CENR/5-99/001
Ecological Risk Assessment
in the
Federal Government
Committee on Environment and Natural Resources
of the
National Science and Technology Council
May 1999
-------
About the National Science and Technology Council
President Clinton established the National Science and Technology Council (NSTC) by
Executive Order onNovember 23,1993. This cabinet-level council is the principal means for
the President to coordinate science, space, and technology policies across the Federal
Government The NSTC acts as a "virtual" agency for science and technology to coordinate the
diverse parts of the Federal research and development enterprise. The NSTC is chaired by the
President Membership consists of the Vice President the Assistant to the President for Science
anS Technology, Cabinet Secretaries and Agency Heads with significant science and technology
responsibilities, and other senior White House officials.
i ' 4 . I!"' ", IN i ! , ' ' . | I '.f i i
An important objective of the NSTC is the establishment of clear national goals for Federal
science and tecjinqjpgy investments in areas ranging from information technology and health
re^arch, to improving transportation systems and strengthening fundamental research. The
Council prepares research and development strategies that are coordinated across Federal
agencies to form an investment package that is aimed at accomplishing multiple national goals.
i
To obtain additional information regarding the NSTC, contact the NSTC Executive Secretariat at
(202)456-6102.
About the Committee on Environment and Natural Resources
agencies
The Committee on Environment and Natural Resources (CENR) is one
the NSTC, and is charged with improving coordination among Federal
environmental and natural resources research and development, establishing
between science and policy, and developing a Federal environment
research and development strategy that responds to national and international
of five committees under
involved in
a strong link
and natural resources
issues.
To obtain additional information about the CENR, contact the CENR Executive Secretary at
(202)482-5916.
-------
Ecological Risk Assessment
in the Federal Government
Committee on Environment and Natural Resources
National Science and Technology Council
-------
The purpose of this report is to review the major uses of ecological risk assessment by Federal
agencies and to provide case studies that illustrate established and potential applications of
ecological risk assessment in the Federal Government. The report presents the established roles
of ecological risk assessment in Federal decision making, highlights the ecological risk
assessment framework and terminology for its application to the various topics discussed, and
promotes the use of ecological risk assessment to address the wide array of environmental issues.
-------
Executive Office of the President
National Science and Technology Council
Committee on Environment and Natural Resources
Washington, D. C.
May 1999
Dear Colleague,
We are pleased to provide a copy of the report "Ecological Risk Assessment in the Federal
Government". Federal agencies face a major challenge in assessing and evaluating numerous
and varied ecological problems, ranging from the potential impacts of climate change to loss of
biodiversity, habitat destruction, and the effects of multiple chemicals on ecological systems.
Federal agencies have different responsibilities for addressing these problems: some have
regulatory functions, others serve as natural resource trustees, and some must address ecological
risks associated with their own activities. These differing responsibilities highlight the need for
flexible problem-solving approaches. Increasingly, ecological risk assessment is being suggested
as a useful tool for helping poliymaken: to address this wide array of ecological problems.
This report demonstrates the application of ecological risk assessment by Federal agencies to a
wide array of environmental issues and illustrates how other types of ecological and scientific
assessments might benefit through the use of ecological risk assessment approaches. Continued
progress in environmental protection requires the application of sound science to support risk
management and decision-making. This report indicates the utility of ecological risk assessment
for linking scientific information with Informed decision-making in the Federal government.
