Lessons Learned on Planning and Scoping
for Environmental Risk Assessments
Prepared by the Planning and Scoping Workgroup of the
Science Policy Council Steering Committee
U.S. Environmental Protection Agency
Washington, DC
January 2002
-------
Lessons Learned on Planning and Scoping
for Environmental Risk Assessments
Prepared for the U.S. Environmental Protection Agency
by the Planning and Scoping Workgroup of the
Science Policy Council Steering Committee
Workgroup Members
Michael Firestone (OCHP) and Edward S. Bender (SPC Staff), Co-Chairs
George Bollweg (Region 5) Wanda Jakob (OPP)
Gerald Carney (Region 6) David Guinnup (OAQPS)
Nader Elkasabany (OPP) Allen Vaughan (OPP)
James Rowe (SPC staff) William Wood (ORD)
Jack Fowle (OA) Jan Young (OSW)
U.S. Environmental Protection Agency
Washington, DC
January 2002
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental Protection
Agency policy and approved for publication and distribution. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
-------
Table of Contents
Acknowledgments v
Executive Summary vi
Chapter 1 Purpose, Overview and Organizational Focus 1
1.1 Background and Purpose 1
1.2 Overview of the SPC's '97 CR Planning and Scoping Guidance 1
1.2.1 Determine the overall purpose and risk management objectives for the risk
assessment 1
1.2.2 Determine the scope, the problem statement, participants, and resources
for the assessment 2
1.2.3 Determine the risk dimensions and technical elements that may be
evaluated in the assessment 2
1.2.4 Formulate a technical approach including a conceptual model and an
analysis plan for conducting the assessment 2
1.3 History and Related Current Agency Activities 3
1.4 Key organizing issues 3
1.4.1 Risk Expert - Decision Maker Dialogue 3
1.4.2 Stakeholder Involvement 4
1.4.3 Planning and Resources 4
1.4.4 Defining the Scope 5
1.4.5 Conceptual Model 5
1.4.6 Analysis Plans 6
1.5 Case Studies in this Handbook 7
Chapter 2 Highlights from Case Studies 9
2.1 Risk Assessment and Risk Management Dialogue 10
2.2 Planning and Resources 10
2.3 Stakeholder Involvement 11
2.5 Development of a Conceptual Model 12
2.6 Analysis Plan 13
2.7 Planning and Scoping of National versus Place-based Assessments 13
2.8 Basic Lessons 14
Chapter 3. References 16
Appendix A. Risk Assessment Terminology A-l
Appendix B. Case Study on Concentrated Animal Feeding Operations B-l
B. 1 Background History B-l
B.2 Highlights and Key Findings B-l
B.2.1 Risk Assessor - Risk Manager Dialogue B-l
ii
-------
B.2.2 Defining the Scope B-2
B.2.3 Planning and Resources B-2
B.2.4 Stakeholder Involvement B-2
B.2.5 Development of a Conceptual Model B-3
B.2.6 Analysis Plan B-7
B.3 Lessons Learned B-7
Appendix C. The Reregistration of Pentachlorophenol: Case Study C-l
C.I. Background/history C-l
C.2 Purpose C-l
C.3. Highlights and Key Findings C-2
C.3.1 Risk Assessor-Risk Manager Dialogue C-2
C.3.2 Defining the scope C-2
C.3.3 Planning and Development of a conceptual model C-2
C.3.4 Stakeholder Involvement C-5
C.3.5 Lessons Learned C-7
C.4 Details on Conceptual Models for PCP C-9
C.5 Narrative for the Human Health Conceptual Models for Pentachlorophenol (PCP)
and Its Contaminants C-13
Appendix D. Cumulative Risk Initiative (CRI) for Cook County IL and Lake County IN
(formerly Chicago Cumulative Risk Initiative, CCRT) D-l
D.I. Background and History D-l
D.2. Highlights and key findings D-2
D.2.1 Stakeholder involvement D-2
D.2.2 Planning and scoping, conceptual model, analysis plan, resource
considerations D-2
D.3 Risk Assessor - Risk Manager Dialogue D-4
D.4. Lessons Learned D-5
Appendix E. Planning and Scoping for the National-Scale Assessments E-l
E.I The National Air Toxics Program Assessment E-l
E. 1.1 Purpose for this assessment E-l
E. 1.2 Stakeholder Involvement E-l
E. 1.3 Details of the Conceptual Model E-2
E. 1.5 Analysis Plan for Cumulative Risk Assessment and Characterization . . E-5
E. 1.6 Lessons Learned E-5
in
-------
E.2 RCRA Surface Impoundment Study-Technical Plan for Human Health and
Ecological Risk Assessment E-6
E.2.1 Background E-6
E.2.2 Purpose and objectives for the study E-7
E.2.3 Phase I (Conceptual Models and Analysis Plan) E-7
E.2.4 Phase II (Assessment Plan) E-9
E.2.5 Internal and External Stakeholder Process E-13
E.2.6 Lessons Learned E-14
IV
-------
Acknowledgments
This document reflects contributions of many individuals whose talents and toils we
gratefully acknowledge.
The case studies mentioned in this document are works in progress, not final resolutions.
These individuals deserve special thanks for sharing their work and for responding to comments
from their peers at the practica. Nader Elkasabany, Wanda Jacob, and Allen Vaughan (Office of
Pesticide Programs) presented and developed the case for Pentachlorophenol. Carole
Braverman presented and George Bollweg (both from Region 5) developed the case for the
Cumulative Risk Initiative. Gerald Carney (Region 6) presented and developed the case for
Concentrated Animal Feeding Operations. We are also indebted to Joseph Swick, a Region 6
risk manager, who defined risk management needs for the CAFO case in our third practicum.
David Guinnup and Roy Smith (Office of Air Quality Planning and Standards) provided figures
and background for the discussion of the National Air Toxics Assessment. Jan Young and
Elizabeth Cuthbertson (Office of Solid Waste) provided figures and detailed information for the
discussion on the Surface Impoundment Study and feedback from their Science Advisory Board
review.
The work on case studies proceeded through practica (practical applications of the
planning and scoping guidance) which were conceived by Dorothy Patton. Drs. Mark Harwell
and Jack Gentile, University of Miami, provided instruction based on their experience with the
ecological risk assessment process at two practica. Mary McCarthy-OReilly orchestrated the
venue and logistics for the practica that stimulated healthy discussions. During the course of
three practica, over 100 participants applied the guidance to case studies. Their thoughtful
inquiry, patience, and perseverance has improved the planning and scoping methods and added
details, beyond the initial guidance, which we present here.
The document was also improved by thoughtful suggestions on stakeholder involvement,
terminology, and presentation from many individuals including: Donald Barnes, David Bennett,
Patricia Bonner, Michael Callahan, Kerry Dearfield, Deirdre Murphy, Jeff Morris, and Anne
Sergeant.
Our sincere thanks,
The Planning and Scoping Workgroup Co-Chairs
Michael J. Firestone, Ph.D. Edward S. Bender, Ph.D.
Office Children's Health Protection Office of Science Policy
v
-------
Executive Summary
The purpose of this handbook is to provide early feedback to agency scientists and
managers regarding our experiences with planning and scoping as the first step in conducting
environmental assessments since the 1997 "Guidance on Cumulative Risk Assessment - Part 1.
Planning and Scoping was released (http://www.epa.gov/ordntrnt/ORD/spc/cumrisk2.htm). This
handbook is meant to reinforce the concept that formal planning and dialogue prior to the
conduct of an environmental assessment can improve the final assessment product in terms of
relevancy to an environmental decision and addressing the concerns of decision makers,
scientists, economists and stakeholders (where applicable). This handbook is also meant to be a
catalyst to encourage agency managers to adopt formal planning and scoping as part of EPA's
culture, especially when conducting significant and/or unique environmental assessments. While
this handbook is primarily intended to assist agency managers and scientists, it is hoped that its
"lessons learned" can also be informative for anyone involved directly or indirectly in the
process of developing environmental assessments. It is important to recognize that this
handbook does not represent rigid rules which must always be followed, but rather helpful ideas
for improving our efforts toward assessing environmental problems.
The figure below depicts how planning and scoping fits into the iterative process to
assess and manage environmental risks.
Risk Assessment/Management Decision Process
New Management Needs
Planning
And
Scoping
(Technical,
Stakeholder,
And Manager
Dialogue)
Economic, Poll-Science,
and Social Analysis
VI
-------
This handbook provides lessons learned from case studies evaluated during a series of
practica following release of the 1997 guidance cited above. It is organized around the following
key steps:
risk expert - decision-maker dialogue (management objectives)
stakeholder involvement
planning and resource considerations
defining the scope of an environmental assessment
development of a conceptual model
production of an analysis plan
The order in which these steps are taken during any specific assessment may be influenced by
the "driver" - for a national assessment, the driver may be a new law which requires up-front
agency planning, while a local place-based problem may very well be driven by local
stakeholder concerns and involvement. General characteristics for these steps are summarized in
the table below for place-based and national assessments:
Steps
Place-based Characteristics
National Characteristics
Stakeholder Involvement
Focuses on diverse groups: a)
the affected/interested public
and b) regulated parties
Tends to be dominated by
expert and advocacy opinions
through formal processes.
Defining the Scope and
Problem Formulation
Broad discussion, amenable
to public concerns and issues.
Legal basis must be satisfied.
Additional issues based on
technical concerns.
Resources and Planning
Provide public education on
problem and process.
Accommodate and support
public participation.
Provide technical forum, add
legal and facilitation support
if needed.
Development of a Conceptual
Model
Extensive community input
helps refine exposure
scenarios and health
concerns.
Technical and legal input
tends to follow the regulatory
framework for managing
risks.
Risk Management Objectives
Externally driven, multi-
agency responsibilities.
EPA proposals, modified
based on comments.
Analysis Plans
EPA sets ground rules and
definitions, public expands
content
EPA, regulated and affected
parties negotiate.
The case studies evaluated in this handbook include:
vn
-------
> Registration of Pentachlorophenol under FIFRA (PCP)
> General water permit conditions for a Concentrated Animal Feeding Operations (CAFO)
> Cumulative Risk Initiative for citizen petitions under TSCA (CRT)
» National Air Toxics Assessment of hazardous air toxics (NAT A)
> RCRA Surface Impoundment Study to screen for hazardous wastes (SIS)
Key lessons gleaned from the subject case studies include:
1. Early and extensive involvement of the risk manager (decision maker) helped focus the
process toward a tangible product.
2. Purporting that planning and scoping will be quick and easy is likely to be
counterproductive; it is a lot more work than people assume. However, it ultimately
saves time by helping to organize everyone's thinking and should result in a better
quality assessment.
3. Stakeholder engagement is essential at the beginning, because their patience is directly
proportional to their sense of influence in the process. They have been helpful in
identifying important public health endpoints that were not initially considered by EPA in
the process of developing a conceptual model.
4. Conceptual models are helpful in demonstrating how one program relates to other
regulatory activities as well as the relationships between stressors and effects beyond
traditional regulatory paradigms.
5. Debate over terminology and brainstorming sessions was necessary to reach a consensus
in the practica. A clear set of definitions would aid this process.
6. The planning and scoping process cannot be prescriptive, because the context of each
situation is different. Planning and scoping is particularly valuable when the assessment
will be complex, controversial, or precedential. At this time, planning and scoping
should precede cumulative risk assessments.
7. Clear objectives, resource commitments, and estimated schedules from management will
drive the approach and level of detail that can be considered.
8. Explaining uncertainty to stakeholders is critical despite a hesitancy to reveal all that is
known and not known about chemicals risks. While revealing these uncertainties may
lead to criticism and political ramifications, it can also develop a sense of trust,
credibility, and support for the decision making process.
Vlll
-------
Chapter 1 Purpose, Overview and Organizational Focus
1.1 Background and Purpose
This document supplements the "Guidance on Cumulative Risk Assessment. Part 1.
Planning and Scoping," July 1997 (hereafter referred to as the 1997 Guidance), and is intended
for use by EPA/non-EPA individuals who are interested in the application of planning and
scoping to environmental risk assessment and risk management decision making.
Following release of the 1997 Guidance, EPA held a series of practica utilizing several
case studies to illustrate the planning and scoping process. The cases considered during the series
were: Big Darby Creek (first practicum only); Pentachlorophenol; Cumulative Risk Index
Analysis of a Concentrated Animal Feeding Operation; and the Cumulative Cumulative Risk
Initiative. The attached case studies illustrate applications of the 1997 Guidance and lessons
from that experience. As the 1997 guidance has received wider circulation, additional questions
have arisen about its implementation, applications to other assessments and risk management
decisions, documentation of the results, and approaches to dealing with stakeholders. The
practica and various case studies have increased our experience and understanding of the
techniques so that lessons have been learned that will guide future applications of the planning
and scoping process. This document is NOT intended to provide guidance for analyses of
cumulative or other risks; guidelines and specific program guidance serve that function.
1.2 Overview of the SPC's '97 CR Planning and Scoping Guidance
The 1997 Guidance was developed as much to raise awareness of the need to consider the
cumulative risks of exposure to multiple stressors as it was to highlight the importance of a
Planning and Scoping dialogue with the assessment and management of risk. Risk assessments
use scientific analyses to inform risk management decisions. The analysis may be used to screen
candidate stressors or sources for possible adverse health or environmental effects, evaluate
planned actions or existing activities and stressors, or support licensing or permits for product
development and treatment of the waste streams. Planning and scoping occurs before the risk
assessment begins. During planning and scoping, risk experts (including those involved in
assessing risk such as ecologists, toxicologists, chemists, along with other technical experts such
as economists and engineers) and decision makers work together as a team, informed by
stakeholder input, to develop the rationale and scope for the risk assessment and
characterization. Listed below are the key steps identified in the 1997 guidance. Although these
steps are shown as a sequence, each step may go through several iterations as additional
information is gathered.
1.2.1 Determine the overall purpose and risk management objectives for the risk
assessment.
The dialogue between the decision maker and risk experts begins with a discussion on
risk management objectives and information needed to manage risks for a particular case. The
1
-------
risk experts and decision maker work as a team to determine how to obtain technical input,
whether and how to involve stakeholders, and develop a conceptual model and an analysis plan
for the risk assessment. The purpose and risk management objective guide the risk assessment
and data collection and establish some guidelines for estimating resources.
1.2.2 Determine the scope, the problem statement, participants, and resources for the
assessment.
The boundaries of the problem help define the scope. For example, does the risk occur in
a local community or nationally? Problem statements describe the problem or risk situation to
be investigated in the risk assessment. Participants should include those with appropriate
technical expertise (e.g., scientists, economists, and engineers) and stakeholders (i.e., interested
and affected parties) as part of a risk assessment team. Available resources and the schedule for
a decision define the resources and time that can be expended to obtain data and analyze the
information.
1.2.3 Determine the risk dimensions and technical elements that may be evaluated in
the assessment.
The 1997 guidance describes a cumulative risk outline covering the stressors, sources of
stressors, environmental pathways, routes of exposure, time frames for exposures, populations
exposed, and effects and specific elements to consider for each dimension. This outline is a list
of possibilities for what elements will and will not be addressed in the risk assessment. Gaps in
knowledge may be filled with assumptions, estimates, or default values. Feasability of obtaining
data, its relevance to the problem or risk management objectives, stakeholder concerns about
risks, cost, and timing may all affect what elements are included. The rationale for inclusion and
exclusion of specific elements should be documented.
1.2.4 Formulate a technical approach including a conceptual model and an analysis
plan for conducting the assessment.
The conceptual model is both a diagram and narrative description of the theoretical
linkages between stressors and adverse effects. It helps demonstrate the plausible cause and
effect relationships and the endpoints of concern for the risk assessment. The model (either data
driven or hypothetical) can show how multiple stressors might be accumulated or how one
stressor may lead to multiple effects. The analysis plan discusses critical data gaps, potential
data sources and their value to the risk assessment and deliberations on how the analysis is
expected to proceed. In cases where an element of risk is likely to be important but cannot be
quantified due to lack of data, the assessor must highlight this deficiency, using professional
judgement or estimates(if possible) to approximate the missing data. Judgements and
approximations must be clearly noted and explained to the relevant risk manager and/or relevant
stakeholder participants in the final risk characterization. The analysis plan also represents an
agreement between the assessor and decision maker about the initial scope and level of effort
that will be applied to the assessment.
