Environmental Toxicology and Chernisiry, Vol. 11, pp.

1663-1672, 1992
Printed in the USA. Pergamon Press Ltd.

Annual Review
A FRAMEWORK FOR ECOLOGICAL
RISK ASSESSMENT AT THE EPA
SUSANB. NORTON,? DONALD J. RODIER,? JOHNH. GENTILE,$
WILLIAM H. VAN DER SCHALIE,~WILLIAMP. WOOD^
and MICHAELw. SLIMAK*t
7U.S. Environmental Protection Agency, Washington, DC 20460
1U.S. Environmental Protection Agency Laboratory, Narragansett, Rhode Island 02592

(Received 9 January 1992; Accepted 13 July 1992)

Abstract - Ecological risk assessments evaluate the likelihood of adverse ecological effects caused
by stressors related to human activities such as draining of wetlands or release of chemicals. The
term stressor is used to describe any chemical, physical, or biological entity that can induce adverse
effects on ecological components (i.e., individuals, populations, communities, or ecosystems). In
this review article, a historical perspective on ecological risk assessment activities at the U S . Envi-
ronmental Protection Agency (EPA) is followed by a discussion of the EPA’s “Framework Report,”
which describes the basic elements for conducting an ecological risk assessment. The “Framework
Report” is neither a procedural guide nor a regulatory requirement within the EPA. Rather, it is
intended to foster a consistent approach to ecological risk assessments within the Agency, identify
key issues, and define terminology.

Keywords - Risk assessment Ecological stressors

INTRODUCTION or guidance on factors influencing policy. Such is-

Environmental problems are often complex, sues are reserved for future guidance, and plans for
with multiple causes and diverse ecological effects. developing such guidance, based on the approach
Examples include the effects of global climate described in the “Framework Report,” are de-
change, habitat loss, acid deposition, and multiple scribed in the last section of this article. We hope
chemicals present in the environment. Dealing with that this article will broaden the audience for the
such problems requires a flexible decision-making “Framework Report” and help stimulate discus-
process that can accommodate this diversity while sions of the many issues that have been highlighted
providing some measure of the uncertainty associ- throughout the guideline development process.
ated with decisions that are made. At the U.S. En-
The nature of ecological risk assessment
vironmental Protection Agency (EPA), there is
increasing interest in using ecological risk assess- Ecological risk assessments evaluate the likeli-
ments as a basis for environmental decisions. hood that adverse ecological effects will occur as
This article examines the EPA’s past, present, a result of exposure to stressors related to human
and possible future utilization of ecological risk activities, such as draining of wetlands or release of
assessment approaches. The core of the discussion chemicals. The term stressor is used here to describe
is a summary of the EPA’s recently published any chemical, physical, or biological entity that can
“Framework Report,” which describes the basic induce adverse effects on ecological components,
elements of, or a framework for, ecological risk that is, individuals, populations, communities, or
assessment and has been proposed as a basis for ecosystems. Adverse ecological effects encompass
conducting ecological risk assessment within the a wide range of disturbances ranging from mortal-
EPA [l]. ity in an individual organism to a loss in ecosystem
The “Framework Report” does not contain sub- function. Thus, the ecological risk assessment pro-
stantive guidance on factors that are integral to the cess must be flexible while providing a logical and
risk assessment process, such as analytical meth- scientific structure to accommodate a broad array
ods, techniques for analyzing and interpreting data, of stressors and ecological components.
Ecological risk may be expressed in a variety of
*To whom correspondence may be addressed ways. Whereas some ecological risk assessments
1663
1664 S.B. NORTONET AL.

