• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Ecological Risk Assessment and Ecological Epidemiology for Contaminated Sites

Posted on: Wednesday, 15 March 2006, 12:00 CST

By Suter, Glenn W II

ABSTRACT

After 20 years of development, ecological risk assessment is widely accepted. However, it is evolving in response to a variety of technical and societal pressures. First, pressure for greater simplicity and standardization arise from the expectation that risk assessments should require little time and resources but be defensible. Second, the advance of the environmental sciences and increasing awareness of the complexity of ecological responses generate pressure for greater realism. Third, the dominance of human health risk assessment generates a pressure to integrate ecological risk assessment with that dominant field. Fourth, the demand for cost-benefit analysis creates pressure for integration with environmental economics. Finally, the need to connect the practice of ecological epidemiology with risk-based decision-making creates a pressure of the formation of a single integrated ecological assessment practice.

Key Words: risk assessment, contaminated sites, ecology, decision- making.

INTRODUCTION

After 20 years of development, ecological risk assessment is widely accepted. However, it is evolving in response to a variety of technical and societal pressures. This article proposes that the best means of responding to these various pressures is to develop two distinct ecological risk assessment practices. First, risk assessments should begin with a screening assessment that is standardized and can resolve the status of most environmental hazards. second, for those assessments that require an estimate of risk, technical support should be made available that puts the best methods, models, and data at the practitioner's disposal.

In 20 years, ecological risk assessment has gone from a United States Environmental Protection Agency (USEPA)-funded project at the Oak Ridge National Laboratory in Tennessee to a standard tool for environmental regulation and management in most industrialized nations. It now engages the efforts of many thousands of scientists in government, industry, and even academia. Although the ecological risk assessment paradigm has many variants, they all have the same basic structure of a preliminary step that defines goals, the system to be assessed, and the approach to be used in the assessment; an analytical step that defines exposures and the relationship between exposure and effects; a characterization step that uses the results of analysis to estimate risks and associated uncertainties; and finally, provision for either iterating the assessment or informing a management decision (USEPA 1998; Power and McCarty 1998). In sum, ecological risk assessment has become a mature practice. At the same time, technical and societal pressures are pushing ecological risk assessment in various directions. Assessors and the users of risk assessment must modulate or redirect these pressures to ensure that ecological risk assessment becomes more useful. This article examines some of these pressures and suggests one strategy for response.

GREATER SIMPLICITY AND STANDARDIZATION

The pressures for standardization are apparent to any practitioner. If standard data sets are used in simple methods and models in a standard framework, it is possible to do more assessments per unit of time and resources. In addition, there is less need for expertise in the practitioners, because the options for employing expertise are constrained. Further, the results of standard assessments are more easily defended. A standard method is, by definition, not arbitrary or capricious and defenses of standard assessments can appeal to prior legal and scientific reviews. These are all important attributes given the criticism that science-based regulation is inefficient and too subject to technical challenges (Houck 2004). Examples include not only standard frameworks, but also generic assessment endpoints (USEPA 2004), standard ecosystem models (USEPA 2003b), standard data sets such as ECOTOX, and standard screening benchmarks (Suter 1996; USEPA 2003b).

GREATER REALISM

A counter pressure pushes for incorporation of the best science into assessments that address the peculiarities of each individual assessment. For example, when it became apparent that simply enforcing single chemical water quality criteria was not meeting the goals of the U.S. Clean Water Act, effluent and ambient water toxicity tests were developed and bioassessment methods based on biological surveys (Ohio EPA 1998). Similarly, tests of contaminated media and biological surveys are used to increase the realism of contaminated site assessments (Suter et al. 2000). However, even this increased degree of realism is often not adequate. For example, the controversy over the role of pesticides in regional declines of frog and toad species is not resolved by the standard tests or surveys (Linder et al. 2003). The difficulty is illustrated by the toxicity of carbaryl, which is highly variable depending on pH, temperature, the presence of competitors and predators, and the species of amphibian being tested (Boone and Bridges 1999; Bridges 1999a, b; Relyea 2003). If the stakes are high enough in terms of the values of both the agent and the receptors, the costs of studies needed to assess such complexities are justified.

INTEGRATION OF HEALTH AND ECOLOGICAL RISKASSESSMENTS

Ecological risk assessment and health risk assessment have been developed independendy and practiced separately. This has obvious disadvantages in terms of efficiency and providing decision-makers with a coherent and consistent description of the risks. For example, expressing health and ecological risks at a contaminated site in a consistent manner allows the decision-maker to balance those risks against the ecological damage done by removing contaminated soil, vegetation, or other media. A potentially more important advantage is the expansion of health risk assessment from concern with the direct effects of chemicals, radiation, and other agents on human individuals to include indirect effects mediated by die environment. These indirect effects range from increased exposures to pathogens and algal toxins due to environmental disturbances to reduced health and well-being due to a diminished experience of nature (Wilson 1998; Reach 1998; Conway 1999; Van Dolah 2000). In addition, the concept of wildlife as sentinel species for humans is receiving increasing attention (Colborn and Thayer 2000). Hence, ecological risks are risks to humans as well as to nonhuman organisms and ecosystems. Integrated health and ecological assessments that quantify those relationships could increase the influence of ecology in decision-making, but will require new science, new assessment, methods, and more assessment effort. The World Health Organization's framework for integrated risk assessment can provide a basis for taking those steps (WHO 2001; Suteretal. 2003).