-------
1.3 History and Related Current Agency Activities
Since the 1997 Guidance was released, the Agency has engaged in several planning and
scoping activities. In 1998 and continuing forward, the Office of Pesticide Programs began
developing guidance to implement requirements for aggregate exposure and cumulative risk
assessment of pesticides with common mechanisms of toxicity under the Food Quality
Protection Act. The Office of Air and Radiation developed a planning and scoping document
and conceptual models as part of its National Air Toxics Assessment
(http://www.epa.gov/ttn/atw/sab/ natareport.pdf). The concepts of planning and scoping were
derived from concepts that were detailed for Ecological Risk Assessment Guidelines and
illustrated in five watershed case studies developed by the Office of Research and Development
and the Office of Water. In 2000, a draft Framework for Cumulative Risk Assessment was
developed, that described planning and scoping and problem formulation
(http://epa.gov/ncea/raf/frmwrkcra.htm). The Agency's current thoughts on stakeholder
participation were presented in 2000 as an Interim Policy on Public Involvement. The EPA Risk
Characterization Handbook (2000), describes the steps in planning and scoping and the benefits
that may accrue by focusing the assessment, addressing appropriate questions for the risk
management decision, and examining underlying assumptions and alternate hypotheses about the
risks involved (http://epa.gov/ord/spc/rchandbk.pdf).
1.4 Key organizing issues
The key issues associated with planning and scoping are listed as a general sequence of
steps; however, planning and resource considerations pervade each of the issues discussed in this
section. The steps are iterative and interrelated. For example, questions that may arise with the
analysis plan can lead to refinements of the scope or expansion of the participants (see figure).
1.4.1 Risk Expert - Decision Maker Dialogue
Traditionally decision makers and risk experts at EPA (or a state) initiate the process of
planning for a risk assessment. They develop the list of participants based on the issues, risk
management concerns, affected parties, and technical experience. However, in some instances,
the stakeholders request that the agency conduct a risk assessment. Through discussions at the
practica and other shared experiences, it has become clear that when the issues relate to
communities or specific locations, local representatives should be included in the dialogue. The
National Research Council (1996) and the Presidential/Congressional Commission on Risk
Assessment and Risk Management (1997) described a similar kind of dialogue with stakeholders
where values and opinions were discussed in a deliberative phase and data and technical
conditions were debated and evaluated by scientific and technical experts in an analytic phase of
the dialogue. The case studies that follow illustrate several different approaches for conduicting
this dialogue in planning and scoping.
Decision makers help define objectives, schedules, available resources, and approve the
analysis plan for the risk assessment. They must decide whether stakeholder input is needed and
-------
if so, what roles stakeholders might have (e.g., information exchange, develop recommendations,
or develop agreements). Risk experts may facilitate the discussion of the scope and lead
development of the conceptual model and analysis plan. Usually they collect stakeholder data,
identify data requirements, select models and default conditions, and explain the rationale for
these choices.
1.4.2 Stakehol der Involvement
The extent of stakeholder involvement and commitment to process outcome will depend
in part upon their level of interest and their confidence in that process (Glicken, 1999).
Stakeholders can help provide information on their concerns, values, and in the case of
communities and workers, personal data on exposures and life style. Stakeholders also provide
feedback on the relevance and clarity of the risk management objective, scope for the
assessment, timing, conceptual model, and analysis plan. Stakeholders may also provide details
about releases of stressors from sources, their activity and exposure patterns, and concerns of the
community (Folk and Finney, 1992).
Stakeholders may provide technical expertise in hazard and exposure assessment and
technology, as well as, economic, social, political and legal areas. For example, affected parties
may help identify concerns and costs so they can be considered in the problem assessment and
the general deliberation process. With a clearer statement of costs, benefits, uncertainties, and
other implications available to the assessor, experts, and stakeholders, a wider range of risk
assessment options may be characterized or developed, including some that may be more
innovative, more protective, voluntary, and more economical. In addition, exposed communities
or groups can often provide critical information on potential or actual exposure scenarios, health
and/or ecological endpoints, and highly exposed/highly susceptible subpopulations and/or
lifestages that should be considered in the risk assessment. They may also provide invaluable
insights into public values and perceptions on the risk of concern, the preliminary remedial
actions being considered and public acceptance of those remedies.
1.4.3 Planning and Resources
It is important that the planning exercise be a transparent effort so that the basis for the
final environmental decision (and the alternative options, limitations, and approaches considered
but not selected) is clearly understood early in the process by the public and regulated
community. Thus, the reasons to limit the technical scope of the assessment must be stated
explicitly and must include details on limitations of resources, data, the impact of risk elements
on the risk estimate and methods available.
Place-based (e.g., the CRI case study) versus national scale (e.g., the PCP re-registration
case study) planning and scoping exercises will necessarily involve different orientations and
resource requirements (including those for possible stakeholder involvement) in constructing the
conceptual model and analysis plan.
-------
Local versus National Scale Problems
If place-based:
- define the boundary of the problem area
- determine the relevant pollution sources
- identify the stressors of possible concern
- scope out the location of targets
- define specifc lifestages of possible concern for health
assessments
- identify and inform relevant local/regional stakeholder groups,
citizens
If national scale:
- define general and specific subpopulations of concern within the
national boundaries; establish a clear rationale for their inclusion
- define the stressors and their sources
- identify indicators of human health, ecological effects (e.g.,
epidemiology data, USGS trend data)
- identify and inform the relevant national stakeholder groups
1.4.4 Defining the Scope
Defining the scope -
what's in and what's not - is
based on the six dimensions of
Cumulative Risk (Population at
Risk; Stressors; Sources; Routes
and Pathways of Exposure;
Endpoints; and Time Frame. In
the cases studies, potential
elements for each dimension
were developed by brainstorming
and discussions with technical
individuals (scientists,
economists, engineers, and
planners) and stakeholders.
Specific elements were selected
on specific bases, such as
resources; available data; ability
to measure, regulate, or control. The scope describes the currently identifiable context of the
environmental risk that will (or can) be included in the assessment (see text box for examples).
1.4.5 Conceptual Models
To develop an accurate picture of the risks and risk management options, the conceptual
model must explain the elements for each risk dimension; i.e., populations at risk (human,
ecological entities, landscape or geographic concerns), sources of stressors, stressors, pathways
and routes of exposure, assessment endpoints, and time frames of exposure. Stakeholders and
outside expert participants are helpful in exploring the elements in the conceptual model. For a
regulated chemical, for example, the industry representative will usually have the most definitive
information on its chemical synthesis, production and use, which can more completely define the
sources of stressors, potential loadings, and pathways of exposure. Exposed groups or
individuals can confirm or more accurately reflect the qualitative or quantitative aspects of the
exposure pathways, including routes of exposure and the relevant time frames involved in the
proposed model. Stakeholders may suggest alternative methods of looking at the problem that
may allow more flexible approaches to remediation of the risks, the development of additional
conceptual models not originally considered, or novel, non-regulatory solutions to a problem.
The conceptual model should be accompanied by a detailed narrative explaining the
rationale for the elements and their linkage in the conceptual model. The simple diagram of a
generic conceptual model given below illustrates the application of the terms in the 1997
Guidance document. Sources are activities that generate or release stressors. These may include
industry, municipal waste and wastewater treatment, solid waste disposal, transportation,
agriculture, and natural resource management. Stressors are chemical, physical, or biological
-------
Six Questions that define a Risk with examples of Cumulative Risk
Population: Who/What/ is at risk?
-Example-Hispanic toddlers or the process of
nitrogen fixation
Sources: What are the relevant sources of
2.
3.
4.
5.
6.
stressors?
-Example-Auto exhaust or exotic species
Stressors: What are the stressors of concern?
-Example-Lead or overfishing
Pathways/Routes of Exposure: What are the
relevant environmental pathways/routes of
exposure?
-Example- Surface water/Drinking water
ingestion or skin contact
Endpoints: What are the effects due to
exposure? (Assessment/measurement endpoints)
-Example-cancer/estimated number of cases
Time Frames: What are the relevant time
frames of exposure to a stressor or mixtures of
stressors?
-Example-one generation or 40 hours/week
agents that cause an adverse
effect. The stressors move from
the source to the receptors
through pathways (e.g., air or
surface water), where they may
be converted or metabolized in
some way. Exposure occurs in
similar ways for plants and
animals, although ecological
entities, like communities and
ecosystems, are exposed in more
complex ways. Receptors
express the effects of the
stressors, usually in response to
the dose or quantity of stressor
they experience. Under health
endpoints, the generic model
(figure 1-1) adds confounding
factors that contribute to how
the effect is expressed. In the
figure, adverse ecological effects
lead to adverse "quality of life"
effects. Quality of life issues are
often concerns for siting or expanding existing facilities or projects. Although they are beyond
the traditional considerations of risk assessment guidelines, quality of life issues were of great
concern to people in the case studies.
1.4.6 Analysis Plans
The analysis plan is the final stage of planning and scoping before the risk assessment.
The analysis plan identifies data needs, information sources and technical approaches for
evaluating risk hypotheses presented in conceptual models and other important issues identified
during planning and scoping that may be pursued during the risk analysis phase. Those
hypotheses considered more likely to contribute to risk can also be targeted. The rationale for
selecting and omitting risk hypotheses is incorporated into the plan and includes discussion of
data gaps and uncertainties. It also may include a comparison between the level of confidence
needed for the management decision with that expected from alternative analyses in order to
determine data needs and evaluate which analytical approach is best. When new data are
needed, the feasibility and cost of obtaining them can be taken into account. The analysis plan is
strongest when it contains explicit statements for how measures were selected, what they are
intended to evaluate, and which analyses they support. Uncertainties associated with selected
measures and analyses, and plans for addressing them, should be included in the plan when
possible.
-------
1.5 Case Studies in this Handbook
The cases considered during the 1998-9 practicum series were: General water permit
conditions for a Concentrated Animal Feeding Operations (CAFO); Registration of Pentachloro-
phenol under FIFRA (PCP); and Cumulative Risk Initiative for citizen petitions under TSCA
(CRT). Planning for an ecological risk assessment of Big Darby Creek was discussed at the first
workshop, but that watershed case study is not included in this document.
In addition, other cases are included to show additional aspects of planning and scoping:
a) the National Air Toxics Assessment (NATA) which is a national screening activity for risks
from urban air toxic organic chemicals and b) the Resources Conservation and Recovery Act
Surface Impoundment Study(SIS) to screen for cumulative risk from hazardous constituents in
wastewater treatment ponds. The results of the practica case studies are summarized in
Appendices B-D and lessons learned from the other cases are summarized in Appendix E.
The material in the appendices describe how each of the cases addressed the planning and
scoping process. The CAFO case, was developed from an approach that Region 6 developed for
watershed protection. The conceptual model considers cumulative effects from permitting
CAFOs on human health, ecological resources, and quality of life. The PCP case involved
detailed technical and risk management discussions which lead to detailed models and analytical
plans for the risk assessment. Stakeholders were involved from the beginning in the CRI. They
found that planning and scoping helped develop trust between citizen groups and EPA and
commitment to a long term study of community hazards. The additional two national studies,
NATA and SIS, illustrate how analytical plans and detailed models can help analysts, decision
makers and customers focus on the most pressing problems. All of these cases are, in a sense,
works in progress. More lessons will be learned in the future.
-------
Figure 1-1. Conceptual Model
with Examples of Possible Elements and Linkages
Sources
Stressors
Pathways / Exposure!
Routes
Receptors
Ecological
Endpoints
Ecological Endpoints
Condition of Aquatic
Ecosystem
Condition of
Terrestrial/Wetlands
Ecosystems
Populations
environmental
processes
Critical species or
species of special
concern
Human/Societal Endpoints
Cancer
Heart Disease
Respiratory disease
Immunologic disease
Gastrointestinal dis.
Pysch. condition
Farm culture change
Demographic change
Comm. Infrastructure
Aesthetics
Loss of recreation
Property values
-------
Chapter 2 Highlights from Case Studies
The primary goal of planning and scoping is to identify scientific and technical
information and stakeholder concerns about potential environmental risks that is relevant to
inform decisions on risk reduction and management. This is essential in the planning phase to
produce a more focused, cost-efficient risk assessment which is defensible (provides a record of
initial decisions made, approaches taken and parties involved), allows future accountability and
Planning and Scoping Steps
^^^ *~FV^9f» ~rt
Who needs
to be
involved?
* Begin dialogue on nature of concern and analysis to inform risk management decisions
* Identify participants (technical, affected and interested parties).
* Decide what will and will not be included.
* Develop a conceptual model of
potential cause and effects.
An analysis plan.
Figure 2.1 Key Steps and Questions for Planning and Scoping
provides a more rational framework for appropriate revisions as needed. Figure 2.1 above shows
key questions and activities to accomplish the general steps in planning and scoping. The
discussion in this chapter is organized along each of these steps, even though several cases did
not follow all of the steps. The process is iterative and involves feedback and adjustment as new
information is gathered from participants and as decisions are made to refine the scope of the
assessment. It allows for consideration of economic and other data beyond the traditional risk
assessment process.
-------
2.1 Risk Assessment and Risk Management Dialogue
Traditionally, EPA risk assessors have been asked to estimate the risks of one or more
chemicals, risk assessments were performed, and risk managers would then discuss the
assumptions and basis for the assessment, and in many cases, the assessment would be revised to
reflect changes in the scope, assumptions, or problem from the risk manager or from public
comments.
The 1997 Guidance on planning and scoping recommends that the purpose for the risk
assessment be discussed at the outset between the risk assessor and risk manager. From case
studies with communities, we have learned that the risk assessor may actually be a team of
scientists, economists, and engineers from EPA and from stakeholders. Risk managers, in some
cases may be stakeholders and local public officials. This team should plan the process for
defining the problem, conducting the assessment or technical activities, inviting/involving
stakeholder participation, risk management, and evaluation. All projects are more effective if
this team develops a proposal, including tentative schedules that reflect available resources and
time. The team could also delegate or suggest leads for discussions.
Agency guidance, glossaries of technical, regulatory and other terms needing definition,
and public information packages should be developed to guide the participants in the dialogue.
The legal basis for the decision and pertinent policy requirements should also be described.
Planning and scoping is used to frame the activity and the documentation that will be
developed. The scope and level of effort that goes into the document is often bounded by the
purpose or application that is intended by the manager and the legal requirements or authority of
the sponsoring organization. For example, screening assessments require less detail and input
from outside parties and the output is usually needed quickly. For cumulative risk assessments,
problem formulation will set those boundaries. It is important not to initially narrow the focus
based on methods, data, and information that is available at this time, but to allow the scope to
reflect concerns from stakeholders. The problem statement can be revised as the analysis plan is
developed.
2.2 Planning and Resources
The dialogue discussed earlier should define the process and schedule for planning and
scoping. For example, in the RCRA Surface Impoundment Study, the assessment deadline was
established by a consent decree and the planning and peer review steps were established to
assure the regulatory deadline was achieved. During planning and scoping, many choices are
made about the quality of data required for a risk management decision and the scope of the
assessment which may affect the time and resources required for conducting the assessment. In
the CRI case study, the lack of data on human exposure, dwindling resources, and lack of
stakeholder confidence about the assessment of indoor air quality lead to a change in the
approach from a cumulative risk assessment to a cumulative hazard assessment.
10
-------
2.3 Stakeholder Involvement
Depending on the nature and complexity of the issue, planning and scoping may bring
stakeholders into the process during the early deliberations about the problem and on what can
and cannot be included in the risk assessment. If EPA is convening the process for a community,
such as the CAFO case study, stakeholders need some background about the process and how
they may participate. To receive effective stakeholder contributions, we need to get some kind
of commitment from them, if possible, to accept the validity of the results of the assessment,
regardless of what it shows. In some cases, particularly for national rules, individual
stakeholders may not be willing or able to participate in the full process and so it may be
necessary to provide them with technical support, suggest the appointment of a representative or
spokesperson to represent an adversely affected group, and/or provide special access, summaries
or data systems they can consult at their convenience. Websites worked well in the CRI,
providing access to monitoring data. Other regions have used special information and tutorial
sessions preceding public meetings. Professional organizations and members of the regulated
community routinely respond to public notices in newspapers, websites, or the Federal Register.