may provide true probabilistic estimates of risk, the use of laboratory bioassays and the quotient
others may be deterministic or even qualitative in method have provided results useful for decision-
nature. In these cases, the likelihood of adverse ef- making with a reasonable commitment of resources,
fects is expressed through a semiquantitative or there has been increased interest in strengthening
qualitative comparison of effects and exposure. the assessment of ecological effects. Advances in-
The appropriate application of risk assessment clude incorporating toxicity data for higher levels
helps meet the EPA’s goal of targeting environmen- of organization (e.g., through mesocosms), using
tal protection resources at the problems and the simulation models to project effects at lower tro-
geographic areas posing the greatest risks. The fol- phic levels to higher trophic levels, and developing
lowing section briefly discusses the use and devel- techniques to assess exposure to multiple chemicals.
opment of ecological risk assessment approaches in The Office of Water is required by the Clean
EPA programs. Water Act to restore and maintain the biological
integrity of the nation’s waters and, specifically, to
Ecological risk assessment at EPA ensure the protection and propagation of a balanced
Since its establishment in 1970, the EPA has at- population of fish, shellfish, and wildlife. The EPA
tempted to protect both human health and ecolog- also develops methods, including biological mon-
ical resources. Typically, however, activities most itoring and assessment methods, for establishing
closely related to human health have received the and measuring water-quality criteria. These statu-
highest priority. Nonetheless, assessment of risk to tory requirements have encouraged the Office of
ecological resources has been an important activ- Water to develop innovative approaches to eco-
ity for many programs at the EPA. The recent re- logical assessment. The Water Quality Act of 1987
port by the Science Advisory Board [2] strongly (Public Law 100-4) amends the decade-old Clean
supported an increased emphasis on risk-based de- Water Act and redirects its focus from the technol-
cision making and ecological risk assessment. The ogy approach, based on end-of-pipe standards, to
board’s recommendations included that (a) the full-scale implementation of the water-quality ap-
EPA should target its environmental protection ef- proach, based on ambient receiving water standards.
forts on the basis of opportunities for the greatest State water-quality standards and designated
risk reduction; (b) the EPA should attach as much uses form the backbone of the water quality-based
importance to reducing ecological risk as it does to approach, and the EPA criteria are developed as
reducing human health risk; and (c) the EPA national recommendations to assist states in devel-
should improve the data and analytical methodol- oping their standards. The most commonly used
ogies that support the assessment, comparison, and risk-based approaches to the evaluation of water
reduction of different environmental risks. quality are the application of chemical-specific wa-
The ecological risk activities of two EPA of- ter-quality criteria and whole-effluent toxicity cri-
fices-the Office of Pesticides and Toxic Sub- teria. Both criteria have three components, the first
stances and the Office of Water-illustrate the use characterizing the ecological effects and the latter
and development of ecological risk assessment at two assessing the exposure:
the EPA. The Office of Prevention, Pesticides and
Toxic Substances is concerned about potential im- 1. Magnitude - what concentration of a pollutant
pacts of pesticides and toxic chemicals on organ- (or a pollutant parameter such as toxicity) is
isms, including aquatic and terrestrial communities. allowable
Its legal mandates come from the Federal Insecti- 2. Duration-the period of time over which the
cide, Fungicide, and Rodenticide Act (FIFRA) and predicted in-stream concentration occurs (this
the Toxic Substances Control Act (TSCA). specification limits the duration of concentra-
Both programs under the Office of Prevention, tion above the criteria)
Pesticides and Toxic Substances- the Office of 3. Frequency - how often criteria can be exceeded
Pesticide Programs (OPP) and the Office of Pol- without unacceptably affecting the community.
lution Prevention and Toxics (OPPT) -assess risks
to ecological resources by an ecotoxicological ap- The approaches to risk-based water-quality crite-
proach: Laboratory toxicity bioassays are used to ria are being expanded by the Office of Water in
characterize ecologicaleffects; exposure is estimated their development of several new types of criteria,
by using either monitoring data or models; and the including chemical-specific sediment criteria and
risk is estimated by comparing the two with a sim- wildlife criteria, and biological criteria based on
ple quotient, risk = exposure/effects. Although community structure.
 Framework for ecological risk assessment at EPA 1665

