SCOPE 53 - Methods to Assess the Effects of Chemicals On Ecosystems

1.1 INTRODUCTION

Ecological policy and management decision makers require sound scientific information upon which to base and justify their decisions; therefore, environmental research and monitoring must provide the full range of information required by decision makers. Historically, studies of the effects of chemicals on flora, fauna, and the environment have focused almost exclusively on identifying how those effects benefit or harm humans. As a result, environmental decision makers have selected natural resources that are particularly valued for food, shelter, medicine, recreation, or other uses. Only in the last few decades have relationships been recognized between these critical natural resources and the many components of the environment not of direct use to humans. As knowledge of the interconnectedness of the Earth's ecosystems increases, so does the recognition of the importance of obtaining comprehensive information about the effects of chemicals on all components of ecosystems. The success of concepts such as "sustainable development" is inconceivable without an improved understanding of the function and integrity of ecosystems and of how they are affected by human-induced stresses. This growing awareness of the importance of "healthy" ecosystems is coupled with a heightened interest in the ethical questions related to managing the uses of our natural environment.

Recognizing that all ecosystems are of indisputable importance to our future, useful approaches to make decisions to protect them must be developed and disseminated. Because the cost of effective environmental protection can be significant (e.g., the expense of developing and using alternative methods in manufacturing, proper waste treatment and disposal methods, recycling; installing pollution controls and monitoring their effectiveness), efforts to control the introduction of chemicals to the environment must be justifiable. Decisions to incur the cost of protecting the environment are influenced by factors such as:

1. the value of the ecosystem in terms of societal and cultural priorities, economic considerations, and ecological functions;  

2. the certainty that an effect will occur at expected or observable levels; 

3. the geographic extent and magnitude of the effect; 

4. the certainty of the cause of an observed effect; and 

5. the management options available to reduce the risk of adverse effects.

This volume presents approaches and methods to study the effects of chemicals on ecosystems in order to collect information that promotes risk-based decision making. Chapter 2 provides a discussion of the conceptual approaches to conduct ecological research and monitoring that includes characterizing "healthy" ecosystems, setting and maintaining objectives for ecosystem management, and using ecological risk assessment as a model for studying the effects of chemicals on ecosystems. Chapter 3 provides an overview of available methods to study the effects of chemical stressors in aquatic and terrestrial ecosystems. Chapter 4 focuses on the importance of large geographic-scale (regional) environmental monitoring and assessments.

The remaining chapters present papers contributed for review by the participants of SGOMSEC 10 Workshop, held in Montpellier, France, in March 1993. These papers describe and evaluate available methods to monitor chemical contamination and their effects on ecosystems.

1.2 ECOLOGICAL RISK ASSESSMENT: A TOOL FOR DECISION MAKING

Ecological risk assessment is useful to provide a framework to obtain information needed to make management decisions. Ecological risk assessments help to identify existing problems, anticipate the risks of planned actions, establish regulatory program and research priorities, and provide a scientific basis for regulatory actions. Most importantly, environmental risk assessment provides the basis to prioritize management and regulatory efforts because objective comparisons of a variety of situations can be made. Using risk as a common denominator allows the decision maker to select the most effective management actions. Following implementation of management actions, environmental risk assessment can also help to measure the effectiveness of those actions.

Before any data are collected or any research takes place, the researcher must consider how the environmental risk assessment will be carried out and what specific questions must be answered. The process of conducting an environmental risk assessment requires carefully identifying assessment objectives (environmental values to be protected), deciding the appropriate scale and level of biological organization, assembling multidisciplinary data collection and assessment teams, rigorously interpreting results using both quantitative and qualitative methods, and communicating the results in a manner that facilitates risk management. Chapter 2 presents a framework for environmental risk assessment drawn from several recent sources that present unified principles for assessing the ecological risks of toxic  chemicals and other stresses. The components of the suggested framework are illustrated in Figure 1.1. This risk assessment framework is intended to facilitate integration of human health and ecological risks of toxic chemicals and other environmental stresses, and, as far as the US is concerned, was developed by combining essential features of the National Academy of Sciences/National Research Council (NRC, 1983) framework for health risk assessment and the US Environmental Protection Agency (USEPA, 1992) framework for ecological risk assessment. This proposed framework is discussed within the contributed papers. This framework has four components: (1) hazard identification; (2) exposure assessment and exposure-response assessment; (3) risk characterization; and (4) risk management.

1.2.1 HAZARD IDENTIFICATION

This step determines whether a particular danger exists, if the effects associated with the hazard are sufficiently significant to warrant further study or immediate management action, and the kinds of data required to determine the level of risk. This systematic planning step establishes the goals and focus of the assessment, and identifies the major factors to be considered. A major factor to be considered at this step is the selection of ecologically based endpoints relevant to the ultimate decisions to be made. Information used for hazard identification include short-term or screening toxicity tests and reviews of existing information that characterize the potentially affected ecosystems and the contaminants in question.