Stakeholders may also require technical assistance to comprehend agency technical reports or
other data. Agency policy requires that EPA accommodate requests from public citizen groups
wherever possible, e.g., technical assistance grants in the Superfund Program. One lesson from
the CRI is that citizen stakeholders' patience in that case was high because they had a sense that
their comments were being considered in the process.
Stakeholders in community risk assessments can provide invaluable insights about
background and baseline conditions which contribute to risks. They can also provide details
about personal habits and activity patterns which should be considered particularly for patterns
of exposure and mitigation strategies. Stakeholders frequently view the federal government as
one huge entity that has common access to all information and broad authority to control any
activity or facility that is contributing to their potential exposure and adverse health conditions.
Some stakeholders also assume that personal lifestyles (e.g., smoking, health care, exercise, and
diet) have little to do with environmental
risks. The challenge to the Agency is to
get the public to realize the importance
of those personal lifestyle choices and
that an even-handed assessment must
consider all significant sources of risk
which have a significant impact on the
ultimate environmental decision.
Public participation [in planning
and scoping] leads to the incorporation
of new kinds of information in
environmental decision-making and it
has shifted the model from one where the
government defines the process and
pPA's Public Involvement Functions
1. Identify the interested and affected public
2. Provide information and outreach to them
3. Establish public consultation activities
4. Assimilate information and provide feedback
5. Plan and budget for public feedback
6. Consider technical or financial assistance as
needed
(EPA, 2001. Interim Policy on Public
Involvement in Regulatory Decisions)
11
-------
invites stakeholders to participate toward one in which "every affected group participates" (Bear,
1994). EPA should plan activities to support public involvement, provide background on the
problem and EPA perspectives for risk, discuss expectations and needs with stakeholders, and
keep stakeholders informed and involved on the key decisions. These principles have been
formalized in recent Agency policy. Agency staff need advice and training on how to obtain
useful information on risk perceptions, stakeholder concerns and values from stakeholders.
2.4 Defining the scope
It is very helpful to review the definitions of the risk dimensions before brainstorming
begins (i.e., develop as many relevant ideas and approaches as feasible). The group should
include stakeholders or their spokespersons as well as the traditional assessors and other
expertise. In our workshops, exhaustive lists were developed initially, and then criteria were
applied to narrow the list. In more than one instance, we made the mistake of narrowing the list
based on data availability; however, input from actual stakeholders often led to expanded lists of
stressors, sources, or exposure scenarios. Time frames (length of exposure, frequency, etc.) are
part of the working definition of cumulative risk, but they may not apply or be needed for some
assessments.
2.5 Development of a Conceptual Model
Btep^o^Developin^Conceptua^Iodeh^J
1. Brainstorm what could be included.
2. Prioritize the elements for each
dimension.
3. Document reasons for any deletions.
4. Develop linkages among the elements.
Each conceptual model was case
specific. The conceptual model may be a
simple diagram (as for screening) or a
complex, multi-level graphic representation
of the sources, stressors, environmental
pathways, routes of exposure, and receptors.
In several cases, conceptual models were
developed in a hierarchical fashion, with
hyperlinks to show details for technical discussions of the data requirements and hypothetical
cause and effect relationships between stressors and receptor effects.
In practice, the conceptual model is a valuable tool for communication with stakeholders
and as a flowchart for planning the analysis. For some audiences, the broad overview was
sufficient. In the CAFO case, the model included background conditions, showed feedback from
stress on the aquatic ecosystem to secondary impacts on recreation and property values. We also
found that training in special software is needed so models can represent both broad and specific
relationships.
12
-------
2.6 Analysis Plan
The analysis plan is the final stage of planning and scoping and has been adapted from
the Ecological Risk Assessment Guidelines, section 3.5. The plan describes data needs
(qualitative and quantitative), data quality objectives, sampling approaches, and analysis steps
for the risk assessment. The Surface Impoundment Study Technical Plan and the National Air
Toxics Assessment planning and scoping document provide extensive details about how data are
collected, combined, and analyzed for screening risk assessments.
2.7 Planning and Scoping of National versus Place-based Assessments
There are some significant differences in the process for planning and scoping of
community or place-based assessments and national assessments which focus on a sector or
category of sources across the entire country. These differences are important from a number of
standpoints. For example, the general public is usually less interested in the national
assessments, so stakeholders tend to reflect technical concerns and economic interests of national
organizations from the regulated community and environmental interests. Some general
observations gleaned from these case studies and others about the components of planning and
scoping for these broad categories of assessment appear in Table 2.1.
13
-------
Table 2.1. General Characteristics of the Steps in Planning and Scoping for
National and Place-Based Risk Assessments.
Steps
Place-based Characteristics
National Characteristics
Stakeholder Involvement
Focuses on diverse groups: a)
the affected/interested public
and b) regulated parties
Tends to be dominated by
expert and advocacy opinions
through formal processes.
Defining the Scope and
Problem Formulation
Broad discussion, amenable
to public concerns and issues.
Legal basis must be satisfied.
Additional issues based on
technical concerns.
Resources and Planning
Provide public education on
problem and process.
Accommodate and support
public participation.
Provide technical forum, add
legal and facilitation support
if needed.
Development of a Conceptual
Model
Extensive community input
helps refine exposure
scenarios and health
concerns.
Technical and legal input
tends to follow the regulatory
framework for managing
risks.
Risk Management Objectives
Externally driven, multi-
agency responsibilities.
EPA proposals, modified
based on comments.
Analysis Plans
EPA sets ground rules and
definitions, public expands
content
EPA, regulated and affected
parties negotiate.
2.8 Basic Lessons
1. Early input from decision makers and stakeholders is essential Risk assessments need to
identify and evaluate the problems that these groups want to solve. The nature of the decision,
the degree of public concern, the level of scientific understanding, and the complexity of the
issue will profoundly affect the input that is needed and how it can be obtained. Public
involvement policies and procedures should be consulted to assure requirements are met.
2. Stakeholder participation is most beneficial when the participation process,
expectations, and responsibilities for all parties are discussed and accommodated up-front
EPA should develop background materials and plan for stakeholder support if public
involvement is desired. For example, there could be discussion with stakeholders on their roles
(e.g., as advisors or decision makers), possible outcomes from the assessment, and what EPA
plans to provide.
14
-------
3. When the risks are complex, the context for exposure and risk management must be
considered in planning and scoping steps. The approach will likely be tailored by an iterative
process. Sometimes screening or range finding analyses precede formal risk assessments.
Stakeholder input on their potential exposure, diet, and lifestyle can be considered in exposure
scenarios and risk communication.
4. Conceptual models can help reveal assumptions, provide common background and
definitions for participants, and explain choices for the assessment They can be developed
at different levels of detail to explain technical issues and management options and to focus
input and data collection. Conceptual modeling provides opportunities for integrating the
analysis across sources, receptors, and endpoints.
5 Risk managers are critical players in the planning and scoping process In some cases,
the public is the principal risk manager and a key decision maker. Risk managers should clarify
risk management objectives to help formulate the problem for the risk assessment.
6. Planning and scoping has improved individual assessments of risk even though the
analysis may be qualitative, focus on hazard, or be limited to a single stressor The process
is especially useful for complex, controversial, or precedential assessments. Stakeholder input
and discussion helps validate the process for selecting risk management options.
7 Analysis plans that provide roadmaps and data inventories for the risk assessment
inform participants and decision makers as well as risk assessors. Planning and scoping is
especially desirable to use for situations which involve cumulative risk assessments, multiple
stakeholder groups, multiple stressors, a high degree of uncertainty, input and data from multiple
groups, and high costs for analysis.
8. All of the case studies described are works in progress. For some cases, initial stages of
analysis have begun; for other cases, the approach is being refined and applied to new problems.
As risk management decisions are made and implemented, we can use this experience to reflect
further on the role of planning and scoping.
15
-------
Chapter 3. References
1. Bear, D. 1994. Public Participation in Environmental Decision making. American Bar
Association Standing Committee on Environmental Law Washington, DC
2. Finney, C. andR. E. Polk 1992. "Developing Stakeholder Understanding, Technical
Capability, and Responsibility: the New Bedford Harbor Superfund Forum" Environmental
Impact Assessment Review vol 15 p.517-541
3. Glicken, J. 1999. "Effective Public Involvement in Public Decisions" Science
Communication vol 2, no 3 p. 298-327
4. National Research Council. 1983. Risk Assessment in the Federal Government: Managing
the Process. National Academy Press, Washington, D.C.
5. National Research Council. 1994. Science and Judgement in Risk Assessment. National
Academy Press, Washington, D.C.
6. National Research Council. 1996. Understanding Risk: Informing Decisions in a Democratic
Society. National Academy Press, Washington, DC.
7. Presidential/Congressional Commission on Risk Assessment and Risk Management. 1997.
Risk Assessment and Risk Management in Regulatory Decision-Making. Washington, DC.
8. USEPA. 1998. Guidelines for Ecological Risk Assessment EPA/630/R-95/002F April 1998.
9. USEPA. 1997. Guidance on Cumulative Risk Assessment. Parti. Planning and Scoping.
Science Policy Council. July, 1997.
10. USEPA. 2000a. Risk Characterization Handbook. EPA 100-B-00-002. Science Policy
Council. December 2000
11. USEPA. 2000b. Public Involvement Policy. Interim Final. November 2000.
12. USEPA. 2001a. Developing Management Objectives for Ecological Risk Assessments.
Risk Assessment Forum. (Draft).
13. USEPA. 2001b. Framework for Cumulative Risk Assessment. Risk Assessment Forum.
(Draft).
16
-------
Appendix A. Risk Assessment Terminology
This is a partial list of risk assessment terms that are often associated with risk
assessment practice. The list is not exhaustive, but it does include terminology used in this
guidance and other terms that are closely related to the planning and scoping of risk assessments.
Agent-Suter et al. (1994) suggested it as an alternative for the term stressor. It is considered to
be more neutral than stressor, and is used in EPA's Guidelines for Exposure Assessment.
Aggregate exposure - the sum of dietary and residential exposures to pesticide chemical residues
with a common mechanism of toxicity from multiple sources and multiple routes of exposure
(Food Quality Protection Act, 1996).
Analysis- The analytical phase of the risk assessment in which the potential for adverse effects is
calculated based on the hazard identification, dose-response assessment, and the exposure
assessment.
Assessment endpoint- an explicit expression of the actual environmental value that is to be
protected, operationally defined by an ecological entity and its attributes.
Comparative Risk Assessment- A process that generally uses an expert judgement approach to
evaluate the relative magnitude of effects (relative risk) and set priorities among a wide range of
environmental problems (US EPA, 1993b). In some cases this may be done as a preliminary risk
assessment.
Cumulative Risk Assessment- involves the consideration of the aggregate ecologic or human
health risk to the target entity caused by the accumulation of risk from multiple stressors,
[multiple pathways, sources] (US EPA, 1995).
Cumulative effects- 1) the sum of all environmental effects resulting from cumulative impacts
(Liebowitz et al., 1992), and 2) the combination of effects from all pesticide chemical residues
which have a common mechanism of toxicity (Food Quality Protection Act, 1996).
Cumulative impacts-the sum of all individual impacts occurring over time and space, including
those of the foreseeable future (CEQ, 40 CFR Sect. 1508.7)
Conceptual model- a diagram or written description of the predicted key relationships between
the stressor(s) and the assessment endpoint(s) for a risk assessment.
Dimensions of risk- these are components of risk from the 1997 guidance (see USEPA, 1997),
including sources of stress, stressors, pathways and routes of exposure, receptors, and effects.
Disturbance- any event or series of events (such as a physical stressor) that disrupts ecosystem,
community, or population structure and changes resources, substrate availability, or the physical
A-l
-------
environment.
Elements of Risk- these are specific aspects of each dimension that may be included in an
assessment. Elements comprise the scope of the risk that will be described in the conceptual
model and analysis plan (USEPA, 1997).
Environmental Impact Assessment- an assessment required by the National Environmental
Policy Act to evaluate fully potential environmental effects associated with proposed federal
actions.
Exposure-the contact or co-occurrence of a stressor with a receptor.
Integrated Risk Assessment- a process that combines risks from multiple sources, stressors, and
routes of exposure for humans, biota and ecological resources in one assessment with a defined
point of focus (See also cumulative risk assessment).
Receptor-the entity which is exposed to the stressor.
Relative Risk Assessment- a process that involves estimating the risks associated with stressors
or management actions that often uses qualitative risk techniques.
Risk Assessment- a process that evaluates the likelihood that adverse effects such as disease or
injury) may occur as a result of exposure to a chemical, physical, or biological agent.
Source- an entity or action that releases to the environment or imposes on the environment
chemical, biological, or physical stressor or stressors.
Stakeholder - a person, group of people, an organization (public or private), a business, or other
party that has an interest in terms of knowledge or jurisdiction or is affected in terms of their
health, property rights, or economy by an environmental risk(s).
Stressor- Any physical, chemical, or biological entity that can induce an adverse response.
Stress Regime- (1) a characterization of multiple exposures to stressors, (2) a synonym for
exposure, or (3) a series of interactions of exposures and effects resulting in secondary effects.
Because of its potential for confusion, the term is not used in guideline documents.
A-2
-------
Appendix B. Case Study on Concentrated Animal Feeding Operations
B.I B ackground Hi story
Concentrated Animal Feeding Operations (CAFOs) are a common and significant
concern throughout EPA Region 6. CAFOs are large farms (often occupying a quarter square
milesignificant in terms of watershed areas) and they produce enormous quantities of waste
that is discharged into on-site lagoons. These lagoons and associated operations are permitted
under the Clean Water Act's National Pollutant Discharge Elimination System (NPDES) and
require environmental impact reviews under the National Environmental Policy Act (NEPA).
For some watersheds that are not meeting state-prescribed standards, there may be Total
Maximum Daily Load analyses and additional restrictions or penalties imposed. General
statewide permits which cover many CAFOs expired in 1998, and Region 6 wanted to consider
if cumulative impacts from CAFOs and other existing regional sources (agriculture, oil and gas
exploration, roads and transportation infrastructures, and domestic waste) may exceed applicable
water quality standards, pose threats to groundwater supplies, or degrade air quality. The risk
evaluation was requested by Region 6's Compliance Office to meet the NEPA requirement to
review waste lagoons for NPDES permits. In addition, there was public concern over the rapid
expansion of CAFOs and their impact on surrounding communities.
There was no method or approach to determine when a watershed reaches a significantly
polluted state. Region 6 developed Cumulative Risk Index Analysis (CRIA), a novel approach
based upon a mathematical algorithm that established the potential for significant environmental
risk for each CAFO. Cumulative risks are identified through evaluation of: 1) Areas of regulated
and unregulated CAFOs; 2) environmental vulnerabilities (e.g., ground water depth or soil
permeability); and 3) impacts from known CAFO projects (water quality, vector/odor, wildlife
habitats) specific to each water shed subunit.
Watershed Unit Subarea _. f,T , ,.,.. _. -T
^r,T* /rr. i * £c j * TT? . i j x Degree of Vulnerability Degree of Impact
CRIA= (Total Affected Area -^ Watershed area) x / i ,-, - x / i ,-, v.
,,,,,., (scale of 1-5) (scale of 1-5)
(scale of 1-4) v ' ^ '
CRIA facilitates communication of technical and regulatory data upon which better agency
decisions can be made. The CRIA is designed to better understand the effectiveness and results
of CAFO controls. The tool is not intended to be used alone but in concert with other
environmental program perspectives and data (i.e., endangered species and fish and wildlife
service, state environmental agencies with cultural resources' concerns).
B.2 Highlights and Key Findings
B.2.1 Risk Assessor - Risk Manager Dialogue
During the third workshop, a Region 6 risk manager assisted the case presenter. He said
that the risk management objectives for Region 6 were two-fold:
B-l
-------
1) develop new general permits for CAFOs in the state of Oklahoma
2) identify the point where cumulative impacts may exceed the current permit requirements.
B.2.2 Defining the Scope
Region 6's Case Presenter explained that waste lagoons are generally regulated on the
basis of nitrogen/nitrate concentrations, while ignoring phosphate. However, Region 6 (at least)
may be leaning toward regulating phosphate instead. There is a tradeoff between avoiding a
discharge to the atmosphere (the basis of limiting nitrogen) and avoiding a phosphorous buildup
in the soil. There is also a water quality dimension: phosphate runoff to streams and lakes or
percolation of nitrogen to groundwater. How much excess phosphate or nitrogen might result in
human health consequences is not yet known. Also, the total amount of nitrogen being released
from all sources (not just CAFOs) is unknown.