Although the programs in the Office of Preven- point in the risk assessment process. At that point,
tion, Pesticides and Toxic Substances and the Of- the risk assessment stops, the necessary data are ac-
fice of Water have a longer history, all of the EPA’s quired, then the assessment resumes. Verification
programs are beginning to address ecological con- and monitoring can help determine the overall ef-
cerns. Responding to the increased emphasis on fectiveness of the framework approach, provide
ecological issues and the need to provide for some necessary feedback concerning the need for future
uniform procedures, the EPA’s Risk Assessment modifications of the framework, help evaluate the
Council (senior managers with significant respon- effectiveness and practicality of policy decisions,
sibilities for assessment and reduction of risks) di- and indicate the need for new or improved scien-
rected the Risk Assessment Forum to develop tific techniques [3].
ecological risk assessment guidelines using the same Finally, whereas risk assessment and risk man-
open process that was used in developing human agement are distinct processes, Figure 1 indicates
health risk assessment guidelines. The first docu- two points of interface between these two processes
ment produced from the process, the “Framework during discussions between the risk assessor and
Report,” is discussed below. risk manager. At the initiation of the risk assess-
ment, the risk manager can help ensure that the risk
EPA’S FRAMEWORK FOR ECOLOGICAL assessment will provide information relevant to
RISK ASSESSMENT making decisions on the issues under consideration,
The “Framework Report” describes a basic and while the risk assessor can ensure that the risk as-
flexible structure for conducting ecological risk as- sessment addresses all relevant ecological concerns.
sessments within the EPA. The framework provides Effective communication is also important at the
principles not only for estimating risks from chem- end of the risk assessment process to provide the
icals, but also for predicting impacts from non- risk manager with a full and complete understand-
chemical stressors (e.g., habitat loss from human ing of the assessment’s conclusions, assumptions,
activities), and retrospectively for assessing site-spe- and limitations.
cific impacts. The “Framework Report” is neither The remainder of this section discusses the three
a procedural guide nor a regulatory requirement major phases of ecological risk assessment shown
within the EPA. Rather, it is intended to foster a in Figure 1 : problem formulation, analysis, and
consistent approach to ecological risk assessment risk characterization. To illustrate the described
within the EPA, identify key issues, and define concepts and issues, a simplistic, hypothetical ex-
terminology. ample concerning nutrient loads to an estuary will
The framework is conceptually similar to the also be discussed.
approach used for human health risk assessment
but is distinctive in its emphasis in three areas. Problem formulation
First, ecological risk assessment can consider ef- Problem formulation is a planning and scoping
fects beyond those on individuals of a single spe- process that links the regulatory or management
cies and may examine a population, community, or goal to the risk assessment. Its end product is a
ecosystem. Second, there is no single set of ecolog- conceptual model that identifies the environmen-
ical values to be protected that can be generally ap- tal values to be protected (the assessment end
plied. Rather, values are selected from a number of points), the data needed, and the analyses to be
possibilities based on both scientific and policy con- used.
siderations. Finally, there is an increasing aware- The initial steps in problem formulation include
ness of the need for ecological risk assessments to the identification and preliminary characterization
consider nonchemical as well as chemical stressors. of the stressor, the ecosystem potentially at risk,
The framework for ecological risk assessment is and the ecological effects. Performing this analy-
illustrated in Figure 1 . The risk assessment process sis is an interactive process; for example, gathering
is shown within the bold line. Figure 1 also illus- information on the characteristics of a stressor
trates the interaction of risk assessment with data helps to define the ecosystems potentially at risk
acquisition, data verification, and monitoring. In from the stressor as well as the ecological effects
the “Framework Report,” a distinction is made be- that may result. The ecosystem within which effects
tween data acquisition (which is outside the risk as- occur provides the ecological context for the assess-
sessment process) and data analysis (which is an ment. Knowledge of the ecosystem potentially at
integral part of an ecological risk assessment). The risk can help identify ecological components (i.e.,
need for additional data may be identified at any individuals, populations, communities, or ecosys-
1666 S.B. NORTONET AL.

Ecological Risk Assessment

Discussion
between the
Risk Assessor
a-
Risk Manager
(Planning)

Discussion between the

Risk Assessor and Risk Manager

n Risk Management +- - - - - - -

Fig. 1. Framework for ecological risk assessment

tems) that may be affected and stressor-ecosystem vironmental values that are to be protected; mea-
interactions relevant to developing exposure scenar- surement end points are measurable responses to a
ios. Available information on ecological effects can stressor that are related to the valued characteris-
help focus the assessment on specific stressors and tics chosen as the assessment end points [4].
on ecological components that should be evaluated. Assessment end points are the ultimate focus in
Information compiled in the first stage of prob- risk characterization and link the measurement end
lem formulation is used to select ecologically based points to the risk management process (e.g., pol-
end points that are relevant to decisions made icy goals). When an assessment end point can be di-
about protecting the environment. An end point is rectly measured, the measurement and assessment
a characteristic of an ecological component (e.g., end points are the same. In most cases, however,
increased mortality in fish) that may be affected by the assessment end point cannot be directly mea-
exposure to a stressor [4]. Two types of end points sured, so a measurement end point (or a suite of
are distinguished in the framework: Assessment measurement end points) is selected that can be re-
end points are explicit expressions of the actual en- lated, either qualitatively or quantitatively, to the
 Framework for ecological risk assessment at EPA 1667