INTEGRATION OF RISKAND BENEFITS ASSESSMENTS

Increasingly, risk assessors are expected to specify the benefits of risk reduction. For example, they are asked to quantify the increased numbers in a salmon run that would result from removal of a dam or the increase in stream miles that would be restored to a natural level of diversity if a waste from a feedlot was treated. This demand is in part because many environmental problems are less obvious, so more justification is required for remedial actions. Perhaps more importandy, the results of public opinion polls and of elections indicate that the political climate has changed with respect to the environment. As a result, utilitarian rationales are increasingly supplanting rights and ethical obligations as bases for regulatory and management decisions. This creates a serious challenge to ecological risk assessors. Assessors have developed methods for determining the likelihood that a safe exposure level will be exceeded, but have seldom specified the benefits of avoiding that exceedence. How can assessors quantify the many benefits of restoring an ecosystem or preserving a species? This challenge is also an opportunity. The many services of the environment are underappreciated and their quantification should increase the pressures for protection and restoration (Costanza et al. 1997; Daily 1997).

INTEGRATION OF ECOLOGICAL RISKASSESSMENT AND ECOLOGICAL EPIDEMIOLOGY

Risk assessment is, by definition, concerned with the uncertain future consequences of actions. Hence, risk assessments are based on physical, mathematical, or statistical models. Monitoring of the real world, which is inherently about the past, can only tell us if prior assessments and management actions succeeded or failed. Although it is clearly desirable for risk assessors to know the state of a system before management actions are taken (to improve models of future states) and after (to validate the models), biological monitoring data are seldom available to ecological risk assessors.

A separate assessment tradition has grown up around the use of biological survey data to drive management actions. This approach, known as ecoepidemiology, begins with observations of biological conditions (biological surveys), determines whether the conditions are impaired (bioassessment), and recommends actions based on inferences concerning the causes o\f impairment (causal analysis). Practitioners of bioassessment tend to be dismissive of risk assessment and its models (Karr and Chu 1997). However, assessments based on only survey data cannot address the outcomes of actions other dian no action.

There are increasing practical pressures to merge ecological risk assessment and ecoepidemiology. Risk assessors need more grounding in the reality of the field, and bioassessors need better grounding in the reality of decision-making. The connecting concept is causality. Biological surveys and associated monitoring activities can demonstrate associations of effects with potential causes but cannot demonstrate that those associations are causal. Causal analysis requires the development of conceptual models, exposure- response relationships, mechanistic models, and other tools of risk assessors (USEPA 2000; Suter et al. 2002). On the other side, decisions are not well informed unless causal risk models link alternative actions to future conditions.

Environmental managementwould benefit from a unified practice of risk assessment and ecoepidemiology. A framework for that unified practice is shown in Figure 1. The diagram is a cycle of identification of agents of concern, risk assessment of the management options including no action, management decision-making and subsequent actions, monitoring of the results, assessment of the biological system, analysis of the causes of ecological impairment, and finally those causes are agents that might be subject to risk assessment. Ideally, this basic cycle would end when bioassessment reveals no remaining impairment, but it might also end with a management decision to do nothing or to take an action but not monitor the results. The cycle would start with an outside impetus. Most commonly, the impetus is a mandate to assess risks from identified agents of concern. These mandates include registration of a new pesticide, a remedial program for a contaminated site, or response to a product spill. Alternatively, the impetus may be for a monitoring program that includes biological surveys. These may be routine environmental monitoring programs, monitoring mandated by regulations, or ad hoc monitoring because offish kills, reports of deformed frogs, and so on. Required outside input to the cycle includes management alternatives to be assessed and other management considerations such as costs and stakeholder preferences. The ascending left side of the diagram is ecological epidemiology. The descending right side is risk-based decision-making.

Figure 1. A framework for integrated ecological assessment and management. Steps in the assessment cycle are represented by rectangles, regulatory or policy impetus for assessment by parallelograms, and input from other types of assessment by rounded rectangles. Dotted lines indicate alternative pathways.