Participants in the third practicum considered several alternatives to define the
geographic scope of the assessment: One was to look at only two counties in the Oklahoma
panhandle for permit renewals; another was to consider the nationwide perspective of risks
anywhere in the country; a third was to focus locally but identify considerations (almost as
"asides") that would apply in other areas. Ultimately, the group decided to focus the risk
assessment on the watersheds affected by CAFOs in a single county in the Oklahoma panhandle
as the basis to develop a risk assessment for all CAFOs in Oklahoma.
B.2.3 Planning and Resources
In the practica, the participants brought different levels of knowledge to each session.
Most were unfamiliar with CAFOs and few had experience with water quality issues. For region
6, planning and scoping has been used to expand this project to a national strategy on permitting
for CAFOs. They have pooled resources with state and county agricultural officials, regional
land-use planners, natural resource agencies, and universities and colleges to strengthen the data
base and analytical capability.
B.2.4 Stakeholder Involvement
Region 6 program managers and staff involved with NEPA enforcement, NPDES
permits, watershed quality, groundwater, surface water, risk assessors, RCRA, Superfund, and
GIS comprised the in-house experts. Stakeholders for this case include academics, industry
(primarily swine production but also beef producers who may have a future stake), state and
county regulators, EPA headquarters (NEPA, agriculture sector), Department of Agriculture's
Natural Resource Conservation Service, national and local environmental groups, and
community residents.
At the third workshop, Dr. Lauren Zeise of the California Environmental Protection
Agency was invited to lead a discussion on applying the lessons from the National Research
B-2
-------
Council's Understanding Risk: Informing Decisions in a Democratic Society. Due to limited
time, participants concentrated on two of Dr. Zeise's questions, 1) who the stakeholders were
and 2) how to engage them. The group recognized the value of stakeholder involvement in the
preparation of the conceptual model, and identified several ideas for expanding stakeholder
involvement in the process:
Sharing the conceptual model with local residents and stakeholders (may need to engage
disadvantaged groups)
Holding scoping meetings with plans and then follow up to the public comments
Providing training opportunities for citizens to follow the project over the long-term
Developing interactive communications tools based on the conceptual model
Region 6 has developed background materials based on the planning and scoping process
for public involvement. They routinely brainstorm with community groups to develop ideas for
conceptual models on the key human health and environmental concerns.
B.2.5 Development of a Conceptual Model
The Concentrated Animal Feeding Operations (CAFO) model includes input and
deliberation from three practica, regional meetings, and stakeholder comments. The model
includes the major elements of each planning and scoping dimension. The relationships flow
from the sources that produce categories of stressors through environmental pathways and routes
of exposure to affect receptors. The endpoints were for ecological, health or economic system
(quality of life) effects. The setting for this model is a watershed in the panhandle of Oklahoma.
Sources
The CAFO is the primary source of stressors that are considered in this model. Also
within the watershed some of the same stressors may be contributed by existing sources which
may include: Agriculture (primarily livestock and row crops); oil and gas exploration; roads and
vehicular traffic; and domestic waste treatment facilities (both private and public). Existing
sources form a background for the type of stressors that the CAFO may add to the watershed. In
the figure, solid vertical lines are used to show common linkages between elements within a
dimension.
Stressors
Stressors were aggregated with consideration to their pathways and routes of exposure to
particular receptors, and common endpoints of concern. Nutrients includes phosphate, ammonia,
and water soluble nitrogen compounds that may be released from the land application or
discharge. The linkage to the air/aerosol pathway is represented by a dotted line to indicate the
group considered it to be insignificant. Air/aerosols are primarily volatile organic compounds
released from lagoons or the barns directly to the air. Associated chemicals include antibiotics,
pesticides, and nutritional supplements released from land application or surface water
B-3
-------
discharge. Erosion and sediments include physical particles from land application, infrastructure
construction, and transportation of supplies, animals, and wastes at the CAFO. Habitat alteration
includes soil compaction, construction of the facility, fragmentation of habitat, and changes in
vegetation. Groundwater loss reflects the net consumption of water by the facility. Odor is an
obvious public concern along with noise from the facility and traffic to and from the site. The
Nitrate stressor includes nitrite that are special concerns for groundwater. Methane and
greenhouse gases are stressors associated with the odor that are known to have health
consequences as well. Pathogens are shown as stressors for surface and ground water pathways.
CAFO workers are most likely to be exposed by inhalation and direct contact, but that is not
shown in the conceptual model. Pests, including mosquitoes, rats, and flies, may carry disease
beyond the facility by a terrestrial route. They also contribute to the nuisance factor of the
facility.
Environmental Pathways/Routes of Exposure
The environmental pathways are represented by four elements shown to aggregate the
processes of transport, transformation, decomposition, accumulation, and transfer. The route(s)
of exposure for the receptors are associated with each medium and represent best judgment of
the group about those that are most likely significant. Surface water flow in Oklahoma is
intermittent and generally does not serve as a drinking water supply. The strongest linkages are
between nutrient stressors and ecological receptors, especially aquatic and wetland ecosystems.
Stressors in air or aerosols are most strongly linked to human health receptors. The hypothesized
linkage between the stressor nutrients and air would be linked to all ecological receptors. It is
not shown, because it was considered to be insignificant. Terrestrial/habitat alteration integrates
the principal changes to the structure of the watershed and the habitats it provides. Presumably,
the land is already used for grazing or row crops and human health is not significantly affected
by this stressor. There is also an important link between the terrestrial and habitat alteration
stressors and socioeconomic receptors.
Receptors
As discussed during the workshops, receptors are the entities that are exposed to the
stressors. These entities exhibit the effects (endpoints). There are three groups of elements:
ecological, human health, and socioeconomic. The aquatic, terrestrial, and wetland's ecosystems
are considered interrelated and most of the stressor-effects linkages apply to all three. For human
health, infants are shown as a special group because of concern about nitrite in private drinking
water wells and the possibility of methemoglobinemia. The other elements (sensitive
populations, CAFO workers, other off-site residents and minorities) are likely to have the same
stressor-effects linkages.
B-4
-------
Assessment Endpoints
The elements under this dimension of the model are highly aggregated. They relate
directly to concerns and values that the stakeholders and risk manager may hold. They may also
relate directly to risk management objectives. In most cases, these are not directly measurable
endpoints (also called measurement endpoints). Again, there are three groups: ecological,
human health, and societal. The ecological endpoints highlight significant aspects of each
ecosystem and the whooping crane as a possible threatened and endangered species which may
occur in these watersheds. Health endpoints are clustered as three diseases and dermatitis and
the special concern for methemoglobinemia. This cluster also includes asthma. It is linked to
four receptors. Note that cancer is not among the principal effects. The last two clusters are
societal endpoints which transcend traditional human health. In the diagram, three receptors
(sensitive populations, other residents, and minorities) are linked to these clusters. CAFO
workers are excluded because they have an economic interest in the CAFO. Psychological
condition of the surrounding community is an endpoint which interacts with elements in both
clusters. Odor, noise, pathogens, pests and habitat change are of particular interest for analysis
and risk management of the CAFO in a watershed.
The CAFO discussion lead to a very detailed conceptual model, with a very high degree
of aggregation. In some cases, such as nutrients, the level of aggregation in the model was too
great to fit the control technologies (i.e., nitrogen and phosphate removal require different
technologies in waste water systems). Expert judgement was used to select which linkages were
highlighted. While it was easier to communicate the broad set of potential concerns to the
public, the development of the analysis plan was difficult to extract from the broad aggregation.
Therefore, a more detailed version of the model was retained for technical use.
Since the practica, Region 6 has used written materials and detailed sub-models to show
how stressors like nitrogen and phosphate affect human health and ecological receptors. The
additional details (e.g., for individual stressors such as nitrogen and phosphorus, or pathways of
exposure) allow analysts to separate potential risks so their significance and susceptibility to
management options can be evaluated. Aggregation in the model does not necessarily mean that
the risks are combined.
B-5
-------
CAFO Conceptual Model
Sources
Stressors
Pathways/Routes
Surface
Water
\ -L
1 Air/aerosols
(inhalation/
contact)
/
' I
Terrestrial/
Habitat Alter.
I i
At-Risk Components
Ecological
Minorities/
EJ Population
Endpoints
Ecological Endpoints
Condition of Aquatic
Ecosystem
Condition of
Terrestrial/Wetlands
Ecosystems
Birds
Wetlands structure
Playa lakes
Rangeland structure
Habitat mosaic
T/E Species
Whooping Crane
Human/Societal Endpoints
Hemoglobinemia
Dermatitis
I Respiratory disease
Immunologic disease
Gastrointestinal dis.
Pysch. condition
Farm culture change
Economic
System
-»>
Demographic change
Comm. Infrastructure
Aesthetics
Loss of recreation
Property values
-------
B.2.6 Analysis Plan
The practica participants did not develop an analysis plan for the CAFO, but they did
discuss how it might be done. They said that the analysis plan should provide the rationale for
limiting the scope of the risk assessment, because it is not possible to address everything in the
conceptual model. The criteria for analyzing a stressor-effect linkage should include feasibility,
likely significance (contribution to the overall risk), and data availability or likelihood of
obtaining it. The analysis plan should describe the tools to be used and explain the procedures
and rationale for the analysis. It should discuss data limitations, assumptions, and uncertainties.
Under NEPA, permit applicants usually prepare and submit their own Environmental
Impact Analysis (EIA), which requires them to do their own planning and scoping. Since
national program offices and regions generally do not require EIA/EISs for permits, EPA may
want to add EIA-relevant questions to its permit application. Region 6 is working to: identify
data requirements and sources in collaboration with other agencies, set priorities for missing
pieces, formulate hypotheses for the linkages within the model, develop a crosswalk with the
CRIA and general permit process, and establish a generic schedule.
B.3 Lessons Learned
1. Early involvement of the risk manager (decision maker) helped focus the process toward a
tangible product. Once the discussion focused on establishing the terms and conditions of a
general NPDES permit for the facility, the data requirements, a general approach, and processes
for involving the public emerged quickly.
2. Participants developed a conceptual model which identified specific public health endpoints
that were not covered in the Region 6 CRIA. Region 6's case presenter indicated that the Region
probably will go back and add them. EPA staff who were not familiar with the NPDES program
or CAFOs assumed the roles of community stakeholders. The workshops helped the region to
practice communication techniques with stakeholders for other problems.
3. Initially, the key public concern with
CAFOs was for odor. Although this was not
related to known health risks, it could still be
included in this discussion and management
options will be included in the general permit.
After the workshops, further literature
reviews uncovered research on health effects
of some odors, particularly from ammonia compounds.
Region 6 used the planning and scoping
process to create a strategy for dealing with
CAFOs and has also applied it to other
community based issues, even where a risk
assessment may not be developed.
4. Debate over terminology and brainstorming sessions were necessary to reach a consensus. A
clear set of definitions would aid this process.
B-7
-------
5. The planning and scoping process cannot be prescriptive, because the context of each
situation is different. Clear objectives, resource commitments, and estimated schedules from
management will drive the approach and level of detail that can be considered.
6. Consideration of "measurement endpoints" during formulation of the conceptual model may
unduly restrict the model because of concerns over data availability.
7. The workshop participants observed that EPA needs to overcome a cultural bias within the
agency that risk assessment is an internal function. Stakeholder engagement is essential at the
beginning.
8. The intent of NEPA is to include all stakeholders in the scoping process. Experience with
NEPA improved how ideas, recommendations, and agreements were solicited from stakeholders.
Region 6 used planning and scoping to create a strategy for dealing with CAFOs, which
they discussed with regional and program managers, other federal agencies, and stakeholders.
The planning and scoping process has also been applied to other community based issues, even
where a risk assessment may not be developed.
B-8
-------
Appendix C. The Reregi strati on of Pentachlorophenol: Case Study
C.I. Background/history
Pentachlorophenol (PCP), a heavy duty wood preservative, is an organic oil-borne
pesticide first registered in the United States in 1948 to prevent wood decay from fungal and
insect damage. PCP is formed by the high temperature chlorination of phenol, which results in
the formation of microcontaminants (dioxins, furans, and hexachlorobenzene) in PCP.
In 1978, USEPA issued a Federal Register Notice initiating an administrative process to
consider whether pesticide registrations for wood preservative chemicals should be cancelled or
modified due to adverse toxicological effects noted in animal toxicity studies. The Agency
issued notices of "Rebuttable Presumption Against Registration" (RPAR) for PCP based on
teratogenicity and fetotoxicity findings. In addition, the Agency determined that PCP use posed
the risk of oncogenicity due to the presence of microcontaminants (dioxins/furans/HCB). The
Agency subsequently published Position Documents to address comments made by stakeholders
on the Federal Register Notice. The conclusion of the RPAR process (now called "Special
Review") in 1984 and final settlement agreements with registrants in 1986 restricted PCP uses
and modified its terms and conditions of registration. The RPAR process also resulted in
cancellation in 1987 of certain non-wood preservative uses of PCP as a herbicide, defoliant,
mossicide, and mushroom house biocide. In 1993 uses of PCP were terminated as a biocide in
pulp and paper mills, oil wells, and cooling towers.
Currently, two U.S. manufacturers produce PCP, and approximately 100 wood treatment
plants apply the pesticide to wood. Treatment plants vary considerably in age and design. Utility
companies nationwide use 92.5% of all PCP-treated lumber for utility poles and cross arms.
Secondary uses include railroad crossties, wood pilings, fenceposts, and commercial/residential
structures, such as decks, fences, and walkways.
C.2 Purpose
The Office of Pesticide Programs/Antimicrobial Division (OPP/AD) is reassessing the
potential risks of PCP on human health and the environment. The reassessment is driven by a
FIFRA (Federal Insecticide, Fungicide, and Rodenticide Act) requirement for a Reregi strati on
Eligibility Decision (RED) on heavy duty wood preservatives, due in 2001, and will conform
with the Food Quality Protection Act (FQPA). Although the reregi strati on of PCP is regulated
under FIFRA, PCP also is regulated under the Clean Water Act, Clean Air Act, and Resource
Conservation and Recovery Act.
Reregi strati on of PCP is unusual in that it requires risk assessment harmonization with
Canada under the North American Free Trade Agreement (NAFTA). Mexico is not an active
participant in this process, but EPA and Canada keep Mexico informed.
C.3. Highlights and Key Findings
C-l
-------
C.3.1 Risk Assessor-Risk Manager Dialogue
The purpose of planning and scoping in this case was to develop a risk assessment to
inform a pesticide use Reregi strati on Eligibility Decision (RED) document for heavy duty wood
preservatives. Participants of all three workshops accepted this regulatory purpose without
debate.
C.3.2 Defining the scope
Workshop participants accepted the decision to limit the reassessment process to PCP
and address the other two wood preservatives (CCA and creosote) separately. Due to the time
limit, participants of the third workshop decided to focus mainly on PCP and dioxins/furans, but
not to address hexachlorobenzene. The participants recommended that the risk assessment for
HCB be performed separately and later integrate the results with the PCP component.
C.3.3 Planning and Development of a conceptual model
Ecological
The participants of the first two workshops identified the components of six dimensions
of ecological risk that form the elements of a conceptual model for ecological effects.
The participants considered the use of treated wood for utility poles and their disposal after use
as the most important sources of PCP release into the environment. Utility poles are widely
distributed even in residential neighborhoods, and while a lot of used wood is incinerated, some
is sold and even acquired by unsuspecting homeowners. Facilities that treat wood were not
considered a major source because drip pads and recapture technologies are supposed to prevent
release to the environment. However, participants in the third workshop suspected that treatment
plants may be an important source of PCP to the environment during the drying process at the
plant and possibly due to runoff. OPP/AD does not account for misuse in risk assessment, but
participants generally agreed that the conceptual model could encompass it.
Using the dimensions of ecological risk developed during the first two workshops,
OPP/AD had prepared a tentative PCP conceptual model for ecological effects (Figure 1). The
third workshop reviewed that model and recommended several changes:
Create one inclusive ("generic") conceptual model and sub-models as appropriate to
expand and highlight specific areas of focus.
