their products over the full product life cycle, from resource extraction to ultimate end use and disposal (SETAC, 1991).
If environmental goals and regulatory approaches change, then assessment and measurement endpoints for ecological risk assessments must also change. From the sustainability perspective, the key management questions relate to the implications of environmental change for human and environmental welfare. Two aspects to the problem of how to formulate ecological risk assessments in terms provide useful answers to these questions: the scale of the technical studies supporting exposure and exposure-response assessments and the units in which risks are characterized and communicated. As noted previously, scientific studies must deal with tractable-sized systems and with durations of no more than a few months for laboratory studies or a few years for field studies. Space and time scales for management are quite different: the unit of management is likely to have an area of hundreds or thousands of km2, and the planning timeframe is likely to be measured in decades. Extrapolating from scientifically tractable scales to scales of management interest requires the use of statistical and mathematical models (Suter and Barnthouse, 1993). Several different kinds of extrapolations exist. Experimental studies such as toxicity tests performed on single species must be extrapolated to effects on ecosystems containing many interacting species; Bartell et al. (1992) and Suter (1993b) have recently summarized methods for making these extrapolations. Methods to quantify characteristics of landscapes that are mosaics of many different ecosystems have now been developed (Turner and Gardner, 1991), and models that link atmospheric transport of pollutants to large-scale ecological effects have been developed and used to support regulation of sulphur and nitrogen emissions (Hordjik, 1991). These techniques cannot substitute for rigorous experimental and observational studies, but they are essential for interpreting the management-scale implications of empirical information.
The second component of the extrapolation problem is the translation of ecological characterizations of exposures and effects into estimates of risks to human-ecological welfare. Conventional descriptors of population and ecosystem status, e.g., numbers, biomass, trophic structure, and productivity, are meaningful to ecologists, but not to decision-makers or to the public. Means must be found to translate these descriptors into measures of gained or lost value to society. Increasingly, the method chosen to make this translation is economic valuation. Economic measures such as marketable discharge permits and pollution taxes are replacing conventional emissions limits and performance standards as the preferred means of regulating industrial pollutant emissions (Schmidheiny, 1992). In the United States, the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA) permits federal agencies, states, and other institutions responsible for managing natural resources to obtain monetary compensation for damages to those resources caused by oil spills and other forms of chemical contamination (USDI, 1991). These approaches to environmental management , require that accurate monetary values be assigned to changes in the quality of the environment. These values must include marketable goods such as timber and fish, indirect services of nature such as pollution abatement and flood control, recreational use, and aesthetic values unrelated to actual use of the resources. A substantial literature on economic valuation of wetland ecosystems already exists (Scodari, 1990); the literature on ecosystem valuation in general is rapidly growing (Orians, 1990; Costanza, 1990). Although generally accepted principles do not yet exist, ecologists can expect eventually to be more frequently asked to express the results of ecological studies in terms that can be used as starting points for economic valuation studies.
A focus on assessing the overall quality of ecosystems, on restoration and maintenance of ecosystem quality, and on management of multiple human influences on ecosystems will probably also require changes in the kinds of data collected in chemical effects studies. The range of activities in chemical effects assessment will have to be broadened to address questions of rehabilitation, of causal relationships, and of impaired human welfare. Key scientific issues include (1) distinguishing the influence of chemicals from the influence of other stressors, (2) predicting and measuring recovery from reduced chemical inputs, and (3) estimating impacts of chemicals on the use of the environment by man, including recreation, flood control, natural biodegradation capacity, and aesthetic enjoyment. The first two of the above issues relate to retrospective risk assessment as defined above. Observed adverse ecological conditions must be causally related to specific sources of stress so that managers can determine which stress to reduce. Koch's postulates (Suter, 1993b) and Hill's criteria (Hill, 1965; USEPA, 1992) have both been suggested as decision rules to determine ecological causal relationships. Both approaches require that empirical data include information on toxicity, mode of action, spatial distribution of contaminants and exposed organisms, temporal sequence (i.e., putative cause must regularly precede observed effect), and exposure-response relationships from both experimental and observational studies.
The expected rate and degree of recovery resulting from different potential restoration activities must be predicted so that the beneficial effects of the alternatives can be evaluated and compared to the costs. This requirement is simply another form of extrapolation, and similarly requires the same kinds of extrapolation techniques described by Bartell et al. (1992) and Suter (1993a). If monitoring following institution of restoration actions were routine, our ability to predict restoration following future actions would be greatly enhanced (Yount and Niemi, 1990).
The third issue, that of estimating impacts of chemicals on human use of the environment, implies that ecologists and social scientists must cooperate in designing chemical effects studies. Ecological studies will have to be integrated with socioeconomic studies so that data will be available to interpret the socioeconomic implications of the measured or predicted ecological changes.