Figure 1.1. Risk assessment framework that integrates human health risk (NRC, 1983) and ecological risks of chemicals (USEPA, 1992)

1.2.2 EXPOSURE ASSESSMENT

This step is the determination of exposure to the hazardous agent in question. This process includes measurement or prediction of movement, fate, and partitioning of chemicals in the environment. This step is typically accomplished through chemical analysis of site media or ecological receptors and/or mathematical modelling.

1.2.3 EXPOSURE-RESPONSE ASSESSMENT

This process is the determination of the relation between the magnitude of exposure and the probability of occurrence of the expected effects. Information useful in this step includes toxicity data (chronic toxicity, mode of action, sensitivities of particular species), mesocosm or field test data, field surveys comparing exposed and unexposed sites, and population or ecosystem modelling.

1.2.4 RISK CHARACTERIZATION

This step involves describing the nature and magnitude of risks, including, the inherent uncertainties, expressed in terms understandable to decision makers and the public. This step integrates information from the previous steps, and communicates it to decision makers in a manner relevant to the decisions being made. Because the purpose of risk assessment is to support decision making, communication with decision makers is a critical step in an environmental risk assessment.

1.2.5 RISK MANAGEMENT

This process is the one whereby decisions are made about whether an assessed risk needs to be managed, and the means for accomplishing it, for the protection of public health and environmental resources. Managing risks involves making decisions based on the information collected in the previous steps of the risk assessment along with a consideration of social and cultural values, economic realities, and political factors.

Monitoring and research are used to provide the data necessary to conduct an environmental risk assessment. Important aspects to consider when planning monitoring for a risk assessment include the assumptions involved in determining which environmental values are to be protected by the risk management decisions; which levels of biological organization must be studied to obtain the data necessary to answer the pertinent questions; and the temporal and spatial scales relevant to the study. Each aspect is discussed in greater detail throughout this volume.

1.3 ECOSYSTEM OBJECTIVES

The first steps in performing an environmental risk assessment are the identification of the resources at risk and the definition of desired or acceptable conditions of those resources. Ecological "values" to be protected must be identified clearly to help define the monitoring objectives and the assessment endpoints of concern. Social, cultural, political, economic, and ecological values must contribute to the selection of assessment endpoints. Frequently, risk assessments focus only on resources of recognized economic value to human society. This narrow definition of the resources at risk ignores the critical components of an ecosystem that support and preserve the resources traditionally valued, because ecological values cannot be quantified easily or in terms that are meaningful to environmental policy makers. Risk assessments based solely on a narrow economic definition of valuable resources ultimately may fail to detect and quantify the risks that threaten those resources indirectly through effects on other components of the ecosystem. Furthermore, management efforts that neglect ecological values ultimately may be ineffective at protecting economically valuable resources. An approach is needed to translate the ecological value of ecosystem components that are not valued by human society into information that contributes to management decisions or can be used to justify these decisions. Such an approach must account for the fact that definitions of ecosystem values will always be influenced primarily by the needs and desires of human society and provide guidance to relate ecologically important components of an ecosystem to direct and indirect human uses of the system.

Once the values of an ecosystem are defined, its condition or degree of "health" must be characterized relative to some standard of condition in order to assess risk. Defining a "healthy" ecosystem provides the standard against which to evaluate its current condition and to predict the potential changes in that condition over time (i.e., trends) in response to a stress. Unfortunately, the current state of ecological science is insufficient to enable ecologists to define the optimum condition of most ecosystems. In most cases, defining the optimum level of species diversity within an ecosystem or to compiling a taxonomic list that precisely represents the optimal condition of a community is impossible. Several properties or attributes have been proposed to determine whether an ecosystem is in optimal condition of "health:" homeostasis, the absence of disease, diversity or complexity, stability or resilience, vigour or "scope for growth;" and the balance between ecosystem components.

All of these have shortcomings that limit their utility. For example, homeostasis, stability, and resiliency cannot be expected in ecosystems in early stages of succession. Vigour also depends on the successional stage, and the balance between ecosystems is difficult to evaluate without a reliable reference. Characterizing the condition of an ecosystem on the basis of diversity assumes a relationship between diversity and stability that cannot be generalized confidently. Despite these limitations, evaluation of the condition of an ecosystem should be possible by studying the structural and functional properties demonstrated to be essential in determining the condition of ecosystems.