Consider changing the emphasis for PCP (but not necessarily PCP's microcontaminants)
from the treated wood and disposal sources to wood treatment at the plant and disposal.
Participants suggested that treated wood in use as utility poles may not be a major
concern because PCP metabolizes rapidly under aerobic conditions and has a short half-
life and thus may not migrate far from the pole.
Describe miscellaneous sources (does it mean decks, retaining walls, garden borders?)
and whether they are important to the ecology or human health.
C-2
-------
Add furans to the dioxin stressor (i.e., dioxins/furans).
Reexamine and highlight the critical pathways from the top of the model to the bottom.
Add ingestion or food chain as a pathway.
Add assessment endpoints and measures so the model can be used to develop practical
management goals.
OPP/AD agreed to revise their conceptual model for ecological effects based on the
group's recommendations and distribute it to workshop participants for comment after the third
workshop. The revised conceptual models are presented below along with a narrative
description.
Human Health
The third workshop began developing a conceptual model for human health. Participants
concurred that while one generic human health model is needed, it may be easier to construct
separate submodels for each stressor or source and then combine them to create one
comprehensive model. The participants also recognized that the generic conceptual models for
ecological and human health should be comparable at some levels, such as sources and stressors.
When aggregating submodels to create a complete picture, it should be possible to determine
which components go together and which can be de-emphasized or eliminated from the
assessment.
The group agreed to use a materials-flow approach to try to identify potential human
exposures to a stressor from a source. The half-life in the environment and potential of the
chemical to bioaccumulate were also considered useful in identifying relevant pathways of
exposure.
The group briefly discussed other sources of stressors, such as PCP manufacturing, PCP
transportation, treatment of wood at the plant, transportation and handling of treated wood, wood
in use, remedial treatment (usually ground-line treatment of utility poles), and transportation and
disposal. The group decided to concentrate on one source onlywood treatmentand one
stressor"clean" PCP (i.e., without microcontaminants)due to the time limitation imposed by
the workshop, and complete a draft submodel for clean PCP (Figure 2 ). They then began a
conceptual submodel for the dioxins and furans and acknowledged that the Agency's
reassessment of dioxins and furans will affect OPP/AD's assessment.
Most participants agreed that the sources should be the same for PCP, dioxins/furans, and
hexachlorobenzene submodels. They also indicated that the sources for human health should be
similar, if not the same, as those identified for the ecological conceptual model.
Participants concurred on several components for the clean PCP conceptual submodel
using wood treatment at the plant as the primary source of PCP exposure to humans (see the
sidebar).
C-3
-------
While developing the PCP submodel, several participants acknowledged the importance
of defining the terms used in the models. For instance, participants had to try to clarify
definitions for pathway and route and found that ecological and human health risk assessment
guidelines use some terms differently. They also acknowledged that in their conceptual models
for human health they reversed the order of the dimensions listed in the guidance, putting the
stressor before the source.
Participants agreed that a clear rationale is needed to explain why a component may be
eliminated from a model. For instance, the indirect occupational pathway and non-occupational
visitors to the treatment plant probably could be eliminated from the clean PCP submodel
because their risks are so minimal. Using professional judgment, they expected that the oral
route for the direct occupational pathway probably represents a minor risk compared to dermal
and inhalation, especially since oral exposure (hand to mouth) is preceded by dermal exposure.
Also, workers may wear gloves and wash their hands before eating. Adults were selected
because younger people would not be allowed to work at the plant. Since data indicate that clean
PCP does not have reproductive, developmental, or immunological effects, these endpoints
probably could be eliminated from the clean PCP submodel. However, they may be relevant to
PCP's microcontaminants.
Participants discussed eliminating non-occupational pathways (neighbors as well as
visitors) from the submodel on the basis that if workers are protected then little or no risk can be
expected to reach people in the neighborhoods. However, risk managers pointed out that actually
the neighborhoods might be at greater risk because controls, such as protective clothing,
available for workers are not generally used by residents in the neighborhoods. Also, OSHA
typically allows less stringent risk standards for workers than EPA allows for residents.
Several participants recognized that duration and frequency of exposure are important
variables that influence endpoints and they are relevant for ecology and human health. They are
best captured in the text supporting the models rather than in the models themselves. Most
participants preferred the terms short-term and long-term rather than acute and chronic.
Participants then attempted to identify the major components of a dioxin/furan micro-
contaminant submodel for PCP, again using the wood treatment plant as the source (Figure 3).
Note that the lines drawn on the conceptual model and the submodels represent only
selected linkages for illustrative purposes. A complete conceptual model would require filling in
all appropriate linkages, and the arrows on the conceptual model ideally would be drawn in
various widths to indicate the strength or importance of each linkage.
A complete conceptual model and submodels for PCP would show differing levels of
depth and detail that the reader could view as desired. At the most aggregated level, the
conceptual model would show only the most important items and linkages. The level of detail
would increase with subsequent diagrams. That way, a nonscientific audience could visualize the
essence of the problem without getting lost in a highly detailed diagram, and a technical
C-4
-------
audience could view the more detailed diagrams and locate all plausible linkages and feedback
loops.
A revised conceptual model for PCP and the contaminants is presented below along with
a narrative description. A combined eco/human health conceptual model has not yet been
completed, but is under study.
C. 3.4 Stakehol der Involvement
Dr. Lauren Zeise of the California Environmental Protection Agency led participants of
the PCP group in the third workshop in applying the lessons from the National Research
Council's Understanding Risk: Informing Decisions in a Democratic Society. She posed six
questions that risk assessment planners should consider during planning and scoping:
1. Who are the interested and affected parties (stakeholders)?
Participants identified a wide variety of stakeholders: trade associations for manufacturers and
users (e.g., American Wood Preservatives Institute AWPI, American Wood Preservatives
Association AWPA, Chemical Manufacturer Association CMA); PCP manufacturers, utility
companies and others, such as workers (linemen, treatment), Penta Task Force (registrants),
NAFTA (Canada as co-regulator, Mexico as affected party), other EPA offices (e.g., ORD,
OAR, OW, OSW, OPPT), EPA regional offices, other agencies regarding OSHA issues, public,
Congress, and environmental groups such as National Coalition Against Misuse of Pesticides
(NCAMP).
2. What should be deliberated? When?
The group indicated that what is deliberated should depend on the stakeholder group. OPP/AD
has established regulatory negotiation procedures for deliberating with industry and is
considering broadening stakeholder involvement in this process.
Several participants suggested meeting early with stakeholders to scope the risk assessment and
to decide what should and should not be in the assessment. Secondary and tertiary exposures of
concern, impact on resources, and data needs to address management options also may be
discussed at this time. Intra-Agency deliberations may be required on disposal issues and on
contaminant issues, such as dioxin, which EPA's Office of Research and Development is
currently assessing. Most participants agreed that public involvement should be limited to appro-
priate assessment endpoints and that regulatory negotiations should not be conducted in public
forums. However, one person suggested that OPP/AD discuss the entire conceptual model, not
just the endpoints, with the public.
During the intermediate stages of the reassessment process, OPP/AD could distribute a draft of
the science chapter of the RED and meet with all stakeholders to make mid-course corrections.
OPP/AD usually hold meetings with industry to describe data gaps, the approach to the risk
C-5
-------
assessment, and assessment endpoints before the draft RED is completed. OPP/AD also hold
meetings on the draft RED with th National Coalition Against the Misuse of Pesticides. A final
meeting with various stakeholder groups may be helpful to discuss the findings and avoid
surprising anyone.
OPP/AD cautioned that when preliminary findings are widely distributed before the risk assess-
ment is complete, people may misinterpret EPA's intentions. This occurred when OPP/AD put
the preliminary risk assessment for organophosphate (OP) pesticides on the Internet in a pilot
program requested by the registrant. On the basis of this preliminary assessment, which raised
some health concerns, several OP users stopped buying OP products for fear EPA would soon
ban their use.
3. What approaches might be taken in interacting with stakeholders during deliberation? (Focus
groups, etc.)
Participants suggested that meetings would fulfill most needs. However, the Internet could also
be used to distribute information to the general public if costs allow. The current practice of
involving the public is through EPA's Science Advisory Board and Scientific Advisory Panels.
Dozens of meetings generally occur during the long pesticide registration process. Most of them
occur only with the registrants, but a few involve the public (e.g., NCAMP). Although OPP/AD
has not decided when and how to share information with the public, OPP/AD has no plans to
open the reregi strati on meetings with registrants to the public, mainly due to insufficient
resources.
4. How should participants be selected?
Participants acknowledged that some stakeholders, such as workers, environmental groups, and
the public, have not been directly involved in the regulatory process for PCP. Participants
recognized that public stakeholders for the PCP risk assessment are difficult to define because
there is no specific site to point to.
5. Should the program enlist outside help in establishing the deliberative process?
Since a formal regulatory process is already established, participants questioned the value of
outside facilitation and saw no need for it in the development of the RED.
6. What is the external constraints for deliberation (budget, time, legal)?
Participants acknowledged budget, time, and legal constraints as applicable to the pesticide
reregi strati on process.
C.3.5 Lessons Learned
Participants discussed briefly what they had learned from this case study. Their remarks
are summarized below:
C-6
-------
A. Ecological and human health _. . ... . .
° ^ . , Risk assessors and risk managers already
assessment planning can be ,, ,, . . .
, . j 7 it u ... perform some aspects or planning and
harmonized conceptually, but a . . . , . . .
. . .... , , scoping but the planning and scoping process
common vocabulary will be needed. 11. ^ 1 j i A ^
helped organize thinking, develop trust
. , ,. among participants, and should produce a
B. Both top-down and bottom-up .. r r r
, . , , . r . , quality assessment.
approaches to developing conceptual n J
models are usefulthere is no wrong
method.
C. Explaining uncertainty to stakeholders is a problem that needs up-front planning. There is
a hesitancy to reveal all that is known and not known about chemicals and their risks
because it reveals uncertainties that can lead to criticism and political ramifications. Yet,
uncertainty is inevitable since some questions will have no answers.
D. The interplay between risk assessors and risk managers was valuable. For instance, it
provided insight into the potential PCP risks to workers versus neighbors.
E. Planning and scoping for cumulative risk has proven to be very valuable. Risk assessors
and risk managers already perform the components of planning and scoping but generally
in an unorganized protracted way. Early planning and scoping helps organize everyone's
thinking and should result in a smoother and better quality assessment than is possible
without it.
F. Planning and scoping offers a good opportunity to identify problem scenarios early and
the potential risk management options to address them. It also offers an opportunity to
determine if data exist to compare management options and to develop contingency plans
for potential risks.
G. There are a lot more questions than answers, which is why an iterative planning and
scoping approach involving many different but knowledgeable people is valuable for
developing a good conceptual model.
H. Planning and scoping is not risk assessment, the deliberation involves a broader set of
participants in a dialogue.
C-7
-------
The utility and the time involved in planning and scoping need to be presented honestly
to risk managers. Purporting that planning and scoping will be quick and easy is likely to
be counterproductive; it is a lot more work than people assume. However, it ultimately
saves time by explicitly organizing an assessment that would have to be done at least
implicitly anyway. Also, documenting the planning and scoping leads to clearer thinking
and greater credibility, and it captures the thinking for others (stakeholders, risk
managers, and next generation risk assessors).
C-8
-------
C.4 Details on Conceptual Models for PCP
Sfressor
HCB
r
.ource
Pathways
Exposed
Populations
Suh-
poprrlations
Treate
Soil
h
d Wood
Water
Air
Terrestrial
h
Aquatic
Sub-terrestrial Terrestrial Freshwater Marine/e^stuarine
F F F I F F
orgasms Mammals Birds Invertpbrates Reptiles/amphibians Plants Fish
Population levels Ecosystem
arg*
Organizational individual levels
IP VP|$
Furl TI oinfv
Neurot
Measurements
Beha
disort
Species I
Reduction ii
Extin<
oxicity Sub- chronic Chr
to Jcity toxi
dor Moribundity Grc
onic Develo
city tox
wth Polyi
iversity
i Diversity
:tion
imental Reproc
city toxi
actylism Fecui
Reproductive
survival
uctive Endo
:ity disru
idity/ Thj
ility dysfu
crine Ac
ption toxi
roid Mor
ute Car
city gen
tality Li
'&j
;ino- Mi ita-
city gen city
i^er In vitro tests
lors In vivo tests
-------
Stressor
Source
Pathways
Exposed
Populations
Sub-
poprjlafion!?
PCP
Treate
Soil
iWood
Water
Air
Terrestrial
Aquatic
Sub-terrestrial Terrestrial Freshwater Marine/^stuarine
organisms Mammals Birds
Organizational
levels
Endpoints
Neurot
Measurements
Beha
disorc
[ndividu
al levels
oxicity Sub- chronic Chr
to Jcity toxi
/ior Moribundity Gri
~T- nr -T- -i
Invertpbrates Rentiles/amnhibians Plants Fish
Populat
Species I
Reduction ii
Extin*
onic Develo
city tox
wth Polyc
ion levels
Ecosystem
Landscape ecology
liversity
i Diversity
;tion
mental Reproc
city toxi
actylism Fecu
uctive Endo
city disru
idity/ Thj
ilitv dvsfu
crine Ac
ption toxi
roid Mor
nction
ute Car
city gen
tality Li
tun
:ino- Mi ita-
city gen icity
ver In vitro tests
mrs In vivo tests
Reproductive
survival
-------
Dioxin/Furan
Stressor
Source
Pathways
Exposed
Populations
Sub-
populations
orgasms Mammals Birds Invertpbrates Reptiles/amphibians Plants
Population levels
Soil
Treate
Terrestrial
j
d Wood
Water
It
Aquatic
Air
Sub-terrestrial Terrestrial Freshwater Marine/e^stuarine
f
Fish
Organizational Individual levels
levels
Ecosystem
Endjjoints
Neurot
Measurements
y".eha
isori
Species I
Reduction ii
Extin*
jxicity Sub- chronic Chr
to Jcity toxi
rior Moribundity Gr(
lers redu
Repro
sun
onic Develo
city tox
wth Poly(
ction/
ductive
dval
»-"-"£, j
iversity
i Diversity
:tion
imental Reproi
city toxi
actylism Fecui
fert
uctive Endocrine Ac ite
:ity disruption toxicity
idity/ Thyroid Mortality
ility dysfunction
Car:ino- Mi ita-
gen city gen city
Liver In vitro tests
tumors In vivo tests
-------
Sfressor
DIOXIN/FURAN
^ f] 11 l"f* f*
Manufacturing Treatment
Pathways Soil
Exposed
Populations
Treate
d Wood Transportation Disposal
Water
Air
Terrestrial
iL
Aquatic
Srrij-
poprrlations
JIH ^^^^^^^J^^^^^^^^^^m ^^^^^^^^J^^^^^^^^^m
*^ *r *l
Sub-terrestrial Terrestrial Freshwater Marine/gstuarine
sms MamnTals" Birds Invertebrates Reptiles/amphihTFns Pla ^
izational individual levels Population levels Ecosyltem
Landscape eco
T
Plants Fish
em
Landscape ecology
Endpoints
Neurot
Measurements
Beha
dison
Species Diversity
Reduction iji Diversity
Extinction
jxicity Sub-
to
rior Moribi
lers
chronic Chr
ticity toxi
indity Gr(
redu
Repro
Dnic Develo
city tox
wth Poly(
ction/
ductive
imental Reproc
city toxi
actylism Fecui
fert
f
uctive Endo
;ity disru
idity/ Thj
ility dysfu
crine Ac
ption toxi
roid Mor
nction
ite Car
city gen
tality Li
tun
:ino- Mi
city gen
^er In vit
tors In viv
ita-
icity
o tests
o tests
C-12
-------
C.5 Narrative for the Human Health Conceptual Models for Pentachlorophenol (PCP) and Its
Contaminants
The Antimicrobials Division (AD), Office of Pesticide Programs (OPP), U.S. Environmental
Protection Agency (USEPA), is currently developing the Reregi strati on Eligibility Decision for
pentachlorophenol, a heavy-duty wood preservative (HDWP). The purpose of the RED
document is to ensure that currently registered uses of pentachlorophenol are supported by
adequate science data and risk assessments that reflect current USEPA policy for regulation of
pesticide chemicals. In conjunction with the Science Policy Council's Cumulative Risk
Working Group, AD scientists have undertaken development of conceptual models for
pentachlorophenol. The purpose of conceptual model development is to describe the
relationships among predicted responses of a population of concern and its stressors, including
the environmental routes of exposure. The conceptual model also describes endpoints of
concern and how they will be measured. This approach (of developing conceptual models) is
intended to assist in the process of cumulative risk assessment, defined as "the potential risks
presented by multiple stressors in aggregate." During model development, key questions are
addressed, such as who is affected or stressed (receptors), what are the stressors (physical,
chemical, biological or psychological agents), what are the sources, the time frame of the risks,
and the assessment endpoints. For now, the Agency intends to focus on separate assessments of
adverse human health and ecological effects. A separate conceptual model and narrative for the
ecological effects of pentachlorophenol have also been developed by AD. This model may, in
the future, be integrated with the ecological conceptual models.