assessment end point. For example, a decline in a considered most likely to contribute to risk are se-
sport fish population (the assessment end point) lected for further evaluation in the analysis phase.
may be evaluated by using laboratory studies on Professional judgment is needed to select the most
the mortality of surrogate species such as the fat- appropriate focus for the risk assessment, and it is
head minnow (the measurement end point). important to document the selection rationale.
Assessment and measurement end points may In our hypothetical example, a risk manager
involve ecological components from any level of may be concerned about possible effects of nutri-
biological organization, ranging from individual ent inputs to an estuary. During the problem for-
organisms to the ecosystem itself. In general, the mulation phase, the ways that nutrient inputs may
use of a suite of assessment and measurement end cause effects in the estuary are described. For ex-
points at different organizational levels can build ample, nutrient loads may directly alter benthic
greater confidence in the conclusions of the risk as- community structure; ultimately result in decreased
sessment and ensure that all important end points dissolved oxygen levels, which then may increase
are evaluated. In some situations, measurement mortality rates of fish or invertebrates; or reduce
end points at one level of organization may be re- aquatic vegetation abundance, which then may in-
lated to an assessment end point at a higher level. directly affect wildlife and fish populations that de-
For example, measurement end points at the indi- pend on the plants. Although several of these
vidual level (e.g., mortality, reproduction, and hypotheses may be evaluated further in the analy-
growth) could be used in a model to predict effects sis phase, for the purposes of this example we will
on an assessment end point at the population level focus on effects to the aquatic vegetation. The as-
(e.g., viability of a trout population in a stream). sessment end point may be the maintenance of the
Sound professional judgment is necessary for abundance and distribution of several species of
proper assessment and measurement end point se- aquatic vegetation as established by baseline mea-
lection, and it is important that both the selection surements. The measurement end points could be
rationale and the linkages between measurement growth and reproduction measurements of these
end points, assessment end points, and policy goals species in the laboratory and field.
be clearly stated. More detailed discussions of end
points and selection criteria can be found in Suter Analysis
[4,5], Kelly and Harwell [6], U.S. Department of The analysis phase of ecological risk assessment
the Interior [7], and EPA [8]. consists of the technical evaluation of the data on
The initial evaluation of stressors, the ecosystem the potential effects and exposure of the stressor.
potentially at risk, and ecological effects are inte- The analysis phase is based on the conceptual
grated with the end points to develop a conceptual model developed during problem formulation. Al-
model for the assessment. The conceptual model though this phase consists of characterization of
describes how the stressor might affect ecological ecological effects and characterization of exposure,
components of the natural environment [9]. For the dotted line in Figure 1 illustrates that the two
example, the stressor may cause adverse effects by are best performed interactively. An interaction be-
interacting directly with an ecological component. tween the two elements will ensure that the charac-
A stressor may also cause adverse effects indirectly, terized ecological effects are compatible with the
for example, by affecting the food or habitat on biota and exposure pathways identified in the ex-
which the ecological component of interest depends. posure characterization. The outputs of ecological
The conceptual model includes a description of effects characterization and exposure characteriza-
possible exposure scenarios, which are qualitative tion are summary profiles that are used in the risk
descriptions of how the various ecological compo- characterization phase.
nents co-occur with or contact the stressor. Each It is important to describe clearly and estimate
scenario is defined in terms of the stressor, the type quantitatively the assumptions and uncertainties in-
of biological system and principal ecological com- volved in both analysis steps. In the majority of as-
ponents, how the stressor will contact or interact sessments, data will not be available for all aspects
with the system, and the spatial and temporal of these analyses, and those data that are available
scales. Finally the conceptual model also describes may be of questionable or unknown quality. Typ-
the approaches, analyses, and data needed to con- ically, the assessor will have to rely on a number of
duct the assessment. assumptions with varying degrees of uncertainty as-
Although there may be many ways that a stres- sociated with each. These assumptions will be
sor can cause adverse effects, only those that are based on a combination of professional judgment,
1668 S.B. NORTONET AL.