Sincerely,
Rosina Bierbaum
Co-Chair, CENR
Associate Director for Environment
Office of Science and Technology Policy
D. Jam£s Baker
Co-Chair,! CENR
Administrator, National Oceanic
and Atmospheric Administration
-------
-------
Committee on Environment and Natural Resources
Rosina Bierbaum, Co-Chair
White House
D. James Baker, Co-Chair
National Oceanic and Atmospheric Administration
Leonard Hirsch
Smithsonian Institution
Norine Noonan
Environmental Protection Agency
Martha A. Krebs
Department of Energy
Ghassem Asrar
National Aeronautics and Space Administration
Joseph Bordogna
National Science Foundation
Eileen Kennedy
Department of Agriculture
Elwood Holstein
Office of Management and Budget
Mark Schaefer
Department of the Interior
Kenneth Olden
Department of Health and Human Services
Albert Eisenberg
Department of Transportation
Paul Leonard
Department of Housing and Urban Development
Delores M. Etter
Department of Defense
Melinda L. Kimble
Department of State
Craig Wingo
Federal Emergency Management Agency
Kathryn J. Jackson
Tennessee Valley Authority
Samuel Williamson
Office of the Coordinator for Meteorology
Terrance J. Flannery
Central Intelligence Agency
George Frampton
Council oh Environmental Quality
Subcommittees
Air Quality
Martha A. Krebs, DOE, Chair
Dan Albritton, NOAA, Vice Chair
Bob Perciasepe, Vice Chair
Ecological Systems
Mark Schaefer, DOI, Chair
Mary Clutter, NSF, Vice Chair
Donald Scavia, NOAA, Vice Chair
Global Change Research
Robert W. Corell, NSF, Chair
Ghassem Asrar, NASA, Vice Chair
Mike Dombeck, USDA, Vice Chair
Natural Disaster Reduction
William Hooke, NOAA, Chair
John Filspn, USGS, Vice Chair
Craig Wingo, FEMA, Vice Chair
Toxics and Risk
Norine Noonan, EPA, Chair
Kenneth Olden, NIEHS, Vice Chair
Bob Foster, DOD, Vice Chair
-------
Ecological Risk Assessment Task Group Members
Task Group Co-Chairs
Randall Wentsel, EPA (formerly U.S. Army)
William Sommers, USFS
William van der Schalie, EPA
Chapter 2. Ecological Risks of a New
Industrial Chemical Under TSCA
Maurice Zeeman, EPA (Leader)
Donald Rodier, EPA
J.VinceNabholz,EPA
Chapter 3. Ecological Risk Assessment
Under FIFRA
Anthony Maciorowski, EPA (Leader)
Chapter 4. Nonindigenous Species
Richard Orr, USDA (Leader)
Gwendolyn McClung, EPA
Robert Peoples, USFWS
James D. Williams, USGS
Michael A. Meyer, NASA
Chapters. CERCLA
Randall Wentsel, EPA (formerly U.S. Army)
(Leader)
David Charters, EPA
Mark Sprenger, EPA
Stephen Ells, EPA
John Basietto, DOE
Nancy Finley, USFWS
Alyce Fritz, NOAA
Mary Matta, NOAA
Chapter 6. Agricultural Ecosystems
Susan Ferenc, USDA (Leader)
S. Ronald Singer, USFWS
Evert Byington, EPA
Chapter 7. Endangered/Threatened Species
i
David Harrelson, USFWS (Leader)
Chapter 8. Ecological Assessments in
Ecosystem Management
William Sommers, USFS (Leader)
Suzanne Marcy, EPA
William Van der Schalie, EPA
Gene Lessard, USFS
Thomas Quigley, USFS
Robert Lackey, EPA
David Cleaves, USFS
Edward Novak, DoD
Charles van Sickle, USFS
John Wuichet, U.S. Army
Chapter 9. The Use of Ecological Risk
Assessment Following the Accidental
Release of Chemicals
James Andreasen, EPA (Leader)
Nancy Finley, USFWS
i
Key to Agency Abbreviations:
DoD: U.S. Department of Defense
DOE: U.S. Department of Energy
EPA: U.S. Environmental Protection Agency
NASA: National Aeronautics and Space
Administration
NOAA: National Oceanic and Atmospheric
Administration/U.S. Department of
Commerce
USDA: U.S. Department of Agriculture
USFS: U.S. Forest Seryice/U.S. Department
of Agriculture
USFWS: U.S. Fish and Wildlife Service/U.S.