Stressors
Pentachlorophenol is an organic oil-borne pesticide used in the past for a wide variety of
applications (including herbicidal), but which is currently under restricted use status for
preservation of wood only. During manufacture of pentachlorophenol, dioxin and furan
contaminants as well as hexachlorobenzene are formed as the result of the high temperature
chlorination of phenol. Although considerable progress has been made towards reduction of
these contaminants within manufactured pentachlorophenol, they cannot be completely
eliminated. The wood preservation industry has argued that complete elimination of PCP
contaminants would represent a costly option and may present an undue economic burden.
However, there are efforts being undertaken to determine if the contaminants can be eliminated
completely from manufacture of PCP. At present, because this goal has not been achieved,
separate sub-models for the dioxin/furan and hexachlorobenzene contaminants were constructed
for purposes of assessing cumulative risk from exposure to pentachlorophenol. A further reason
for developing sub-models for the contaminants of pentachlorophenol was based on the distinct
toxicities resulting from exposure to the contaminants as opposed to pentachlorophenol alone. A
human health model which integrates the contaminants with pentachlorophenol is anticipated at
some point in the future.
In order to construct the conceptual model for pentachlorophenol and the contaminants, sources
C-13
-------
of pentachlorophenol had to be identified. In conjunction with the EPA's Planning and Scoping
Working Group, workshops were held to aid in development of these models.
Five sources of pentachlorophenol which could result in environmental exposure as related to
wood preservation were identified. These sources included manufacture of pentachlorophenol
itself; transport of pentachlorophenol to the wood treatment plant; treatment of wood with
pentachlorophenol; use of the treated wood; and eventual disposal of the treated wood once its
useful life has expired. Manufacture of pentachlorophenol and disposal of pentachlorophenol
treated wood were identified as outside the scope of regulation for the Office of Pesticide
Programs. These aspects of pentachlorophenol regulation, while recognized as relevant to the
cumulative risks from pentachlorophenol exposure, are under the regulatory authority of other
offices within EPA. The use of wood treated with pentachlorophenol does not present a
significant source of exposure as most of the pentachlorophenol remains within the treated wood.
Further, use sites for treated wood are restricted mainly to utility poles, further lowering the
potential for exposure. Thus, the treatment of wood with pentachlorophenol was felt to offer the
greatest potential for exposure to this chemical out of all of the identified sources. In the
treatment of wood, there are several opportunities for exposure to pentachlorophenol that do not
occur from the other sources that are within EPA's regulatory authority. Personnel treating
wood with pentachlorophenol will come into contact with the technical material when preparing
wood for pressure treatment, and may also contact the chemical when cleaning equipment used
for pressure treatment, or when handling freshly treated wood. Persons living within the vicinity
of the wood treatment facility or those visiting the facility may also come into contact with
pentachlorophenol through dermal or inhalation contact. These types of scenarios do not exist
for the other sources of exposure to pentachlorophenol; therefore, in relation to the other sources
of exposure, treatment of wood with pentachlorophenol was felt to be one of te most significant
sources for exposure. Thus, it was felt that description of a conceptual model using wood
treatment (in a plant and which represents a key exposure pathway) would be representative of
the other potential sources of exposure to both pentachlorophenol and the contaminants.
The conceptual model for "clean" pentachlorophenol focuses upon the use of pentachlorophenol
in the wood preservation process. Within the realm of wood preservation, there are various
types of treatments that can be performed with PCP. Commercial treatment of lumber, such as
utility poles, usually involves a pressure treatment process in which a quantity of wood is
subjected to treatment with PCP within a long metal cylinder (or retort). Treatment times can
vary based upon the type of wood being treated, but the process is an enclosed one. After
treatment, the treated wood is withdrawn from the retort and placed on a concrete drying pad to
collect any residual chemical that may leak from the treated wood. Non-pressure treated wood
is preserved by dipping or extended soaking in open vats. For remedial ground line treatment of
existing utility poles, brushing, swabbing, spraying, bandage wrap, or low pressure injection
techniques are employed. In each case of treatment, appropriate precautions are specified with
regard to the required protective equipment and clothing. However, the types of treatments just
described can result in or provide opportunities for significant exposure to both PCP and the
contaminants.
C-14
-------
Pathways
Pathways for exposure to PCP during treatment of wood can be either occupational (workers
within the treatment plant who actually handle treated wood and/or PCP) or non -occupational (
visitors to the plant as well as persons living in proximity to the plant). Within the occupational
pathway, direct exposure can occur (such as to workers handling treated wood ) as well as
indirect exposure (workers not handling treated wood but who may be exposed to PCP by virtue
of their job being located within the treatment plant, i.e., administrative workers). Persons living
within proximity to the plant as well as visitors to the plant are considered to have an indirect,
but not necessarily lower, exposure to PCP, through volatilization of PCP or contamination of
soil and water surrounding the treatment plant.
Routes of Exposure
Dermal and inhalation routes of exposure are considered significant routes for human exposure
to PCP in the wood treatment plant setting. Oral exposure through hand-to-mouth transfer can
also occur but is not considered as significant in the wood treatment setting. Oral exposure may
become more significant for other sources of PCP, such as use of treated lumber in residential
settings, especially for infants and children who accidentally ingest soil surrounding treated
wood.
For occupational pathways, both the dermal and inhalation routes are considered significant,
while for non-occupational pathways (visitors and those living in proximity to wood treatment
plants), the inhalation route would be most significant as a route of exposure. Within the
occupational setting, adults are the only subpopulation of concern, as children under 16 are not
expected to be employed in the wood treatment industry. The adult subpopulation includes both
males over 16 years of age as well as females (pregnant and non -pregnant). For the non -
occupational pathway, both adults and children have the potential for exposure.
C-15
-------
Appendix D. Cumulative Risk Initiative (CRT) for Cook County IL and Lake County IN
(formerly Chicago Cumulative Risk Initiative, CCRT)
The planning and scoping focus in the present summary relates primarily to the second
and third phases of CRT, a four phase project: (1) Environmental Loading Profile; (2) Petitioner
Risk Workshop; (3) Hazard Screening Assessment; and (4) Risk Management Response. The
Environmental Loading Profile is not discussed here, and outcomes from the fourth phase (risk
management) have not yet been implemented. "Lessons learned" are thus tentative at this
writing.
D.I. Background and History
In 1995 the Chicago Legal Clinic and 11 Chicago-area community advocacy groups filed
a petition under the Toxic Substances Control Act (TSCA) requesting that the USEPA
Administrator prohibit or further regulate the emissions from eight proposed or constructed
incinerators in the Chicago metropolitan area, including one proposed in Northwest Indiana.
The petitioners believed that neither current statutes nor local siting laws adequately address the
cumulative impacts of multiple sources of toxic pollutants in a geographic area, and requested
that the Administrator prohibit or further regulate the emissions of dioxins, furans, mercury, lead
and cadmium from these sources. In May 1996 the petition was withdrawn in response to a
USEPA offer to participate in an investigation of the multimedia impacts of pollutants in Cook
County, Illinois and Lake County, Indiana. This effort was named the Chicago Cumulative Risk
Initiative (CCRI) and later renamed Cumulative Risk Initiative (CRI) for Cook County IL and
Lake County IN. CRI is an attempt to investigate the issue of cumulative loadings and hazards
from pollutant sources, develop community-based activities to help address these concerns, and
use the results of the analyses to assist in prioritizing the use of regulatory agency resources.
The Agency and petitioners agreed to a four-phase project: (1) Environmental Loading Profile;
(2) Petitioner Risk Workshop; (3) Hazard Screening Assessment; and (4) Risk Management
Response.
As of this writing phases 1 and 2 are complete, and the third (Hazard Screening
Assessment) is near completion. The scope of that assessment reflects stakeholder deliberations,
focuses on cumulative hazard (not "risk" as typically defined by USEPA) associated with non-
criteria air pollutants ("air toxics") in the two county study area, and relies on "off-the-shelf air
pollutant information sources, including USEPA's Toxics Release Inventory and Cumulative
Exposure Project, the Regional Air Pollutant Inventory Development System (RAPIDS), and
outdoor air monitoring data. Emission estimates are "toxicity weighted", while
modeled/monitored outdoor air pollutant concentrations are compared with reference values to
develop "hazard index"-like ratios. The ratios or toxicity weighted emission estimates are used
to derive indicators of cumulative hazard, and then mapped over study area locations. Another
part of the study assembles pollutant hazard information and data on existing human disease
rates and blood lead concentrations to identify geographic areas where potentially elevated
hazards and individuals with potentially elevated susceptibilities are collocated.
D-l
-------
D.2. Highlights and key findings
D.2.1 Stakeholder involvement
CRT emerged as a regulatory agency response to a citizen petition. Stakeholders were
therefore defined by the circumstances of this petition and the response it generated, although
other potential stakeholders were considered during the scoping process (e.g. industry; residents
not represented by the community advocacy groups). Stakeholders included:
11 advocacy groups represented by the Chicago Legal Clinic
Indiana and Illinois state government representatives
Chicago Department of Health
Chicago Department of the Environment
USEPA (Region 5, OPPT, OAQPS, OAR)
D.2.2 Planning and scoping, conceptual model, analysis plan, resource considerations
Participants in the three CRI case study workshops defined the problem, goals,
stakeholders, stressors, sources and endpoints, sketched out a conceptual model of the planned
assessment, and developed a preliminary analytical plan. Discussion included a broad list of
stressors, sources, and endpoints, including some elements that were likely to be outside the
project's scope (e.g. assay for DNA adducts as biomarkers of pollutant exposure). The process
was an iterative one: while participants in each subsequent workshop considered, revised, and
expanded upon the work completed by previous groups, they did not consider themselves bound
to prior decisions.
The first workshop concentrated on identifying the elements of a conceptual model and
developed tools to measure effect level and monitor trends over time. The second workshop
developed a list of sources, stressors, and endpoints and drafted first-order models of human
health and ecological risk. By the third workshop, the idea of preparing a conceptual model for
ecological risk was discarded because ecological concerns were not raised by petitioners.
Participants in the third workshop refined the human health conceptual model and concentrated
on the four tasks associated with the planning and scoping dialogue:
(1) Define the purpose of the assessment. Goals and objectives were developed [NOTE: these
pertain to all four CRI phases, not just the Hazard Screening Assessment (third phase)]:
Goals:
to develop the data upon which to base a strategic plan to improve air quality in the two-
county area by effectively targeting emission reduction activities.
to develop the data upon which to develop a strategic plan to improve public health by
effectively targeting intervention activities.
Objectives:
D-2
-------
Better understand environmental conditions in Cook and Lake counties by examining the
air quality impact of point, area, and mobile sources;
Foster dialogue with stakeholders;
Develop a transferable methodology that can be used in other urban areas; and
Inform enforcement targeting and pollution prevention strategies.
(2) Define the scope of the assessment. The following decisions were made:
1. Conduct a cumulative rather than comparative analysis
2. Focus specifically on children
3. Focus on sources rather than receptors
4. Concentrate on USEPA-regulated sources
5. Use only existing data
6. Cover a broad two-county geographic area rather than smaller geographic subareas
7. Limit study to air pathway/medium
8. Do not associate health outcomes with causes
9. Consider hazard rather than risk (i.e. no explicit exposure assessment).
(3) Develop a cumulative risk (hazard) outline. A conceptual model with eight elements in the
following hierarchy was generated:
Activity^ Sources^ Stressors^Pathways^Media^Route^Populati on ^-Health Effects
Measures/Biomarkers
(4) Formulate the technical approach to the assessment
Discussion focused on the conceptual model, analytical plan, and data (availability, limitations,
sources and outputs). The conceptual model and the analytical plan resulted from the planning
and scoping process. After drafting a broad and inclusive CRT conceptual model, workshop
participants developed "functional" conceptual submodels to address each of the overall
assessment goals, given data availability and limitations. They noted that developing a broad-
based inclusive model from which to draw submodels provided a "tool" that Region 5 could use
to communicate risks (hazards) "in context" to interested and affected parties and to the public
at-large. Each functional submodel is a subset of the broad-based, inclusive CRT model modified
to fit the scope of the proposed assessment, and each addresses one of the two overall assessment
goals.
D-3
-------
Goal 1: To develop the data
upon which to base a
strategic plan to improve air
quality in the two-county area
by effectively targeting
emission reduction activities.
"Functional"Conceptual Submodel: Uses CRT Conceptual
model from"top/down" to focus on source emissions and to
map distribution of pollution in subareas.
Elements: Activity>Sources^Stressors ^Pathways
(Outdoor and Indoor Air only) HVIedia (Outdoor and Indoor
Air only)>-Route (Inhalation only).
Goal 2: To develop the data
upon which to develop a
strategic plan to improve
public health by effectively
targeting intervention
activities.
"Functional"Conceptual Submodel: Uses CRI conceptual
model from "bottom/up" to look at public health issues, i.e.,
health effects on various sub-populations, without tying
effects to causes or to specific sources.
Elements: Populations (Children)-*Health Effects Measures/
Biomarkers.
Participants emphasized that the conceptual submodel for Goal 1 focuses on outdoor and
indoor air pathways and the inhalation routenot because other pathways are not important, but
because of the objectives agreed upon by USEPA and the stakeholder groups. Similarly, the
conceptual submodel developed to address the public health objective in Goal 2 is limited to
addressing effects on children and infants because stakeholder petitioners asked USEPA to focus
on children.
D.3 Risk Assessor - Risk Manager Dialogue
The terms "risk assessor" and "risk manager" are difficult to separate and may not be
particularly useful for purposes of the CRI Hazard Screening Assessment. This is due in part to
the Report's objective (i.e., a hazard, rather than risk assessment) and in part to the
circumstances in which the effort occurred. Argonne National Laboratory and USEPA
developed an interagency agreement in which Argonne was to conduct technical analysis
involved in a hazard assessment, the scope of which was defined by petitioner and governmental
stakeholders. In addition, all stakeholders had representatives reviewing and commenting on
Argonne drafts of the Report's chapters. In many cases this review led to substantial revisions or
novel analyses. Thus, one could argue that stakeholders had both "assessor" and "manager"
roles as those terms are typically used in the context of designing and conducting a risk (hazard)
assessment.
The term "risk manager" is also used in describing the use of risk assessment results for
some decision or action. Because it's likely that both governmental and petitioner stakeholders
will use CRI results in their own way, both are also likely to play this "risk manager" role. It's
anticipated that governmental stakeholders will use the results to assist in prioritizing program
activities and directing resources. Petitioner stakeholders could use the results for similar
purposes, e.g. to argue that elevated hazard estimates in particular geographic areas support the
need for additional air monitoring in those areas.
D-4
-------
D.4. Lessons Learned
"Interim Lessons" related to stakeholder involvement:
1. Despite understandable frustration with the |~Z ~ '. ~" '. ~
, . fl^m , . u 4. The duration of the planning and scoping
slow progression of CRI, relations between j / i \
.... . .. , ,u . process can expand or (in the present case)
petitioner representatives and the current ^ , , . v T ,.,.,/.
, ,- , /TTCT-TIA narrow the resulting scope. In hindsight, this
coordinating government agency (USEPA , , 111111 ..
. c\ ^u 11 j reduced scope probably helped to maintain
Region 5) appear to be generally good, even .. i 11
n. i4.- i T-I u citizen and management support so that the
after a multiple-year process. This may be , . j
due to the "self-selection" of the stakeholders
and the apparent patience of some in waiting
for a project designed largely as a result of their input. One potential lesson is that citizen
stakeholders' patience is directly proportional to their sense of influence in the process.