inferences based on analogy with similar chemicals area per time. For physical disturbance, the expo-
and conditions, estimation techniques, and so sure profile may be expressed in other terms, such
forth, all of which contribute to the overall uncer- as percentage of removed habitat or the extent of
tainty. The uncertainties in these two steps are flooding per year.
brought forward and summarized during risk In our nutrient loading example, exposure may
characterization. be characterized by modeling or measuring the nu-
Characterization of exposure. The objective of trient concentrations in the parts of the estuary
the exposure characterization is to combine the spa- suitable for plant growth, say, in depths of 2 m or
tial and temporal distributions of both the ecolog- less. The temporal and spatial variation would also
ical component and the stressor to evaluate the be addressed, and the uncertainty in the measure-
co-occurrence or contact between the ecological ments and model would be discussed.
component and the stressor. The way exposure is Characterization of ecological effects. The re-
characterized will depend on the stressors being lationship between the stressor and the assessment
evaluated and the assessment and measurement end and measurement end points identified during
points. In the case of physical alterations of com- problem formulation is analyzed in characteriza-
munities and ecosystems, exposure can be broadly tion of ecological effects. The analysis focuses on
expressed as co-occurrence, for example, the co- describing the relationship between the amount of
occurrence of a wetland community with fill ma- stressor and the magnitude of ecological effects
terial. Exposure analyses of individuals often focus elicited. Any extrapolations from measurement end
on actual contact with the stressor, as organisms points to assessment end points are also conducted
may not contact all of the stressor present in an during this phase. Finally, the evidence for a causal
area. For chemical stressors, the analyses may fo- relationship between the stressor and the measure-
cus further on the amount of chemical that is bio- ment and assessment end points is evaluated.
available, that is, available for uptake by the Data from both field observations and experi-
organism. Some chemical exposure analyses also ments in controlled settings can be used to evalu-
follow the chemical within the organism’s body and ate ecological effects. Controlled laboratory and
estimate the amount that reaches the target organ. field tests (e.g., mesocosms) can provide strong
In order to estimate exposure appropriately, the causal evidence linking a stressor with a response
temporal and spatial scale of the stressor distribu- and can also help discriminate between multiple
tion must be compatible with that of the ecologi- stressors. Data from laboratory studies tend to be
cal component. A temporal scale may encompass less variable than those from field studies, but be-
the life span of a species, a particular life stage, or cause environmental factors are controlled, re-
a particular cycle, for example, the long-term suc- sponses may differ from those in the natural
cession of a forest community. A spatial scale may environment. Observational field studies (e.g.,
encompass a forest, a lake, a watershed, or an en- comparison to reference sites) provide environmen-
tire region. Stressor timing relative to organism life tal realism that laboratory studies lack, although
stage and activity patterns can greatly influence the the presence of multiple stressors and other con-
occurrence of adverse effects. Even short-term founding factors (e.g., habitat quality) in the nat-
events may be significant if they coincide with crit- ural environment can make attributing observed
ical life stages. Periods of reproductive activity may effects to specific stressors difficult.
be especially important, because early life stages are The test data are used to quantify the relation-
often more sensitive to stressors and adults may ship between the amount of the stressor and the
also be more vulnerable at this time. magnitude of the response, and to evaluate the
The product of the characterization of exposure cause-effect relationship. Ideally, the stressor-
is an exposure profile that quantifies the magnitude response evaluation quantifies the relationship be-
and the spatial and temporal pattern of exposure tween the stressor and the assessment end point.
for the scenarios developed during problem formu- When the assessment end point can be measured,
lation. Exposure profiles can be expressed by using this analysis is straightforward. When it cannot be
a variety of units. For chemical stressors operating measured, the relationship between the stressor and
at the organism level, the usual metric is expressed measurement end point is established first, then ad-
in dose units, for example, milligrams per kilogram ditional extrapolations, analyses, and assumptions
body weight per day. For higher levels of organi- are used to predict or infer changes in the assess-
zation, such as an entire ecosystem, exposure may ment end point.
be expressed in units of concentration per unit of Commonly used extrapolations include those
 Framework for ecological risk assessment at EPA 1669