Department of the Interior
USGS: U.S. Geological Survey/U.S.
Department of the Interior
-------
CONTENTS
LISTS OF TABLES AND FIGURES ' xvl
PREFACE • xvii
EXECUTIVE SUMMARY
1. INTRODUCTION : M
1.1. ECOLOGICAL RISK ASSESSMENT I'1
1.2. CASE STUDY OVERVIEW , 1-6
1.3. REFERENCES M2
ESTABLISHED USES
2. ECOLOGICAL RISKS OF A NEW INDUSTRIAL CHEMICAL UNDER TSCA 2-1
2.1. SUMMARY i 2~l
2.2. INTRODUCTION 2'2
2.2.1. EPA/OPPT Risk Assessment Approach ..., 2'2
2.2.2. Statutory and Regulatory Background j 2-6
2.3. CASE STUDY: DESCRIPTION OF A NEW CHEMICAL ASSESSMENT
UNDER TSCA i 2~6
2.3.1. Background Information and Objective ...; 2-7
2.3.1.1. Chemistry Report ! 2-8
2.3.1.2. Engineering Report 2-8
2.3.1.3. Exposure Assessment ! 2-8
2.3.1.4. Ecological Hazard Assessment 2-8
2.3.1.5. Ecological Risk Assessment 2-9
2.3.2. Problem Formulation , 2"9
2.3.2.1. Stressor Characteristics .... 2-9
2.3.2.2. Ecosystem Potentially at Risk .. j 2-9
2.3.2.3. Ecological Effects 2-10
2.3.2.4. Assessment Endpoints ....,....: 2-10
2.3.2.5. Measurement Endpoints J 2-11
2.3.2.6. Conceptual Model J 2-12
2.3.3. Analysis, Risk Characterization, and Risk
Management—First Iteration 2-12
2.3.3.1. Characterization of Exposure ... i 2-12
2.3.3.2. Characterization of Ecological Effects—
Stressor-Response Profile 2-14
2.3.3.3. Risk Characterization '. 2-14
2.3.3.4. Risk Management i. 2-17
2.3.4. Analysis, Risk Characterization, and Risk
Management—Second Iteration ;. 2-17
IX
-------
CONTENTS (continued)
2.3.4.1. Characterization of Exposure 2-17
2.3.4.2. Characterization of Ecological Effects 2-17
2.3.4.3. Risk Characterization , 2-17
2.3.4.4. Risk Management 2-18
2.3.5. Analysis, Risk Characterization, and Risk
Management—Third Iteration 2-19
2.3.5.1. Characterization of Exposure 2-19
2.3.5.2. Characterization of Ecological Effects 2-19
2.3.5.3. Risk Characterization: Risk Estimation and
Uncertainty Analysis 2-19
2.3.5.4. Risk Management 2-20
2.3.6. Analysis., Risk Characterization, and Risk
Management—Fourth Iteration 2-20
2.3.6.1. Characterization of Exposure 2-20
2.3.6.2. Risk Characterization 2-20
2.3.6.3. Risk Management 2-21
2.3.7. Analysis, Risk Characterization, and Risk
Management—Fifth Iteration 2-21
2.3.7.1. Characterization of Ecological Effects 2-21
2.3.7.2. Risk Characterization—Risk Estimation 2-21
2.3.7.3. Risk Management—Final Decision 2-24
2.3.8. Discussion of Case Study 2-24