2. Whether other stakeholders (e.g. industry, residents not represented by Petitioners) should
have been involved earlier in the CRI process is likely to remain unknown until "risk
management actions" based on CRI results have been taken. Also unknown is whether or how
including such stakeholders against the wishes of Petitioners might have changed the project for
better or worse.
3. The extended deliberations involved in planning, scoping, writing, reviewing, revising and
completing the CRI Hazard Screening Assessment appear to have both strengths and
weaknesses. On the one hand, the extensive review that occurred in preparing the Report makes
more likely that it addresses the objectives of its designers. The Report also seems to
incorporate recent expert advice on stakeholder inclusiveness in issues relating to environmental
and human health risk assessment (e.g. NRC 1996; Presidential Commission, 1997). On the
other hand, conducting and completing such a process is labor and time intensive/extensive,
costly, associated with substantial management needs (e.g. coordination), frequent delays,
mistakes, misunderstandings and miscommunication ("too many cooks in the kitchen"). Those
initiating complex projects with many participants and multiple-year time lines, in some cases
extending beyond some participants' employment tenure, should carefully consider these costs
(including opportunity costs) during planning and scoping. Whether such costs are merited is
likely to be a difficult and subjective evaluation.
"Interim lessons" related to the planning/scoping process:
1. The duration of the planning and scoping process can expand or (in the present case) narrow
the resulting scope. For example, noise, odors and indoor air quality were discussed by planning
and scoping participants and considered relevant to cumulative hazard assessment. However,
these topics were eventually excluded from the Hazard Screening Assessment as the scope was
narrowed to focus on hazards of outdoor "air toxics". In hindsight, this scope attenuation was
probably a good thing, given the difficulties and expense of completing the Report even without
these topics.
D-5
-------
2. "Focus specifically on children": the Hazard Screening Assessment's children's focus is
mostly limited to the perspective that information on the co-location of potentially elevated
pollutant hazards and more susceptible children (those with asthma, elevated blood lead,
leukemia, upper respiratory tract infections) is valuable. The extent to which environmental
pollutant exposure is related to these diseases was not assessed, nor was pollutant exposure in
the study area explicitly assessed. With regard to the developmental toxicity of the -250
pollutants included in the Hazard Screening Assessment, the available toxicology literature is
rife with data gaps for individual pollutants, to say nothing of mixtures. Only a few of these
pollutants or classes (e.g., lead, mercury, and PCBs) are relatively well characterized for effects
in developing organisms such as children. The effects of exposure to changing-component
pollutant mixtures at varying environmental concentrations over the early lifespan (e.g.
preconception, in utero, infancy, childhood, adolescence) are mostly unknown.
Thus, while addressing the agreed-upon scope, the Report also reinforces the notion of large data
gaps and enormous uncertainty associated with a "focus on children".
3. "Use only existing data" and "off-the-shelf tools"; this decision led to unanticipated problems
that are briefly described:
(a) One problem not addressed until late in the analytic process was verification of the census
tract(s) within which air emission facilities named in the Report were located. This verification
procedure supported the notion that both of the emissions databases utilized (e.g. TRI, RAPIDS)
are likely to contain a fairly high rate (-20%) of inaccurate geographic location information.
(b) uncertain data and estimates: constant change in scientific knowledge and the long duration
of CRT led to difficulties, compromise, and expense. For example, the USEPA inhalation unit
risk (cancer potency) factor for 1,3-butadiene was under review starting early in the CRT process.
Because the revised value and the date of its "finalization" were unknown, outputs based on this
factor were frequently done twice (using both the current and expected value), generating two
versions of many analyses. Although still unknown at this writing, it now appears that a value
between that of the current and expected factor will actually be finalized.
4. Surprises: even something as apparently fixed as initial written objectives became the subject
of debate and modification during review of the CRT Hazard Screening Assessment. One CRT
objective developed during planning and scoping was "develop a transferable methodology that
can be used in other urban areas". During the several year duration of CRT, the separate effort of
the National Air Toxics Assessment (NATA) led by another USEPA office (Office of Air
Quality Planning and Standards, OAQPS) was initiated, with objectives intersecting those of
CRT. NATA activities include a national scale assessment, as well as development of local scale
hazardous air pollutant (HAP) evaluations. These local scale evaluations are likely to comprise
specific local information (e.g. terrain, weather patterns) and refined air modeling protocols, and
could provide a basis for national guidance on local scale HAP evaluations. The potential
conflict between CRT's "transferable methodology" objective and that of the NATA local scale
HAP evaluation element was identified by OAQPS reviewers. To avoid this potential conflict
D-6
-------
and the idea that the CRI Hazard Screening Assessment is intended to provide general USEPA
guidance on local scale HAP evaluations, it was necessary to modify the objective in the Hazard
Screening Assessment to "develop methods that can be adapted for use in other urban areas".
5. Despite the name "Cumulative Risk Initiative", the CRI Hazard Screening Assessment is
quite limited in its "cumulative-ness" (e.g. it excluded water and dietary pollutants), it did not
address risk in that exposure assessment was excluded, and it may be as much a response as an
initiative. Some may be disappointed by these limitations.
6. For community-based assessments, Region 5 personnel have found that planning and scoping
can be aided by addressing the following questions:
1. Who are the parties proposing the assessment?
2. Are there other interested or affected parties?
3. What questions do the parties want the assessment to answer?
4. What analysis will be done to answer these questions?
5. Who will conduct the analysis?
6. When are the assessment results needed?
7. Who will pay for the assessment?
8. How will the assessment results be used?
D-7
-------
Appendix E. Planning and Scoping for the National-Scale Assessments
Two other cases involving national assessments are discussed below. The nature of the
stakeholders and their discussions are very different from our experience with place-based
assessments.
E. 1 The National Air Toxics Program Assessment
The Office of Air Quality Planning and Standards (OAQPS) described their approach to
planning and scoping for the 1996 national-scale assessment of air toxics performed as part of its
National Air Toxics Assessment (NATA) activities. The NATA national-scale assessment is a
geographically broad study of potential inhalation exposures and health risks associated with 33
hazardous air pollutants of concern in urban air. It includes cumulative exposure and risk
assessments. This summary highlights the technical products from planning and scoping and
shows the link to the risk analysis.
E. 1.1 Purpose for this assessment
The information from public comments, monitoring, and assessments developed by
NATA activities will help:
1. Determine priorities for regulatory programs as well as for national, regional, and
community-based initiatives;
2. Assess progress toward statutory and future risk-based GPRA goals;
3. Inform state, local, and tribal programs; support public right-to-know initiatives
with regard to the risks associated with exposure to HAPs; and
4. Support prospective assessments of the benefits attributable to implementation of
statutory air toxics mandates (as required by section 812 of the CAA).
E. 1.2 Stakeholder Involvement
The Office of Air Quality Planning and Standards (OAQPS) solicited the perspectives of
key stakeholders as they developed and implemented the air toxics program and integrated urban
strategy. Key stakeholders include regulatory partners (including State, local, and tribal
governments), environmental justice communities, public health and environmental groups,
small business, industry, and urban developers.
EPA initially received hundreds of stakeholder public comments on the draft integrated
urban strategy notice published in the Federal Register on September 14, 1998, and at several
stakeholder meetings across the country. EPA also held informal discussions with several
stakeholder groups, including representatives from the State and Territorial Air Pollution
E-l
-------
Program Administrators and the Association of Local Air Pollution Control Officials
(STAPPA/ALAPO), the National Environmental Justice Advisory Council (NEJAC), the Clean
Air Act Advisory Committee (CAAAC), and the Conference of Mayors. Additional meetings
will be held to discuss implementation of the integrated urban strategy after the assessment is
completed.
E.I. 3 Details of the Conceptual Model
The following subsections include summary descriptions of the risk dimensions and
elements of the national scale assessment for NAT A, as recommended by EPA's Cumulative
Risk Assessment Guidance f1]. The conceptual model for the NATA appears in Figure 3. The
discussion below covers the rationale for what is included and excluded in the assessment and
assumptions for the analysis plan of this project.
A. Sources
The dispersion modeling (from which the exposure assessment and risk characterization
will arise) will include all major, area, and mobile sources that have been entered in EPA's 1996
National Toxics Inventory (NTI) for the contiguous US, Puerto Rico, and the Virgin Islands.
The 1996 NTI, which has been assembled from information on individual sources submitted by
state and local authorities, is the most recent and best available emissions database for the United
States.
By limiting the initial NATA to inventoried sources, EPA is thereby excluding releases
from sources that are not included in the NTI. This limitation will effectively exclude releases
(1) from natural processes, (2)-to indoor air (e.g., from paints, carpets, etc.), and (3) to surface
water, groundwater, or soil. While EPA takes these releases and their potential to cause adverse
health effects seriously, adequate model inputs (i.e., data on substance identities and release
rates) are still needed to include them in the assessment. Furthermore, because most of these
releases fall outside EPA's mandate to control emissions of HAPs under the CAA, it is uncertain
that the information would be useful to the development of air toxics control strategies.
B. Stressors
The initial NATA will encompass the 33 substances that EPA has identified as urban air
toxics under the Urban Air Toxics Strategy. Later NATA assessments will expand to cover as
many of the 188 Clean Air Act HAPs as available emission and toxicity data will support.
EPA has chosen to limit the initial assessment to the 33 urban HAPs for two reasons. First, these
HAPs, in aggregate, are highly likely to encompass most of the total HAP-related inhalation risk
to human populations. Second, EPA intends to use the initial assessment for NATA as a
principal vehicle to fulfill assessment commitments under the UATS, which is focused
specifically on these HAPs.
C. Pathways/ Media
E-2
-------
The dispersion modeling and exposure assessment will include transport of particles and
gases through air to receptors within 50 km of sources. Atmospheric transformation and losses
from the air by deposition will be included in the modeling, as data permit. The initial
assessment will exclude accretion in water, soil, or food associated with deposition from air, or
bioaccumulation of airborne HAPs in tissues. Although EPA takes potential transport of HAPs
into other media very seriously, tools to model multi pathway concentrations on the national
scale do not yet exist. Future local- or urban-scale assessments will include multi-pathway
calculations, and they will be added to national assessments when adequate models become
available.
D. Routes
The NATA will focus on exposures due to inhalation of ambient air. Human receptors
will be modeled for separate micro environments, including residences, offices, schools, outdoor
work sites, automobiles, etc. The exposure assessment will estimate air concentrations of each
substance within each micro environment, using the outdoor concentration, time of day, air
exchange rate, and other factors. Human behaviors and physiology will be reflected in the
assessment by the amount of time individuals spend in each micro environment, and by the
inhalation rate during their time there.
The NATA will exclude human exposures via ingestion or dermal contact. This is a
consequence of the lack of multi pathway models suitable for calculations at the national scale.
As modeling tools become available to estimate transfers of substances from air to other media,
future national assessments may include dermal and ingestion exposures.
E. Subpopulations
The NATA will characterize risks to 12 distinct human subpopulations, divided into four
age cohorts and three socio-economic cohorts. Subpopulations planned for separate assessment
include (cohorts are inclusive): (1) young children aged 7 or less, (2) older children and
adolescents aged 8-17, (3) adults aged 18-64, and (4) people aged 65 or greater. Each of these
age groups will be divided by income level, at the 0-25, 25-75, and 75-100th percentile income
levels. Risks will be estimated separately for each group. The median and 95th percentile
individuals within each census tract will represent multiple descriptors of risk for all groups
combined.
The initial assessment will exclude non-human receptors. This limitation results from the
extreme complexity of considering potential adverse ecological impacts to the multiplicity of
different ecosystems that exist within such a large area. Future local- and urban-scale
assessments may be expanded to include non-human receptors, contingent on the availability of
necessary resources, data, and methodologies. However, non-human receptors will not be
included in future national-level assessments unless radical new models and tools become
available.
E-3
-------
F. Endpoints and Measures
1. Cancer
"Cancer" describes a group of related diseases that affect a variety of organs and tissues.
Cancer results from a combination of genetic damage and non-genetic factors that favor the
growth of damaged cells. At current cancer incidence rates, approximately one third of U.S.
residents may be expected eventually to contract some form of cancer. Cancer is associated with
a wide range of factors, of which exposure to HAPs is only one. Other causes of cancer,
including genetic susceptibility, background radiation, diet, smoking, and other lifestyle factors,
are thought to be the dominant factors determining total cancer incidence. Against the very high
total cancer rate of about one in three from all risk factors, the rate of cancer incidence associated
with HAPs alone cannot be observed directly. Attributing cancer to specific HAPs is also
complicated by the fact that many cancers do not appear for years or decades after exposure and,
therefore, may have been caused by past exposures in different locations. As a result, the
National Site Assessment will rely on modeled estimates of cancer risk rather than on direct
measurements for assessing risks.
The NATA will incorporate predictions of lifetime cancer risk to exposed populations.
Predictions will consider both EPA's 1986 cancer guidelines and the most recent draft version of
EPA's guidelines for cancer risk assessment, currently undergoing Agency science policy
review. For most carcinogenic HAPs, unit risk estimates developed by linear extrapolation from
high to low doses will not be used to estimate the upper bound of lifetime probability of
contracting cancer from inhalation. Available peer-reviewed dose-response assessments
developed from evidence of a threshold for carcinogenicity, or on sublinear low-dose
extrapolations, will be used as appropriate for specific HAPs.
The upper-bound lifetime cancer risk will be estimated for each HAP that has been
assessed as a known, probable, or possible human carcinogen, and for which a unit risk estimate
is available from a peer-reviewed source. Individual-HAP risk estimates will be calculated for
receptor populations within each census tract. Risks will then be aggregated across carcinogenic
HAPs.
Cancer risks will not be aggregated across weight-of-evidence categories (i.e., combining
known, probable, and possible human carcinogens), to avoid inappropriate mixing of
assessments having widely varying levels of uncertainty.
E-4
-------
2. Effects Other Than Cancer
Adverse health effects other than cancer ("noncancer risks") include a wide range of
health endpoints in all organ systems (e.g., cardiovascular, immune, liver, or kidney). As with
cancer, other factors such as genetics, diet, lifestyle, and other exposures (e.g., smoking) may
exert a dominant influence over incidence of adverse noncancer health effects. Therefore, as
with carcinogens, the NATA will rely primarily on risk estimates rather than on direct
measurements of changes in the incidence of adverse noncancer health impacts due to reductions
in emissions. These estimates will in most cases be expressed in terms of the hazard quotient, or
HQ (defined as the ratio of the inhaled exposure concentration to the reference concentration, or
RfC).
The HQ for effects other than cancer will be calculated for each urban HAP that has a
peer-reviewed RfC or equivalent value. HQs for individual HAPs will be estimated for receptor
populations within each census tract. HQs will then be aggregated across HAPs. Where
evidence exists for non-additive interactions among HAPs (e.g., synergism, antagonism,
potentiation, etc.), these will be considered as appropriate.
Probabilities of adverse non-cancer effects may not be possible to estimate. The
approach will generally aggregate HQ across HAPs on the basis of target organ and by toxic
mechanism if data permit. HQs will be separated according to total uncertainty in the RfC, to
avoid inappropriate mixing of assessments having widely varying levels of uncertainty.
E. 1.5 Analysis Plan for Cumulative Risk Assessment and Characterization
The document also provides a detailed analysis plan, including references to Agency
guidelines and program procedures. The plan describes the data, models, and key assumptions
that will be used in each phase (exposure assessment, dose-response assessment, and risk
characterization) of the risk assessment. Uncertainties associated with each and the use of
models are described, as well as the approach that will be followed.
The initial assessment will include four major steps: (1) compiling a 1996 national
emissions inventory of HAP emissions from outdoor sources; (2) estimating 1996 HAP ambient
air concentrations for the 33 urban HAPs nationwide; (3) estimating 1996 population exposures
to these HAPs; and (4) characterizing potential public health risks due to inhalation of HAPs,
including both cancer and noncancer effects. The document also includes additional information
from the risk assessment process, including a preliminary risk characterization and a detailed
discussion on their plan for aggregating the data.