A framework for ecological risk assessment defines the relationship between ecological science and environmental management. The role of science in risk assessment is to ensure that the actions implemented by environmental managers achieve the goals and objectives defined by society. The science required to achieve the goal of environmental sustainability is substantially broader than the science relevant to past approaches to ensuring chemical safety, which emphasized media-specific concentration limits and technology standards. This changing management perspective implies that eventually the science required to support management of risks related to chemicals in the environment will include (1) more emphasis on diagnostic studies to determine which systems are being affected and by which chemicals, (2) more attention to facilitating recovery of damaged ecosystems subjected to multiple human influences, (3) a greater need for large- scale, long-term studies, and (4) explicit integration of ecological and
socioeconomic concerns. Increased communication will be required both between ecologists specializing in different kinds of ecosystems and between ecologists and social scientists.
These steps do not require the construction of expensive laboratory facilities or the use of high-performance computers. The process outlined in
Figure 17.1 calls only for careful identification of assessment objectives, assembly of multidisciplinary data collection and assessment teams, and rigorous interpretation of results using both quantitative and qualitative methods. These requirements are essentially identical to those for sound environmental impact assessments (EIA), as developed and implemented world-wide since 1970.
In fact, the definition of risk assessment proposed here includes EIA as a form of risk assessment concerned with predicting impacts of projects (e.g., power plants or dams) as part of the project planning process. Wherever an infrastructure for EIA exists, that same infrastructure can be used to support ecological risk assessments.
Abel, R., et al. (1986) In: Proceedings of IEEE Oceans 1986 Organotin Symposium, Vol. 4, pp. 1314-1323. Institute of Electrical and Electronics Engineers (IEEE), Washington, D.C.
Alzieu, C. (1986) In: Proceedings of IEEE Oceans 1986 Organotin Symposium, Vol. 4, pp. 1130-1134. Institute of Electrical and Electronics Engineers (IEEE), Washington, D.C.
Barnthouse, L.W. (1993) The role of models in ecological risk assessment: a 1990s perspective. Environ. Toxicol. Chem. 11, 1751-1760.
Barnthouse, L.W., Suter II, G.W., and Rosen, A.E. (1990) Risks of toxic contaminants to exploited fish populations: influence of life history, data uncertainty, and exploitation intensity. Environ. Toxicol. Chem. 9, 297-311.
Bartell, S.M., Gardner, R.H., and O'Neill, R.V. (1992) Ecological Risk Estimation. Lewis, Boca Raton, Florida, 252 pp.
Bascietto, J., Hinckley, D., Plafkin, J., and Slimak, M. (1990) Ecotoxicity and ecological risk assessment: Regulatory applications at EPA. Environ. Sci. Technol. 24, 10-14.
Blunden, S.J., and Chapman, A. (1982) In: Craig, P.J. (Ed.), Organometallic Compounds in the Environment, pp. 110-150. John Wiley & Sons, New York.
Cohen, Y. (1986) Pollutants on a Multimedia Environment. Plenum, New York.
Costanza, R. (1990) Ecological Economics. Columbia University Press, New York, 525 pp.
Hill, A.B. (1965) The environment and disease: association or causation? Proc. Royal Soc. Med. 58, 295-300.
Hordjik, L. (1991) Use of the RAINS model in acid rain negotiations in Europe. Environ. Sci. Technol. 25, 596-603
Huggett, R.J., Unger, M.A., Seligman, P.F., and Valkirs, A.O. (1992) The marine biocide tributyltin. Environ. Sci. Technol. 26, 232-237.
Hunsaker, C.T., Carpenter, D.L., and Messer, J. (1990) Ecological indicators for regional environmental monitoring. Bull. Ecol. Soc. Am. 71, 165-172.
Jørgensen, S.E. (Ed.) (1984) Modelling the Fate and Effects of Toxic Substances in the Environment. Elsevier, Amsterdam.
Kendall, R.J. (1992) Farming with agrochemicals: the response of wildlife. Environ. Sci. Technol. 26, 239-245.
Kimerle, R.A., Werner, A.F., and Adams, W.J. (1983) Aquatic hazard evaluation principles applied to the development of water quality criteria. In: Cardwell, R.D., Purdy, R., and Bahner, R.C. (Eds.) Aquatic Toxicology and Hazard Assessment, Seventh Symposium, pp. 538-547. American Society for Testing and Materials, Philadelphia.
Kooijman, S.A.L.M. (1987) A safety factor for LC50 values allowing for differences in sensitivity among species. Water Res. 21, 269-276.