2.3.9. Summary of Case Study 2-24
2.4. RISK ASSESSMENT METHODOLOGY DEVELOPMENT 2-26
2.5. RISK MANAGEMENT 2-26
2.6. REFERENCES 2-27
3. ECOLOGICAL RISK ASSESSMENT UNDER FIFRA 3-1
3.1. SUMMARY 3_1
3.2. INTRODUCTION ' 3_1
3.3. REGULATORY CONTEXT FOR PESTICIDE REGISTRATION AND
REREGISTRATION 3-2
3.4. RISK MANAGEMENT 3_2
3.5. RISK ASSESSMENT METHODS IN PESTICIDE
REGULATORY OPERATIONS 3-4
3.5.1. Application of Ecological Risk Assessments in
Pesticide Regulatory Decision Making 3-7
3.5.2. The Risk Identification and Mitigation Process 3-8
3.6. RISK ASSESSOR AND RISK MANAGER COMMUNICATION 3-9
3.7. NEXT STEPS 3_10
3.8. REFERENCES 3_10
-------
CONTENTS (continued)
4. NONINDIGENOUS SPECIES ! 4-1
4.1. SUMMARY •: 4-1
4.2. INTRODUCTION \ 4-2
4.2.1. Definition and Scope of Risk Analyses ', 4-2
4.2.2. Relationship to EPA's Ecological Risk Assessment Framework 4-3
4.2.3. Federal Agencies Involved in Nonindigenous Species Risk Issues 4-4
4.3. DISCUSSION ON THE STATE OF THE PRACTICE 4-4
4.4. CASE STUDIES • 4-5
4.4.1. Risk Assessment on Black Carp (Pisces: Cyprinidae) 4-5
4.4.1.1. Probability of Establishment .....! 4-7
4.4.1.2. Consequences of Establishment...' 4-8
4.4.2. Risk Assessment for the Release of Recombinant Rhizobia
at a Small-Scale Agricultural Field Site ....; 4-10
4.4.2.1. Problem Formulation : 4-11
4.4.2.2. Analysis: Characterization of Exposure 4-11
4.4.2.3. Analysis: Characterization of Ecological Effects 4-12
4.4.2.4. Risk Characterization : 4-13
4.4.2.5. Risk Verification 4-13
4.4.3. Scenario Analysis for the Risk of Pine Shoot Beetle Outbreaks 4-14
4.4.3.1. Assessment Summary 4-15
4.4.3.2. Risk Management Summary ; 4-17
4.5. NEXT STEPS ' 4-18
4.6. REFERENCES 4-19
5. CERCLA '• 5-1
5.1. SUMMARY '. 5-1
5.2. INTRODUCTION : 5-1
5.3. RISK MANAGEMENT '. 5-3
5.4. CASE STUDIES AND EXAMPLES 5-5
5.4.1. Linden Chemicals and Plastics Wildlife Assessment 5-5
5.4.1.1. Problem Formulation , 5-5
5.4.1.2. Hazard Characterization 5-5
5.4.1.3. Assessment Endpoints and Testable Hypotheses 5-6
5.4.1.4. Conceptual Model 5-6
5.4.1.5. Food Chain Model Assumptions .' 5-7
5.4.1.6. Sources of Uncertainty 5-8
5.4.1.7. Clapper Rail Tissue Evaluation 5-8
5.4.1.8. Hazard Quotient Results 5-9
5.4.1.9. Risk Assessment Conclusions ... • 5-9
5.4.2. United Heckathorn Assessment 5-10
5.4.2.1. Site History and Background 5-10
5.4.2.2. Problem Formulation and Conceptual Model 5-10
XI
-------
CONTENTS (continued)