between species, between responses, from labora- points on the curve are identified, the difference in
tory to field, and from field to field. Differences in magnitude of effect at different exposure Ievels can
responses among taxa depend on many factors, in- be reflected in risk characterization.
cluding physiology, metabolism, resource utiliza- In our example, the characterization of ecolog-
tion, and life history. The relationship between ical effects may include stressor-response curves
responses also depends on many factors, including from the laboratory or field that relate the concen-
the mechanism of action and internal distribution trations of nutrients to changes in growth and re-
of the stressor within the organism. When extrap- production of the plant species of interest. Studies
olating between different laboratory and field set- of nutrient loads to other, similar estuaries and the
tings, important considerations include differences associated response of vegetation may also be sum-
in the physical environment and organism behav- marized. The uncertainties and assumptions (e.g.,
ior that will alter exposure, interactions with other that response in the field is the same as that in the
stressors, and interactions with other ecological laboratory) would also be summarized.
components.
In addition to these extrapolations, an evalua- Risk characterization
tion of indirect effects (e.g., effects on food, hab- Risk characterization is the final phase of eco-
itat, or competing species), effects at other levels of logical risk assessment. The profiles of exposure and
organization, effects at other temporal and spatial ecological effects serve as input to risk character-
scales, and recovery potential may be necessary. ization whenever risks are estimated and described.
Whether these analyses are required in a particular In the first step of risk characterization, risks
risk assessment will depend on the assessment end are estimated by integrating the exposure and ef-
points identified during problem formulation. The fects data to yield an expression of the likelihood
need for these types of analyses may also be iden- of adverse effects occurring as a result of exposure
tified during risk characterization after an initial to a certain stressor. Depending on the type of
evaluation of risk. data, the risk may be expressed in a qualitative or
Another important aspect of the characteriza- quantitative fashion. The integration may be per-
tion of ecological effects is to evaluate the strength formed by comparing single exposure and effect
of the causal association between the stressor and values, by comparing distributions within the ex-
the measurement and assessment end points. This posure and effect profiles, or through the use of
information supports and complementsthe stressor- simulation models. The nature of the data and the
response assessment and is of particular importance requirements of the risk assessment will largely de-
when the stressor-response relationship is based on termine which method or combination of methods
field observations. An evaluation of causal evidence will be used. Another important activity associated
augments the risk assessment and contributes to the with this step is the discussion of uncertainties en-
weight of evidence analysis supporting the judg- countered during problem formulation, analysis
ment that a causal relationship exists between the phase, and risk characterization. Uncertainties arise
stressor and response. Many of the concepts applied due to data and knowledge gaps, and the assessor
in human epidemiology can be useful for evaluat- must often use assumptions to bridge these gaps.
ing observational field studies [lo]. An example Although such assumptions are necessary, their use
of ecological causality analysis was provided by often leads to uncertainty in the final assessment,
Woodman and Cowling [ll], who evaluated the which has to be acknowledged through a qualita-
causal association between air pollutants and injury tive or quantitative uncertainty analysis.
to forests. After the risks and uncertainties have been es-
The results of the characterization of ecological timated, the assessor summarizes the results and
effects are summarized in a stressor-response pro- discusses the overall confidence in the risk assess-
file that describes the stressor-response relation- ment. This is achieved by objectively considering
ship, any extrapolations and additional analyses, the sufficiency of the data, evidence of the cause-
and evidence of causality (e.g., field effects data). and-effect relationship, and any ancillary data in
For practical reasons, the results of the stressor- a weight-of-evidence evaluation. The objective is to
response analysis are often summarized as one ref- describe the risk in terms of the assessment end
erence point, for instance, a 48-h LCSO. Although point identified in the problem formulation phase.
useful, such values provide no information about Without this crucial connection, the results of the
the slope or shape of the stressor-response curve. risk assessment may not be readily apparent to the
When the entire curve is used, or when several risk manager. In addition, and perhaps most im-
1670 S.B. NORTONET AL.

portant, the assessor provides an interpretation of continuing the process into the second and third
the ecological significance of the identified risks. phases of guidelines development.
Ecological significance may be described in terms The second phase of the strategy focuses on ac-
of the spatial and temporal extent of the effects; the quiring information on a series of guidance-issue
nature and magnitude of the effects; and, when areas that have been identified during several work-
possible, an estimation of the recovery potential shops as essential to the development of a guideline
once the stressor is removed. Depending on the ob- [12,13]. The second phase of the program will pro-
jectives of the assessment, the risks may also be duce a series of resource reports, comprised of one
placed in a broader ecological context by discuss- or more scientific white papers for each of the guid-
ing the implications of the effects to other compo- ance-issue areas. In addition, a suite of problem-
nents of the ecosystem. oriented case studies will be developed to illustrate
The risk characterization phase of our nutrient the application of the guidelines.
loading example would begin by integrating the nu- The third phase of the program involves the in-
trient levels measured or modeled in the estuary tegration of the white papers, the problem-oriented
with the stressor-response curves. These could be examples, and the “Framework Report” into the
integrated simply by comparing different points first EPA-wide guidelines for ecological risk assess-
along the stressor-response curves with the concen- ment. The goal of this plan is to use the framework
trations anticipated in the field. Alternatively, the as the platform for developing Agency ecological
stressor-response and exposure information could risk assessment guidelines by integrating new sci-
be integrated by using a model that would simulate entific information from each of the issue areas and
the changes in the plant community resulting from case studies.
the nutrient load. The uncertainties of the assess-
ment, including those from the problem and anal- Guidance-issue areas
ysis phases, would be described, and the overall It became apparent during the process of devel-
confidence in the assessment would be discussed. oping the “Framework Report” that there were a
The nature, magnitude, and spatial and temporal number of important scientific issues for which
extent of effects would also be discussed. Finally, additional information and research would be
the effects on the plant community might be placed needed before ecological risk assessment guidelines
in a broader ecological context by discussing the could be developed. The following are the types
implications for the wildlife and fish that depend of issues and needs that emerged from workshop
on the plants. discussions:
Risk characterization forms the basis for a dis-
Scale- the issues related to spatial, temporal,
cussion of the results with the risk manager. The
and biological scale
risk assessment is used by the risk manager, along
Stressors- the need to define exposure for non-
with economic, legal, and social concerns in the
chemical stressors and multiple stressors
risk management process, to evaluate management
End points -the importance of identifying the
options. A consideration of the basic principles of
ecosystems and selecting the end points po-
ecological risk assessment will contribute to a final
tentially at risk
product that is both credible and germane to the
Ecological effects- the need for information on
needs of the risk manager.
estimating direct, indirect, and cumulative ef-
The “Framework Report” is the first step in a
fects across biological scales
long-range effort to improve ecological risk assess-
Recovery- the importance of measuring the po-
ments within the EPA. In the short term, the basic
tential for ecosystem recovery
principles provided in the report are intended to
Variability -the incorporation of estimates of
foster a consistent approach for terminology in and
natural variability into assessments
conduct of risk assessments. The framework, as
Ecological significance - providing information
part of a longer term guidelines development pro-
on the ecological significance of change
cess, will serve as a basis for identifying topics in
Uncertainty - treating uncertainty explicitly in
future guidelines activities.
risk assessments
FUTURE DEVELOPMENT OF EPA’S Causality - information on how to determine
ECOLOGICAL RISK ASSESSMENT GUIDELINES causality in heterogeneously stressed envi-
ronments.
The publication of the “Framework Report”
completes the first phase of the guidelines develop- Two other recommendations emerged from dis-
ment process. This section describes a strategy for cussions. First was the need for case studies to il-
 Framework for ecological risk assessment at EPA 1671