MacKay, D., and Patterson, S. (1993) Mathematical models of transport and fate. In: Suter II, G.W. (Ed.) Ecological Risk Assessment, pp. 129-152. Lewis, Boca Raton, Florida.
NRC (National Research Council) (1983) Risk Assessment in the Federal Government: Managing the Process. National Academy Press, Washington, D.C., 191 pp.
NRC (National Research Council) (1993) Issues in Risk Assessment, Vol. 3: Ecological Risk Assessment. National Academy Press, Washington, D.C.
OECD (1984) OECD Guidelines for Testing of Chemicals. Updated Guidelines. Organisation for Economic Co-operation and Development, Paris.
OECD ( 1989) Compendium of Environmental Exposure Assessment Methods for Chemicals. Environmental Monographs No.27 .Organisation for Economic Co-operation and Development, Paris.
Orians, G.H. (Ed.) (1990) The Preservation and Valuation of Biological Resources. University of Washington Press, Seattle.
Schmidheiny, S. (1992) Changing Course: A Global Business Perspective on Development and the Environment. The MIT Press, Cambridge, Massachusetts, 374 pp.
Scodari, P.F. (1990) Wetlands Protection: The Role of Economics. Environmental Law Institute, Washington, D.C., 89 pp.
SETAC (Society for Environmental Toxicology and Chemistry) (1991) A Technical Framework for Life-Cycle Assessments. Society for Environmental Toxicology and Chemistry , and SETAC Foundation for Environmental Education, Inc., Pensacola, Florida.
Shugart, L.R., McCarthy, J.F., and Holbrook, R.S. (1992) Biological markers of environmental and ecological contamination: An overview. Risk Anal. 12, 353-360.
Sloof, W., van Oers, J.A.M., and De Zwart, D. (1986) Margins of uncertainty in ecotoxicological hazard assessment. Environ. Toxicol. Chem. 5, 841-852.
Suter II, G.W., and Loar, J.M. (1992) Weighing the ecological risk of hazardous waste sites: The Oak Ridge case. Environ. Sci. Technol. 26, 432-437.
Suter II, G.W., Vaughan, D.S., and Gardner, R.H. (1983) Risk assessment by analysis of extrapolation error: A demonstration for effects on fish. Environ. Toxicol. Chem. 2, 369-378
Suter II, G.W., Rosen, A.E., Linder, E., and Parkhurst, D.F. (1987) Endpoints for responses of fish to chronic toxic exposures. Environ. Toxicol. Chem. 6, 793-809.
Suter II, G.W. (1993a) Ecological Risk Assessment. Lewis, Boca Raton, Florida, 505 pp.
Suter II, G.W. (1993b) Retrospective risk assessment. In: Suter II, G.W. (Ed.) Ecological Risk Assessment, pp. 311-364. Lewis, Boca Raton, Florida, 505 pp.
Suter II, W.G., and Barnthouse, L.W. (1993) Assessment concepts. In: Suter II, G.W. (Ed.) Ecological Risk Assessment, pp. 21-47. Lewis, Boca Raton, Florida.
van Straalen, N.M., and Denneman, G.A.J. (1989) Ecological evaluation of soil quality criteria. Ecotoxicol. Environ. 18, 241-245.
Volmer, J., Kordel, W., and Klein, W. (1988) A proposed method for calculating taxonomic-group-specific variances for use in ecological risk assessment. Chemosphere 17, 1493-1500.
Turner, M.G., and Gardner, R.H. (Eds.) (1991) Quantitative Methods in Landscape Ecology. Springer-Verlag, New York.
Urban, J.D., and Cook, N.J. (1986) Ecological Risk Assessment: Standard Evaluation Procedures. Report No. EPA 540/9-85-001. US Environmental Protection Agency, Hazard Evaluation Division, Washington, D.C. Available from NTIS (National Technical Information Service), Springfield, Virginia 22161, as PB 86-247-657.
USDI (US Department of the Interior) (1991) Natural resource damage assessments; notice of proposed rulemaking. Fed. Reg. 56, 19753-19773 (43 CFR Part 11), 29 April.
USEPA (US Environmental Protection Agency) (1992) Framework for Ecological Risk Assessment. Report No. EPA/630/R-92/001. US Environmental Protection Agency, Risk Assessment Forum, Washington, D.C.
World Commission on Environment and Development (1987) Our Common Future. Oxford University Press, Oxford, UK, 400 pp.
Yount, J.D., and Niemi, G.J. (1990) Recovery of lotic communities and ecosystems from disturbance¾a narrative review of case studies. Environ. Manag. 14, 547-569.
|
|
|
The electronic version of this publication has been
prepared at |