Chapter 2 describes an approach for ascribing value to ecosystems known as the "Ecological Benefits Paradigm," discusses the advantages and difficulties of using the approach, and includes suggestions for appropriate assessment endpoints and related indicators. Reasonable approaches are suggested to define the condition of aquatic and terrestrial ecosystems based on the current state of ecological knowledge. The utility of defining quality objectives for ecosystems is also discussed. Quality objectives quantify desirable characteristics of ecosystems, and can be used to gauge the magnitude of risk and the efficacy of management strategies. Chapter 2 concludes with a discussion of the importance of effectively communicating risk to all concerned with the condition and protection of an ecosystem ("stake-holders").

1.4 METHODS FOR STUDYING EFFECTS

Once the objectives and relevant scales of an environmental risk assessment are determined, a data collection plan must be developed. To acquire useful information, such a plan involves identification of methods for both field and laboratory studies and of the tools available to assist an environmental manager in measuring environmental conditions. These tools include chemical analyses, toxicity tests, field surveys and assessments, and special studies such as biomarkers and simulated ecosystem studies (i.e., mesocosms or microcosms). Any of these tools may provide the information needed to assess environmental conditions; however, the use of several of these approaches in combination may increase the level of confidence in the conclusions.

After determining the study objectives and implementing a design, measurement endpoints or indicators must be selected. Indicators provide data about the assessment endpoints determined to be most relevant to the study objectives. Methods to measure indicators can be grouped into several categories: fieldecological surveys of the contaminated site; chemical analyses of samples from the site; and toxicity tests of site media (water, soil, or sediments) either in the laboratory or on site. Special studies, such as bioaccumulation tests, and simulated ecosystem studies, such as microcosms or mesocosms, may also be used.

A choice that must be made in developing an evaluation strategy is whether the question is best answered by field or by laboratory studies. Laboratory studies permit better control of naturally varying environmental factors that can confound interpretation of biological response, and may provide an improved means to address specific questions about pathways or modes of action. Field studies provide an opportunity to translate the potential for impact into an estimate of extent of alterations. Mesocosm studies represent an intermediate approach that can potentially bridge the gap between these two extremes. The nature of the exposure and the type of information needed as input to the risk assessment will determine which is the most fruitful approach.

An important issue involved in the selection of an indicator to evaluate the significance of effects of chemicals is the determination of what level of biological organization is most appropriate to the ecosystem objective being evaluated. Biota can be sampled at the cellular, organismal, population, or community levels. Generally, the higher levels of organization address questions concerning extent and magnitude of impact. Indicators associated with the lower levels of organization tend to measure exposure to stress or contamination, and are generally more useful in identifying specific causes of population and community modifications. The various levels are complementary, and the selection of indicators at alternative levels depends on the questions posed by, and objectives of, the risk assessment.

Quality assurance and quality control are important parts of any data collection plan, but are particularly important in the context of risk assessment. Risk assessment often involves comparing populations in space or time. Failure to standardize methods or quality of the data produced could lead to the conclusion that one population is less affected than another, when the result only reflects differences in accuracy among investigators. The extent of quality assurance necessary depends largely on the scale of the study. Studies with larger spatial scales generally involve a greater number of participating organizations, and, therefore, require more attention to standardization. Studies encompassing longer temporal scales require quantification of the extent to which methodological improvements over time alter results.

Chapter 3 provides an overview of currently available monitoring methods for aquatic and terrestrial systems, and discusses appropriate uses of these methods within the ecological risk assessment framework. Many of these methods are discussed in more detail in the contributed papers.

1.5 MEASURING EFFECTS ON LARGER GEOGRAPHIC SCALES

Once the values of interest have been defined and methods are being considered, the risk assessment must be focused on the proper spatial and temporal scales. To focus the assessment, the geographic extent of the resource at risk and what portion of the resource has already been compromised must be known. Increasingly, scientists and decision makers are becoming aware of environmental problems that occur at large geographic scales: acid deposition, non-point source pollution, and loss of biodiversity. For this reason, environmental managers and decision-makers should focus not only on effects on the local scale, but on the multimedia impacts that occur at larger scales. Most environmental monitoring is limited to the local environment, so that specific pollutants can be linked to the source and controlled. Regrettably, this type of monitoring provides little information about the effects of chemical releases on entire large ecosystems. Such information would allow an enhanced evaluation of the risks of action or inaction.

Defining risk typically involves collecting and interpreting monitoring data that characterize ecological processes. Ecological processes occur on many different scales, and the spatial scale of the assessment is an important consideration. For instance, most point source contaminant inputs to aquatic systems occur on local scales, and the biological assemblages at risk may be limited to the immediate depositional zone. Nonpoint source inputs of contaminants, such as pesticides from agricultural land use practices, may affect assemblages at the watershed level, or over entire regions having similar farming practices and land use patterns. Atmospheric inputs typically affect areas that include multiple watersheds, and may involve global scales, as in the case of ozone depletion. Although environmental risk assessment requires investigations on these large scales, 90 percent of biological studies have been estimated to be conducted on spatial scales of less than 1 m² (Levin et al., 1990). Investigators should consider conducting studies on the scale at which decisions are to be made, or use statistical and mathematical models to extrapolate from scientifically valid scales to levels more useful to managers.