lustrate the risk assessment process as applied to become the primary resources for the Agency’s eco-
various types of problems. Second was the need for logical risk assessment guideline work groups. The
risk management guidance that both identifies the result will be a series of ecological risk assessment
context within which the risk assessment resides guidelines that will include descriptions of principles
and discusses the interfaces between risk assessment and concepts, provide generic guidance, include is-
and risk management. sue- and problem-oriented resource volumes, and
contain specific applications of the risk assessment
Problem-oriented case studies process.
The strategy proposed for developing problem- The first document proposed for this phase is a
oriented case studies has two components: criteria general ecological risk assessment guideline that
for selection and a format for the analysis of the will provide specific guidance for the scientific is-
case studies. The review and analysis of each case sues associated with conducting ecological risk as-
study will follow the three-phase process outlined sessments as outlined in the “Framework Report.”
in the framework. Specific attention will be given This approach is consistent with the scientific com-
to assessing the applicability of the process outlined munity’s recommendations to develop an initial
in the framework document and to the guidance guideline that focuses on the scientific issues inte-
areas discussed above. The case studies will also gral to the risk assessment process. The proposed
provide valuable information on analytical ap- guideline, therefore, represents an expansion of the
proaches, methodologies, and the scientific feasi- principles and criteria developed in the framework.
bility of conducting risk assessments for particular The intention is that the ecological risk assessment
problem settings. guideline provide detailed guidance and a range of
The criteria used to select the case studies in- problem-oriented illustrations for each stage of the
clude basic organizing principles (stressor type, risk assessment process. This approach assures that
level of ecological organization, ecosystem type, the guideline will focus on the important scientific
and spatial and temporal scale), the guidance-issue issues and provide guidance on the use of the guide-
areas, and regulatory needs. For example, case line for specific problems.
studies could be selected to illustrate toxic, non-
toxic chemical, and nonchemical stressors operat- SUMMARY
ing at widely different spatial and temporal scales Increased awarcness of ecological issues has
in different types of ecosystems. emphasized the need for improved ecological risk
The value of the case studies to the ecological assessment methodology. The ecological risk as-
risk assessment guideline process can be viewed sessment process will evolve as new ideas and re-
from several perspectives: search advancements improve our basic knowledge
Framework application- illustrate the applica- about how ecological components interact and how
tion of the description of the risk assessmentstressors alter such interactions.
process to different types of problems The EPA has taken the first steps to develop
Evaluation - evaluate the adequacy of the prin- agencywide guidance for conducting ecological risk
ciples and concepts used in the framework assessments, continuing a long-standing EPA pro-
Guidance areas -provide state-of-the-science in- gram to make the risk assessment process more sys-
formation on specific guidance issues being tematic. Inevitably, it will be a multiyear effort,
developed in parallel white papers just as it has been in the case of human health risk
assessment, where development of those guidelines
Feasibility - determine the scientific feasibility
(models, methods, etc.) for conducting risk has been a product of several years of review and
assessments in a variety of problem settings discussion involving scientists and policy makers.
Agency needs - provide specific examples of Publication of the “Framework Report” now
risk assessment to illustrate a spectrum of provides a structure on which future guidelines can
Agency regulatory needs. be built. The EPA, howevcr, recognizes that eco-
logical risk assessment is a rapidly developing sci-
Preparation of ecological risk ence driven by a desire to expand our capabilities
assessment guidelines beyond assessing single chemical effects on the re-
The third phase of the proposed guideline pro- sponses of individual species. The framework has,
gram will involve the preparation of ecological risk therefore, been designed to accommodate a variety
assessment guidelines by the EPA. The issue-ori- of stressors causing a diversity of ecological effects.
ented resource reports, the problem-oriented The framework is not intended to be static and will
resource reports, and the “Framework Report” will continue to evolve as our experience with address-
1672 S.B. NORTONET AL.