The advantages of this integrated ecological assessment cycle are that die best information is brought to bear on environmental management decisions, no matter what the impetus, and that the efficacy of management actions is determined and used to inform subsequent assessments and decisions. It may be implemented in a way that constitutes adaptive management (Walters 1986). That is, the management actions may be designed so that the results of monitoring can be used not only to experimentally validate the management decision, but also resolve uncertainties in the risk assessment models for the next round of risk-based management. Implementation of this assessment cycle will require greater effort plus some modification of current practices. In particular, bioassessments commonly generate multimetric indices that are not suited for causal analysis or ecological risk assessment because they are not real- world properties that can be mechanistically related to potentially causal agents (Suter 2001).

HOW SHOULD ASSESSORS RESPOND?

Clearly, the demand for greater simplicity and standardization is in conflict with the demands for greater realism and more integration. Assessors must respond to Houck's (2004) criticisms of science-based environmental management by finding ways to bring good science to bear on environmental decision-making without unacceptably impeding remediation and restoration of the environment. I believe that this can be accomplished by a two-part strategy: standardize as much as possible and facilitate the rest.

First, standardize those parts of the process that are appropriate. In general, this means standard methods for screening assessments. Screening assessments are intended to assign hazards to one of three bins: clearly acceptable, clearly unacceptable, or uncertain (more assessment needed). They are distinct from definitive assessments, which estimate risks and their associated uncertainties. Screening assessments serve to narrow the scope of subsequent assessments, but may themselves be definitive if the hazards all fall into one of the extreme bins or if action is forced. For example, one unacceptable chemical in soil could force remediation even if others are acceptable or ambiguous. Screening assessments can be based on standard exposure models and assumptions and standard ecotoxicological benchmarks for contaminated sites or standard factors or other extrapolation models for new chemicals. An example is the development of ecological soil screening levels by the USEPA (2003b). With sufficient standardization, screening ecological risk assessments can be automated. Well-conducted screening assessments should resolve most hazards and many contaminated sites.

Second, facilitate those assessments that quantify specific ecological risks, particularly when assessing non-routine hazards. Examples include chemicals with ambiguous but potentially important effects such as atrazine and perchlorate. They also include observed effects that are ambiguously associated with pollutants like frog deformities and population declines possibly associated with pesticides (Linder et ai 2003). Most often, they are simply cases in which the exposure and effects levels are simply too close to clearly differentiate, given the complexity of real ecosystems. In most cases, the problem is not a lack of scientific approaches to address these problems. Ecology, environmental toxicology, environmental chemistry, epidemiology, remote sensing, systems analysis, statistics, network analysis, and other fields offer an overwhelming array of potentially applicable data, tools, and results. No assessor or small team of assessors will be familiar with all of them, nor will they have the time and resources to explore them and acquire or hire the needed expertise. New methods and results are constantly becoming available. Texts and guidance documents are not sufficiently accessible, their applicability is often unclear, and they become out of date more quickly than they can be revised.

Assessors need help. Ideally, this would be in the form of an expert system, but the sort of well-defined problems and experts that are successfully addressed by such systems are not characteristic of problems in ecological risk assessment. Assessors need systems that provide expertise in the form of guides to inference and access to tools and data. One such system is beginning to be developed to help U.S. states and tribes to determine the causes of impairments of aquatic communities. The Causal Analysis Diagnosis Decision Information System (CADDIS) will support the use of the USEPA's Stressor Identification Guidance (USEPA 2000). It will lead users through the stressor identification process, provide information, links to data sources, and analytical tools, and will provide case studies to show how the resources may be used. Eventually, systems like CADDIS should be expanded to include the full array of measurement, testing, and assessment tools needed to implement the framework shown in Figure 1. These systems will be fully successful only if they are actively developed, expanded, and updated by the user community. No single agency or organization is likely to devote sufficient resources to accomplish that goal. Hence, such systems must assimilate the shared expertise and information of the entire community of ecological assessors. Open source software is a possible model. Conceptually, it would be the product of a self-organizing learning community that generates what complexity theorists call a schema (Gell-Mann 1994).

Once ecological risk assessors have developed the capability of routinely generating credible estimates of risk, they can look outside our field to integrate with the rest of the contributors to environmental decision-making. That is, once assessors can predict the effects of proposed actions on nonhuman organisms, populations, and communities, they can collaborate with economists to estimate the benefits of environmental protection, and work with health scientists, sociologists, and psychologists to elucidate the full range of benefits to humans of the environment.