5.4.2.3. Risk Characterization 5-11
5.4.2.4. Conclusions 5-13
5.4.3. Metal Bank of America 5-13
5.4.3.1. Site History and Background 5-13
5.4.3.2. Problem Formulation and Conceptual Model 5-14
5.4.3.3. Measurement Endpoints and Approach 5-15
5.4.3.4. Risk Characterization 5-16
5.4.3.5. Conclusion 5-17
5.4.4. Data Quality Objectives Process 5-18
5.5. NATURAL RESOURCE DAMAGE ASSESSMENT AND ECOLOGICAL
RISK ASSESSMENT 5-19
5.5.1. What Is Damage Assessment? 5-19
5.5.2. Contrasts Between Ecological Risk Assessment and
Damage Assessment 5-20
5.5.3. Requirement for Coordination of Assessments 5-21
5.6. RISK ASSESSMENT METHODOLOGY DEVELOPMENT 5-21
5.7. SITE REMEDIATION AND THE ROLE OF ECOLOGICAL
RISK ASSESSMENT 5-21
5.8. REFERENCES 5-23
POTENTIAL USES OF ECOLOGICAL RISK ASSESSMENT
6. AGRICULTURAL ECOSYSTEMS 6-1
6.1. SUMMARY 6-1
6.2. INTRODUCTION 6-3
6.2.1. Historical and Current Use of Risk Assessment in
Agricultural Production 6-4
6.2.1.1, Environmental Impacts of Production Practices 6-4
6.2.1.2. Environmental Impacts on Production 6-5
6.2.1.3. Environmental Impacts of Aquaculture 6-6
6.2.2. Applicability of EPA Ecological Risk Assessment Framework and
Guidelines to Agricultural Ecosystems 6-7
6.3. CASE STUDIES ; 6-8
6.3.1. Risk Assessment of USDA Conservation Programs 6-8
6.3.1.1. Environmental Quality Incentives Program 6-9
6.3.1.2. Conservation Reserve Program 6-17
6.3.2. Report on the Ecological Impacts of Nonindigenous Shrimp Viruses 6-19
6.3.2.1. Background 6-19
6.3.2.2. Management Goals 6-20
6.3.2.3. Problem Formulation 6-20
6.3.2.4. Analysis and Risk Characterization 6-22
6.3.2.5. Summary 6-22
XII
-------
CONTENTS (continued)
6.4. RISK ASSESSMENT METHODOLOGY DEVELOPMENT 6-23
6.5. RISK MANAGEMENT .., 6-24
6.6. NEXT STEPS 6-25
6.7. REFERENCES •• 6-25
7. ENDANGERED/THREATENED SPECIES 7-1
7.1. SUMMARY ; 7-1
7.2. THE ENDANGERED SPECIES ACT OF 1973 .., . 7-2
7.2.1. Purpose 7-2
7.2.2. Listing ! 7-3
7.2.3. Species ; 7-3
7.2.4. Candidate Species : 7-3
7.2.5. Recovery ; 7-3
7.2.6. Consultation .' 7-4
7.2.7. Critical Habitat 7-4
7.2.8. International Species 7-4
7.2.9. Exemptions ', 7-4
7.2.10. Habitat Conservation Plans J 7-4
7.2.11. Definition of "Take" 7-5
7.3. ESTIMATING RISK '. 7-5
7.3.1. Estimating the Risk of Extinction 7-6
7.3.1.1. Sources of Risk , 7-6
7.3.1.2. Focusing Conservation Efforts 7-16
7.3.1.3. Distribution of Extinction Times 7-17