ing regional- and global-scale problems increases. REFERENCES

In its present form, the framework, through its dis- 1. U.S. Environmental Protection Agency. 1992. A
cussion of basic principles, operational definitions, framework for ccological risk assessment. EPA
and identification of key issues, will foster consis- 630/R-92-001. Risk Assessment Forum, Washington,
tency within the EPA and assist risk assessors in DC.
2. U.S. Environmental Protection Agency. 1990. Re-
avoiding errors of omission or in pursuing risk as- ducing risk: Setting priorities and strategies for en-
sessment questions that cannot be applied in a reg- vironmental protection. SAB-EC-90-021. Science
ulatory context. Advisory Board, Washington, DC.
Over the next few years, more substantive guid- 3. U.S. Environmental Protection Agency. 1992. Peer
ance will be issued. Although the exact format for review workshop report on a “Framework for ecolog-
ical risk assessment.”EPA 625/3-91-022. Risk Assess-
this guidance has not yet been decided, it will be an ment Forum, Washington, DC.
expansion of the principles discussed in the frame- 4. Suter, G.W. 11. 1990. Endpoints for regional ecolog-
work with emphasis being given to the specific el- ical risk assessments. Environ. Manage. 14:9-23.
ements of the risk assessment process, as described 5. Suter, G.W. 11. 1989. Ecological endpoints. In W.
Warren-Hicks, B.R. Parkhurst and S.S. Baker, Jr.,
by the framework. Just as in the case of the human eds., Ecological Assessments of Hazardous Waste
health risk guidelines, these guidelines will not be Sites: A Field and Laboratory Reference Document.
rigid and will encourage the use of professional EPA 600/3-89-013. U.S. Environmental Protection
judgment within the framework of a logical and Agency, Corvallis, OR, pp. 2-1-2-28.
scientifically sound structure. Development of risk 6. Kelly, J.R. and M.A. Hanvell. 1990. Indicators of
ecosystem recovery. Environ. Manage. 14527-546.
assessment guidelines has historically provided an 7. U.S. Department of Interior. 1987. Injury to fish and
opportunity for discussion and debate among sci- wildlife species. Type B Technical Information Doc-
entists, policy makers, and the public as to what ument. CERCLA 301 Project. Washington, DC.
society values and what level of protection is suf- 8. U S . Environmental Protection Agency. 1990. Eco-
logical indicators. EPA 600/3-90-060. Environmental
ficient. This will certainly be true in the case of the Monitoring and Assessment Program, Washington,
ecological risk assessment guidelines, as the Agency DC.
grapples with translating its broad regulatory man- 9. National Research Council. 1983. Risk assessment in
dates into concrete risk assessment policies and the federal government: Managing the process. Na-
objectives. tional Academy Press, Washington, DC.
10. Hill, A.B. 1965. The environment and disease: Asso-
ciation or causation? Proc. R. SOC. Med. 58:295-300.
Acknowledgement-Suzanne Marcy cochaired the work 11. Woodman, J.N. and E.B. Cowling. 1987. Airborne
group that developed early drafts of the “Framework Re- chemicals and forest health. Environ. Sci. Technol.
port.” Other members of the work group included Mi- 21: 120- 126.
chael Brody, David Mauriello, Anne Sergeant, and Molly 12. U.S. Environmental Protection Agency. 1992. Report
Whitworth. Many peer reviewers, both inside and outside on the ecological risk assessment guidelines strategic
the EPA, made valuable suggestions for revising the planning workshop. EPA 630/R-92-002. Risk Assess-
“Framework Report.” Especially noteworthy were the ment Forum, Washington, DC.
participants in the May 1991 peer review workshop, which 13. U.S. Environmental Protection Agency. 1991. Sum-
was chaired by James Fava, with discussions led by Law- mary report on issues in ecological risk assessment.
rence Barnthouse, James Falco, Mark Harwell, and Ken- EPA 625/3-91-018. Risk Assessment Forum, Wash-
neth Reckhow. ington, DC.