REFERENCES

Boone MD and Bridges CM. 1999. The effect of temperature on the toxicity of carbaryl for survival of tadpoles of the green frog (Rana clamitans). Environ Toxicol Chem 18:1482-4

Bridges CM. 1999a. Effect of a pesticide on tadpole activity and predator avoidance. Environ Toxicol Chem 33:303-6

Bridges CM. 1999b. Predator-prey interactions between two amphibian species: effects of insecticide exposure. Environ Toxicol Chem 33:205-11

Colborn T and Thayer K. 2000. Aquatic ecosystems: harbingers of endocrine disruption. Ecol Applications 10:949-57

Conway RA. 1999. A broader view. Environ Toxicol Chem 18:1077

Costanza R, d'Arge R, deGroot R, et al. 1997. The value of the world's ecosystem services and natural capital. Nature 387:253-60

Daily GC (ed). 1997. Nature's Services: Societal Dependence on Natural Ecosystems. Island Press, Washington, DC, USA

Gell-Mann M. 1994. The Quark and the Jaguar: Adventures in th\e Simple and the Complex. W. H. Freeman, New York, NY, USA

Houck O. 2004. Tales from a troubled marriage: science and law in environmental policy. Science 302:1926-9

Karr JR and Chu EW. 1997. Biological monitoring: essential foundation for ecological risk assessment. Human Ecol Risk Assess 3:993-1004

Keach S. 1998. Assessing Quality of Life Issues in EPA Sponsored State and Local Comparative Risk Projects. US Environmental Protection Agency; Washington, DC, USA. Available at http:// www.epa.gov/comp_risk/cr/qol3.htm

Linder G, Krest SK, and Sparling DW (eds). 2003. Amphibian Decline: An Integrated Analysis of Multiple Stressor Effects. SETAC Press, Pensacola, FL, USA

Ohio EPA. 1998. Biological Criteria for the Protection of Aquatic Life: Volume II: Users Manual for Biological Assessment of Ohio Surface Waters. Report WQMA-SWS-6. State of Ohio Environmental Protection Agency, Columbus, OH, USA

Power M and McCarty LS. 1998. A comparative analysis of environmental risk assessment/risk management frameworks. Environ Science Tech 32:224A-31A

Relyea RA. 2003. Predator cues and pesticides: a double dose of danger for amphibians. Ecol Applications 13:1515-21

Suter GWII. 1996. Toxicological benchmarks for screening contaminants of potential concern for effects on freshwater biota. Environ Toxicol Chem 15:1232-41

Suter GW II. 2001. Applicability of indicator monitoring to ecological risk assessment. Ecol Indicators 1:101-12

Suter GWII, Efroymson RA, Sample BE, et al. 2000. Ecological Risk Assessment for Contaminated Sites. Lewis Publishers, Boca Raton, FL, USA

Suter GW II, Norton SB, and Cormier SM. 2002. A methodology for inferring the causes of observed impairments in aquatic ecosystems. Environ Toxicology Chem 21:1101-11

Suter GW II, Vermier T, Munns WRJr, et al. 2003. Framework for the integration of health and ecological risk assessment. Human and Ecological Risk Assessment 9:281-302

USEPA (US Environmental Protection Agency). 1998. Guidelines for Ecological Risk Assessment. Report nr EPA/630/R-95/002F. Risk Assessment Forum, Washington, DC, USA

USEPA (US Environmental Protection Agency). 2000. Stressor Identification Guidance Document. Report EPA/822/B-00/025. Office of Water, Washington, DC, USA

USEPA (US Environmental Protection Agency). 2003a. AQUATOX for Windows, Release 1: A Modular Fate and Effects Model for Aquatic Ecosystems. Report EPA/823/R-007. Office of Water, Washington, DC, USA

USEPA (US Environmental Protection Agency). 2003b. Ecological Soil Screening Level Guidance. Report OSWER Directive 9285.7-55. Office of Emergency and Remedial Response, Washington, DC, USA

USEPA (US Environmental Protection Agency). 2004. Generic Ecological Assessment Endpoints (GEAEs) for Ecological Risk Assessment. Report EPA/630/P-02/004B. Risk Assessment Forum, Washington, DC, USA

Van Dolah FM. 2000. Marine algal toxins: origins, health effects, and their increased occurrence. Environmental Health Perspectives 108:133-41

Walters CJ. 1986. Adaptive Management of Renewable Resources. Macmillan, New York, NY, USA

WHO (World Health Organization). 2001. Report on Integrated Risk Assessment. Report WHO/IPCS/IRA/01/12. Geneva, Switzerland

Wilson EO. 1998. Integrated science and the coming century of the environment. Science 279:2048-9

Glenn W. Suter II

National Center for Environmental Assessment, Office of Research and Development, United States Environmental Protection Agency, Cincinnati, Ohio, USA

Received 31 January 2005; accepted 28 April 2005.

The views expressed in this article are those of the author and do not necessarily reflect the view or policies of the U.S. Environmental Protection Agency.

This article is a work of the U.S. Government and is not copyrighted.

Address correspondence to Glenn W. Suter II, National Center for Environmental Assessment, Office of Research and Development, United States Environmental Protection Agency, Cincinnati, OH 45268, USA. E- mail: Suter.Glenn@epa.gov

Copyright Taylor & Francis Ltd. Feb 2006

Source: Human and Ecological Risk Assessment

More News in this Category