Part of defining the risk may involve estimating the degree and rate to which the status of the resource is changing, in which case the temporal scale of the assessment must be addressed. The appropriate temporal scale may be related to the timeframe over which the effect manifests itself or to the timeframe over which corrective actions occur. For example, cessation of particulate inputs may resolve smog in less than a year, whereas groundwater contamination may take many years to clear after the cessation of inputs. Assessing groundwater condition based on a single year of data may provide a deceptive, snapshot impression that conditions are bad, when in fact the inputs have stopped and the effects are lessening, but the decrease will be observable only on the time scale of decades.

The spatial scale and temporal scales over which a risk assessment is to be conducted will determine the methods available to document and evaluate effects of chemicals. Some indicators appropriate at local scales may be inappropriate as biogeographic boundaries are crossed. While local scale and site specific approaches are effective assessment methods for certain types of stressors and ecological resources, a larger scale perspective permits a broader, more comprehensive view of environmental problems that may impact large geographic areas. Chapter 4 describes spatial scales frequently used for such assessments and discusses their applicability to selected ecosystem objectives, and also discusses how data collection options are affected by changes in scale, and how assessment techniques may vary.

1.6 GENERAL RECOMMENDATIONS

1. Ecological risk assessment is a useful framework to make decisions and to design monitoring programs that contribute to environmental policy and management decisions. Identifying ecosystem values, reference conditions, and scales of interest are important first steps in the environmental risk assessment process that will guide the selection of monitoring methods. 

2. No single best method exists for comprehensive evaluation of a contaminated ecosystem. The most appropriate methods vary according to the nature of the ecosystem and the contaminant in question, cost and time constraints, and political and social constraints. However, development and employment of standardized methods are essential to perform replicate analyses and to implement quality assurance and control plans that ensure that data are reliable, valid, and comparable. 

3. Investigations at all levels of biological organization and geographic scales are important to understand and document the effects of chemicals on ecosystems.

4. Further development, validation, and standardization of appropriate methods to assess the ecological risks of chemical agents are required. Statistical designs for ecological monitoring must be improved. Dose-response curves should be established at the community and ecosystem levels. Studies of synergistic and antagonistic interactions among chemicals must be expanded. Research is also needed for biomarkers to identify exposures and toxicity in non-temperate biomes. 

5. Accidental discharges of chemicals should be recognized as opportunities to improve our understanding of an ecosystem's inherent resiliency and capacity for recovery. Information gained from accidents can be especially valuable in developing countries where ecosystems are generally poorly characterized and where the fate and effects of pollutants are not well-documented. 

6. Long-term monitoring should continue in order to enable estimation of natural variability, as well as to establish a baseline against which to evaluate the effects of disturbances. Characterization of ecosystems in developing countries is essential. Standard environmental risk assessment approaches (i.e., species diversity and the use of keystone species) may not be useful, if the components of the ecosystem in question have not been determined. Further research to characterize pristine ecosystems is needed to enable scientists to identify and quantify deviations from natural conditions.

7. The importance of time scales should be recognized, especially in the rates of contaminant release and in the resulting effects on biota. 

8. Methods to assess chemical contaminant levels in biological samples involve two strategies: examination of the properties of the ecosystem under study, and examination of biota from the ecosystem for contaminant effects. Either test, used alone, provides an incomplete picture of contamination; using both approaches is recommended to obtain a complete examination of contaminated ecosystems. 

9. Microcosm and mesocosm tests are useful intermediates between laboratory bioassays and ecosystem monitoring. Microcosm tests are generally small-scale laboratory tests that involve a few species, whereas mesocosms are relatively large, constructed in the field, and involve most or all species within an ecosystem. Mesocosm tests are most often conducted in aquatic systems. Whole-lake manipulations have been conducted on the landscape scale and involve intentional manipulation of entire aquatic systems. 

10. Mathematical modelling should continue to be used routinely to evaluate the cycling of materials such as nutrients or metals between biotic and abiotic components of an ecosystem. The simplest models are to be used to estimate chemical loading and chemical mass balances within aquatic systems.

1.7 REFERENCES

Levin, S.A., Harwell, M.A., Kelly, J.R., and Kimball, K.D. (Eds.) (1990) Ecotoxicology: Problems and Approaches. Springer Verlag, New York, 547 pp.

NRC (National Research Council) (1983) Risk Assessment in the Federal Government: Managing the Process. National Academy Press, Washington, D.C., 191 pp.

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.