7.3.2. Limitations of Our Ability To Estimate Risk 7-18
7.4. CONCLUSIONS AND RECOMMENDATIONS .; 7-18
7.5. REFERENCES , 7-19
RELATED SCIENTIFIC ASSESSMENTS
8. ECOLOGICAL ASSESSMENTS IN ECOSYSTEM MANAGEMENT 8-1
8.1. SUMMARY \ 8-1
8.2. INTRODUCTION 8-2
8.3. CASE STUDIES AND EXAMPLES 8-7
8.3.1. Interior Columbia River Basin Scientific Assessment 8-7
8.3.1.1. Framework •. 8-7
8.3.1.2. Integrated Scientific Assessment .;. 8-8
8.3.1.3. Ecosystem Integrity 8-9
8.3.1.4. Composite Ecological Integrity 8-9
8.3.1.5. Socioeconomic Resiliency , 8-10
8.3.1.6. Findings From the Future Management Options 8-10
8.3.2. The Southern Appalachian Assessment 8-10
xni
-------
CONTENTS (continued)
8.3.3. EPA Watershed Assessments 8-14
8.3.3.1. Background 8-14
8.3.3.2. Process 8-15
8.3.3.3. Watershed Case Study Selection 8-15
8.3.3.4. Case Study Teams 8-16
8.3.3.5. Characteristics of Selected Watersheds 8-16
8.3.3.6. Resources to Support Case Study Development 8-17
8.3.3.7. Lessons Learned 8-17
8.3.3.8. Reviews and Current Status 8-18
8.3.4. Examples of U.S. Department of Defense Activities in Ecological
Assessments 8-19
8.3.4.1. DoD's Ecosystem Management Policy 8-19
8.3.4.2. Site Examples 8-20
8.4. RISK ASSESSMENT METHODOLOGY DEVELOPMENT .. „ 8-22
8.4.1. Expanded Use of the EPA Guidelines „ 8-22
8.4.2. Technical and Research Challenges „ 8-23
8.5. RISK ASSESSMENT IN ECOSYSTEM MANAGEMENT
DECISION MAKING 8-23
8.5.1. The Risk Management Cycle and Ecosystem Management 8-23
8.5.2. Risk Management and Decision Quality 8-25
8.5.3. Risk Assessment as a Decision Aid 8-25
8.5.4. Expert Judgment in Risk Management 8-27
8.5.5. Risk Evaluation, Adjustment, and Decision Quality 8-28
8.5.6. Risk Communication and Decision Quality 8-28
8.6. NEXT STEPS 8-29
8.7. REFERENCES 8-29
9. THE USE OF ECOLOGICAL RISK ASSESSMENT FOLLOWING
THE ACCIDENTAL RELEASE OF CHEMICALS '. 9-1
9.1. SUMMARY 9-1
9.2. INTRODUCTION AND LEGISLATION 9-1
9.3. USE OF THE RISK ASSESSMENT PROCESS IN
ACCIDENTAL RELEASES 9-3
9.4. EXAMPLES OF ACCIDENTAL RELEASES 9-5
9.4.1. Case Study: Patricia Sheridan Release of Contaminated Dredge Material . 9-5
9.4.2. Types of Accidental Releases 9-9
9.4.2.1. John Day River Acid Spill 9-9
9.4.2.2. North Cape Oil Spill 9-12
9.4.2.3. Conoco Marine Terminal 1,2-Dichloroethane Spill 9-14
9.5. RISK ASSESSMENT METHODOLOGY AS APPLIED TO
ACCIDENTAL RELEASES 9-15
9.6. NEXT STEPS 9-17
xiv
-------
CONTENTS (continued)