E. 1.6 Lessons Learned
1. Planning and scoping required extensive involvement of the risk manager and technical staff
to develop a rationale for what would be included and excluded from the NATA. It would not
have been a problem if it was recognized and planned, but some of the time was consumed by
E-5
-------
The review of the conceptual model led to
significant savings in the application of the
model for calculating air dispersion,
exposure, and risk estimation. More than a
third of the possible analyses were shown to
be unnecessary to address the problem
formulated in the planning and scoping
discussion.
learning what needed to be done and
interpreting the 1997 guidance.
2. Because this was a national assessment for
screening, stakeholders had technical and
regulatory expertise, so there was little
background or discussion of non-technical
issues. Planning and scoping dealt with
details of defining the dimensions of the risk
assessment and the methods for combining
effects.
3. The conceptual model showed how the program related to other regulatory activities as well
as the relationships between stressors and effects. This context helped clarify how the results
would be explained and presented. Boxes using hyperlinks were very effective ways to present
both an overview and examine the technical details where specific questions occurred.
4. The review of the conceptual model lead to significant savings in the application of the model
for calculating air dispersion, exposure, and risk estimation. More than a third of the possible
analyses were shown to be unnecessary to address the problem formulated in the planning and
scoping discussion.
E.2 RCRA Surface Impoundment Study-Technical Plan for Human Health and Ecological Risk
Assessmentl
The Office of Solid Waste developed a technical plan for complex and cumulative risk
assessments of surface impoundments. The objective of the study was to conduct risk
assessments to determine, within an acceptable degree of certainty, what risks to human health
and the environment are posed by industrial wastewaters managed in surface impoundments.
This technical plan covers several steps which are described in the planning and scoping
guidance. This brief summary of the background and technical plan highlights some of the
techniques this exercise used that show how to implement the planning and scoping process.
E.2.1 Background
In 1996 the Resource Conservation Recovery Act (RCRA) was amended to exempt
decharacterized nonhazardous wastes from land disposal restrictions. Such wastes it was
assumed, have lost their hazardous waste characteristics (i.e., ignitability, corrosivity, reactivity,
and toxicity) through dilution or other treatment. Congress required that the Agency study the
health and environmental risks of exempted wastes managed in surface impoundments or
wastewater treatment systems and evaluate the extent to which existing regulations address any
Surface Impoundment Study-Technical Plan for Human Health and Ecological Risk Assessment, U.S. Environmental
Protection Agency, Office of Solid Waste, Washington, B.C., February, 2000
E-6
-------
such risks (1996 Land Disposal Program Flexibility Act). The scope of the study was expanded
to cover all nonhazardous industrial wastewaters in surface impoundments through consent
decree (Civ. No. 89-0598, EDF vs Browner). Based on statistical surveys, the Agency estimates
that there are 19,000 impoundments at 8,500 facilities within the study's scope. The technical
plan describes a screening process to select facilities, impoundments and constituents for further
assessment.
E.2.2 Purpose and objectives for the study
The study began with a subsample of impoundments to clarify the industries involved,
size range of facilities, constituents (stressors) present, and other key factors that might be
described for planning a risk assessment. After the survey, the Agency developed a technical
plan, then obtained peer review and public comment on the approach. The approach comprised
two phases: Phase I, a screening and prioritization process and Phase II the detailed plan for
analysis. The purpose of Phase I is to eliminate constituents (stressors) and impoundments
(sources) from further analysis which posed negligible risks and prioritize the remaining units
and constituents for further analysis. Phase II is a plan for detailed multimedia modeling which
is continuing. Phase I leads to a decision on whether a risk assessment is necessary and if so,
what will be included. Phase II develops theoretical relationships between constituents and
effects (a conceptual model) including fate and transport of the constituents.
E.2.3 Phase I (Conceptual Models and Analysis Plan)
Health and ecological risks are screened separately in each phase. A simple conceptual
model is presented in Phase I for sources through potential receptors (figure 2-2). The model
also serves as a checklist for screening each unit and constituent. Equations and data sources are
also provided for developing screening factors and each screening factor is presented in the
context of a flow diagram for the overall analysis and in a decision tree for evaluating the
significance of various pathways of exposure for each facility.
The analytical plan specifies that calculated screening risks (Phase I) for each constituent
in a specific impoundment and facility will be combined to generate three cumulative risk
estimates: an impoundment risk, constituent risk, and facility risk. The cumulative risks will be
used in the risk screening and risk distributions as follows. The hazard quotient (HQ) is the ratio
of estimated exposure (dose or concentration) and the appropriate toxicity value (reference dose
of reference concentration) for a single exposure pathway and chemical. The hazard index (HI)
is the summation of HQs across pathways and across chemicals affecting the same target organ.
Following screening and prioritization (based on scores from cumulative risk screening
and cumulative risk distributions), further data collection, model simulation, and refinement of
the risk assessment occurs under Phase II.
E-7
-------
PRIMARY
SOURCE(S)
PRIMARY
RELEASE
MECHANISM(S)
SECONDARY
SECONDARY RELEASE MIGRATION EXPOSURE
SOURCE(S) MECHANISM(S) PATHWAY(S) ROUTE(S)
Ingestion of drinking water
Ingestion of soil
Wind erosion
( , ------- , r ---------
Airborne Dust I ------ X Dispersion ( --- ^i Airborne Dust
{
{
{
' f C
I ,.1 Inhalation of oust 1 Resident
Inhalation of vapors
Pathway evaluated
Pathway not evaluated
Figure E-l. Human health risk conceptual site model and potential exposure pathways.
E-8
-------
E.2.4 Phase II (Assessment Plan)
Impoundment units and constituents identified for the Phase II analysis will be further
characterized with greater precision for potential human health and environmental risks.
As in Phase I, health and ecological risks are screened separately (see human health and
ecological conceptual models, figures E-2 and E-3) using the 3MRA2 conceptual model (HWIR)
for surface impoundments (see figure E-4, dimensions). One of two approaches will be taken.
If, as anticipated, a fairly limited number of units and constituents proceed to Phase II, EPA will
conduct multimedia fate and transport modeling of potential human and ecological risks using
the HWIR multimedia model and using, to the extent possible, the site-specific hydrogeologic
data, watershed parameters, and receptor data provided in the surveys and available through
other data sources such as GIS files. Due to its intensive modeling requirements, only a
relatively limited number of cases will be completed.
Alternatively, if a large number of sites meet the criteria for proceeding to Phase II, EPA will
develop a range of appropriate hydrogeologic an watershed "scenarios" (-20 to 30 representative
scenarios) to simplify the process of data file development and modeling. This will greatly
streamline the use of the HWIR model while maintaining the advantages of this powerful tool to
describe multimedia fate and transport. The Agency is also considering extending the
"representative scenario" approach to include representative ranges of population exposures.
Phase II results will be used to revise the risk profile for the surface impoundment universe
based on more realistic exposure assumptions and multimedia fate and transport modeling.
3MRA is a multimedia, multipathway, multireceptor risk analysis model. See Appendix D of Surface Impoundment
Study Technical Plan for assumptions, limitations, inputs and outputs. Use of this model includes the ability to use many of the
same data files for default parameters that had been developed to support the HWIR effort; the automatic integration of the
various modules for different media thereby minimizing the quality assurance/quality control (QA/QC) necessary for manual
integration of modules; and the feasibility of using the system in screening-level multimedia analyses and comprehensive
multimedia analyses.
E-9
-------
Conceptual Exposure Model for Human Receptors
PRIMARY
SOURCE! S)
PRIMARY
RELEASE
MECHANISM(S)
SECONDARY
SOURCE(S)
SECONDARY
RELEASE
MECIIANISM(S)
MIGRATION
PATIIWAY(S)
EXPOSURE
ROUTE(S)
POTENTIAL
RECEPTOR(S)
1 Kunoffi'Krosion >l WalersI
(post-closure only)
K*
Jnticstiori of drinking water
Inhalation while showering
pngL-stion of fruits &. vegetables f"
_^J Ingcstionofmcat J Farmei
1 Jnecstion of milk 1
J Ingestion of fruils & vegetables J Home
Resident
Home gardener
Farmer
fisher
J
Complete pathway (evaluated)
Incomplete pathway (not evaluated)
1. For each receptor type, adult and two child groups included.
2. Slock water consumed by farm animals only.
3. If wetland, ingestion offish is not included.
Figure E-2. Conceptual exposure model for human receptors.
E-10
-------
Conceptual Exposure Model for Ecological Receptors
PRIMARY SECONDARY
PRIMARY RELEASE SECONDARY RELEASE MIGRATION EXPOSURE POTENTIAL
SOURCE(S) MKCHANISM(S) SOURCE(S) MECHANISM(S) PATHWAY(S) ROUTE(S) RECEPTOR(S)
Surface
Impound- -^
mcnt
^ Complete p£
-J. Runo 11? Erosion L M Watershed ^J Runoff/Erosion
(post-closure only)
S^ '^ Soil >
J >
.^1 Resuspensian L^. ^J droundwater
> Surface W'ller ^ f J J ' ^ Snrffv' ^
II
-Wl Kesuspension 1 > II
^1 1 ' Airhnrni- II
(post-closure only) Particulates Deposition/Diffusion
1 V ^
L>| Volatilization > Airborne >| DisDeision h1
rhwav (evaluated)
Notes:
1. If intermillenlly ITooded wetland, hvdric-soil invertebrates are inc
2. If intcrmiltcnlly Hooded wetland, aquatiu n vertebrates and iish a
y J Root uptake from soil J Ten-estrial plants
\ \ Reptiles
J Ing.bl.onol so.l J BMs
^ "I Ineest.onofterrestnal "1 Mamrmls
^ food items ^
^ J Root uptake from water J Aquatic plants
f Aquatic
_^. J Direct contact with water J invertebrates
|_ j Amphibians
L Fish
_^. J Direct contact with sediment J Benlhic
I. L invertebrates
f f Reptiles
L> J Ingestionot water I Birds
1 Inpestion or aquatic food 1 . , ,
fe H Mammals
^ items ^_
uded;
e not included.
Figure E-3. Conceptual exposure model for ecological receptors.
E-ll
-------
CHEMICALS
Organic chemicals (227)
Metals (17)
Nonmetallic inorganic
chemicals (8)
SOURCE TYPE
Surface impoundment
SOURCE TERM CHARACTERISTICS
Mass balance
Multiphase partitioning
Source degradation
SOURCE RELEASE MECHANISMS
Volatilization
Leaching
Runoff (post-closure, surface failure)
Erosion (post-closure, surface failure)
Particle resuspension (post-closure)
TRANSPORT MEDIA
Atmosphere
Watershed
Vadose zone
Groundwater
Surface water
FATE PROCESSES
Chemical/biological transformation
(and associated products of
transformation)
Linear partitioning
(water/air, water/soil, air/plant,
water/biota)
Nonlinear partitioning
(metals in vadose zone)
Chemical reactions/speciation
(mercury in surface waters)
INTERMEDIA CONTAMINANT FLUXES
Source
Source
Source surface soil
Air
Air
Watershed soil
Surface water
Vadose zone
Watershed soil
Groundwater
FOOD CHAIN/FOOD WEB
Air -»
Farm/habitat soil ->
Vegetation, soil, water ->
Surface water ->
RECEPTORS AND HABITATS
Air (volatilization, resuspension)
Vadose zone (leaching)
Local watershed soil (erosion, runoff)
Watershed/farm habitat soil (wet/dry
deposition, vapor diffusion)
Surface water (wet/dry deposition, vapor
diffusion)
Surface water (erosion, runoff)
Sediment (sedimentation)
Groundwater (infiltration)
Air (volatilization)
Surface water
Vegetation (particulate deposition; vapor
diffusion)
Vegetation (root uptake, translocation)
Animals (uptake)
Aquatic organisms (uptake)
Ecological Habitats:
Terrestrial
Freshwater aquatic
Wetland
Ecological Receptors:
Plants
Invertebrates
Amphibians
Reptiles
Human Receptors*:
Resident
Home gardener
Dairy farmer
Beef farmer
Fisher Birds
Mammals
*For each human receptor type, consider 5 age cohorts
EXPOSURE PATHWAYS
Human
Ingestion (plant, meat, milk, fish, water, soil, breast milk)
Inhalation (gases, particulates)
Ecological
Ingestion (plant, animal, water, soil)
Direct contact (surface water, sediment, soil)
HUMAN AND ECOLOGICAL RISK MEASURES
Cancer (risk probability)
Noncancer (hazard quotient)
Human: population
Ecological: population
Figure E-4. Dimensions of the 3MRA conceptual model for surface impoundments.
E-12
-------
E.2.5 Internal and External Stakeholder Process
The following details the stakeholder process utilized to try to achieve stakeholder acceptance of
the study methodology and thus, the results:
1) At the beginning of the study, EPA staff tasked with performing the work met with
stakeholders, prepared a Federal Register notice requesting comment on the proposed
study methodologies, and consulted with the Science Advisory Board.
2) Using the SAB consultation notes, EPA staff researched relevant background topics
and convened a group of technical experts (under a contract mechanism) to assist in study
design, which was then peer reviewed. Due to time limitations, a detailed written
methodology was not possible, so the study design concepts were conveyed to the peer
reviewers using a briefing format.
3) During the same time period as #2, stakeholders3 (all of whom had provided written
comments in response to the Federal Register notice) requested a face to face meeting to
learn the direction, and what scope of the project OSW envisioned. Since the study was
not part of a regulatory development process, per se, OSW was not constrained by the
Administrative Procedures Act, and was able to meet and talk freely with the
stakeholders.
4) Once the proposed study design methodology was presented to the SAB peer review
subcommittee, the initial planning and scoping process stopped, and OSW proceeded
with study implementation - about 20% of the way through the time budget allowed by
the statute.
5) Concurrent with #2 and #3 OSW ensured that Office-director level management
concurred with the proposed study objectives, and on scope issues (temporal scope,
geographic scope, industries to include vs. exclude).
6) Throughout the initial implementation, stakeholders (both affected industry
representatives and environmental groups) requested several face to face meetings and
had two more opportunities to provide public comment. OGC said the Federal Advisory
Committee Act was not a concern, since the stakeholders were the ones requesting the
meetings and OSW was not looking for consensus from them. OSW shared copies of
contractors' deliverables and this openness may have helped win stakeholder acceptance
of the study methodology. The stakeholders supported the survey instruments, which
may have helped in the OMB clearance process. OSW also informed state environmental
During the comment period, EPA received eight comments: three from trade associations (the Utilities Solid Waste
Activities Group, the Chemical Manufacturer's Association, the American Petroleum Institute), an industry representative (the
General Motors Corporation), two from electric utilities (Virginia Power and Central and South West Services), a combined
comment from a group of environmental organizations (the Environmental Defense Fund, the Green Environmental Coalition,
and the Montana Coalition for Health, Environment and Economic Rights), and a comment from the National Council of the
Paper Industry for Air and Stream Improvement, Inc. which did not directly address the questions in the Federal Register notice.
E-13
-------
agency managers about the study early in its design stage, and communicated directly
with states in the early implementation stage.
7) During the data collection, which was a critical study element, several of the data-
providing stakeholder groups organized workshops to encourage survey recipients (their
members) to provide data for the study. OSW also encouraged the survey recipients by
offering toll-free telephone assistance and putting answers to frequently asked questions
on EPA's internet site.
In conclusion, the Surface Impoundment Technical Plan provides excellent references
and examples of planning techniques for screening and analysis of risks based on conceptual
models and successful stakeholder interactions. It also provides a detailed rationale for
aggregating risks from industrial wastewater facilities.
E.2.6 Lessons Learned
1. The RCRA Study did not rely on the
guidance for planning and scoping, however,
the project did rely on a detailed plan which
included clear objectives, a conceptual
model, and analytical plan. The analytical
plan had a strong statistical basis for its
sampling and interpretation approach. The
statistical basis added quantification and
specificity to the plan.
The conceptual model showed how
stressors, pathways, and effects (both health
and ecological) would be combined and
presented. This presentation helped inform
reviewers of what was planned and enabled
EPA to learn from comments how the
scheme could be modified.
2. The project established and maintained a schedule for developing the analytical approach,
peer review, and analysis that was driven by court deadlines. While the initial schedule appeared
to be generous, ultimately, it helped define how long stakeholder deliberation would last.
3. The conceptual model showed how stressors, pathways, and effects (both health and
ecological) would be combined and presented. This presentation helped inform reviewers of
what was planned and enabled EPA to learn from comments how the scheme could be modified.
E-14
------- |