9.6.1. Ecological Risk Assessment Needs
9.6.2. Contingency Planning
9.6.3. Research on Cleanup Methods ...
9.7. REFERENCES
9-17
9-17
9-17
9-18
10. GLOSSARY 10-1
xv
-------
LIST OF TABLES
1-1. Case studies summary 1_7
2-1. Physical/chemical properties of PMN substance 2-13
2-2. Predicted environmental concentrations (PECs) for PMN
substance (u/L or ppb) 2-14
2-3. PMN substance initial stressor-response profile 2-15
2-4. Summary of five risk characterization iterations 2-15
2-5. OPPT assessment factors used in setting "concern levels" for
new chemicals 2-16
2-6. PDM3 analysis 2-18
2-7. Predicted stressor-response profile for benthic organisms 2-20
2-8. EXAMS E analysis 2-20
2-9. Stressor-response profile for Chironomus tentans 2-21
3-1. Generalized exposure analysis and assessment methods and procedures
used in prospective ecological risk screens of pesticides 3-5
3-2. Generalized ecological effects analysis and risk quotient methods
and procedures used in prospective risk screens of pesticides 3-6
4-1. Frequency of outbreaks of the pine shoot beetle and years between outbreaks 4-18
5-1. Use of ecological data in Records of Decision (RODs) in 1995 5-3
5-2. Clapper rail mercury and PCB tissue levels 5-8
5-3. Measurement endpoints and approach 5-12
5-4. Mean and upper 95% confidence limit (CL) concentrations (mg/kg) of
total PCBs in sediments near the Metal Bank of America site normalized to
dry weight and total organic carbon (TOC) 5-16
LIST OF FIGURES
1-1. Framework for ecological risk assessment (U.S. EPA, 1998) 1-4
2-1. Structure of assessment for effects of a PMN substance 2-4
2-2. Flow chart and decision criteria for the ecological risk assessment of a
PMN substance 2-5
4-1. Risk assessment model from the Report to the Aquatic Nuisance
Species Task Force 4.5
4-2. Combined scenarios for new outbreaks of pine shoot beetle due to the
movement of logs 4.15
6-1. Conceptual diagram of soil/land disturbances 6-11
6-2. Conceptual diagram of irrigation water application 6-12
6-3. Map of cropland acres with conservation needs 6-14
6-4. Map of potential fertilizer loss from farm fields 6-15
8-1. Ecosystem management model 8-4
xvi
-------
PREFACE
This report was prepared by an. interagency work group under the auspices of the
Committee on Environment and Natural Resources (CENR). CENR is charged with improving
coordination among Federal agencies 'involved in environmental and natural resources research
and development, establishing a strong link between science and policy, and developing a
Federal environment and natural resources research and development strategy that responds to
national and international issues. CENR is one of five committees under the National Science
and Technology Council, which was established by President Clinton hi November 1993 as a
cabinet-level council to coordinate science, space, and technology policies across the Federal
Government.
A key issue across the Federal Government is how to evaluate numerous and varied
ecological problems, ranging from potential global climate change to loss of biodiversity, habitat
destruction, and the effects of multiple chemicals on ecological systems. Numerous Federal
agencies have different responsibilities for addressing these problems: Some have regulatory
functions, others serve as natural resource trustees, and some must address ecological risks
associated with their own activities. Ifhese differing responsibilities highlight the need for
flexible problem-solving approaches. Increasingly, ecological risk assessment is being suggested
as a way to address this wide array of ecological problems.
To explore the uses and applicability of ecological risk assessment across the Federal
Government, CENR sponsored workshops in October 1994 and December 1995 to promote
information exchange. The development of this report was initiated as a follow-on to these
workshops to review the major uses of ecological risk assessment by Federal agencies. This
report provides examples of ecological risk assessments conducted hi the Federal Government as
well as ecological assessments that could benefit from ecological risk assessment methodologies.
Recommendations for improving ecological risk assessment are made in the Executive Summary
and at the end of each chapter. The authors would like to thank Michael Rodemeyer, Assistant
Director, Environment, White House Office of Science and Technology Policy (OSTP); Fran
Sharpies, OSTP; and Jim Kariya, U.S. EPA; for their significant contributions to this document.
xvn
-------
EXECUTIVE SUMMARY
i
Ecological risk assessment is a process for organizing and analyzing data, assumptions,
and uncertainties to evaluate the likelihood of adverse ecological effects. The Committee on
i i
Environment and Natural Resources, Subcommittee on Risk Assessment, approved the formation
; I
of an ecological risk assessment work group to write a document to review the major uses of
ecological risk assessment by Federal agencies. Eight task groups were formed with a total of 32
scientists from 9 Federal agencies. The task groups provided examples of current ecological risk
assessment areas (established uses), potential uses where components of ecological risk
assessment are used, and related ecological assessments and other scientific evaluations that
might benefit from the use of ecological risk assessment methodologies. Established uses
included the Toxic Substances Control Act (TSCA); the Federal Insecticide, Fungicide, and
Rodenticide Act (FIFRA); nonindigenous species; and the Comprehensive Environmental
Response, Compensation, and Liability Act (CERCLA). Potential uses included agricultural
ecosystems and endangered/threatened species. Related scientific assessments include oil spills
(accidental releases), and ecosystem management. The work group members generally agreed
that the U.S. Environmental Protection Agency's paradigm for ecological risk assessment and the
associated terminology were a common scientific base from which to address the variety of uses
of ecological risk assessment.
This publication demonstrates the broad applicability of ecological risk assessment as a
flexible, problem-solving paradigm that can support environmental decision-making across the
Federal Government. This document assists in communicating the process by which ecological
risk assessments are performed to scientists and environmental policy makers with some
technical expertise who are unfamiliar with ecological risk assessment. Communicating the
process used for scientific risk assessments performed within the Federal sector is important for
expanding the use of ecological risk assessment into areas where the potential benefits of such an
approach have not yet been realized.
To enhance multiagency coordination, the following recommendations need to be
addressed by the CENR agencies:
• Leverage technical advancements made in one area to fields where ecological risk
assessment is less developed.
• Expand ongoing, multiagency dialogs between agencies and with outside researchers in
order to develop procedures and tools (e.g., workshops, formation of ad hoc groups) for
conducting ecological risk assessment and defining ecological criteria and indicators.
xvm
-------
Address nonindigenous species issues across agency boundaries to expand technical
expertise, focus resources, and reduce redundancy. Promote regional and global
management strategies to address nonindigenous species issues.
xix
-------
-------
1. INTRODUCTION
The ecological problems facing environmental scientists and decision makers are
numerous and varied. Growing concern over potential global climate change, loss of
biodiversity, acid precipitation, habitat destruction, and the effects of multiple chemicals on
ecological systems has highlighted the need for flexible problem-solving approaches that can link
ecological measurements and data with the decision-making needs of environmental managers.
Increasingly, ecological risk assessment is being suggested as a way to address this wide array of
ecological problems.
Scientific publications and presentations at professional meetings on ecological risk
assessment topics have greatly increased in the past few years. Various organizations have
proposed standardized ecological risk assessment paradigms (e.g., U.S. EPA, 1998; NRC, 1993),
and guidance on conducting ecological risk assessments has been or is being developed by
national standardization organizations, States, Federal Government agencies, and other countries
and international organizations. While these efforts have resulted in widespread agreement on
the general ecological risk assessment process, there is still considerable variation in the way the
process is applied in specific situations.
The objectives of this report are to provide examples of the existing uses of ecological
risk assessment by Federal agencies as well as to illustrate how other types of ecological and
scientific assessments used in the Federal Government might benefit through the use of
ecological risk assessment approaches. The report highlights the use of ecological risk
assessment to address a wide array of environmental issues; The intended audience for this
document is scientists and environmental policy makers with some technical expertise who are
unfamiliar with ecological iisk assessment. Ecological risk assessors may find chapters of the
document outside their area of expertise to be of interest.
This introductory section describes ecological risk assessment to those who may be
unfamiliar with the process (1.1), and provides a case study overview (1.2).
1.1. ECOLOGICAL RISK ASSESSMENT
In this report, we use the ecological risk assessment process as described in the U.S.
Environmental Protection Agency's (EPA's) recently published ecological risk assessment
guidelines (U.S. EPA, 1998) as a benchmark for comparisons among the case studies. The
guidelines were prepared by a panel of representatives from across EPA, organized by the
Agency's Risk Assessment Forum. They are the product of nine years of development and peer
review. Preliminary work on guidelines development began in 1989 and included a series of
1-1
-------
colloquia sponsored by EPA's Risk
Assessment Forum to identify and discuss
significant issues in ecological risk
assessment. Based on this early work and on
a consultation with EPA's Science Advisory
Board (SAB), EPA decided to produce
ecological risk guidance sequentially,
beginning with basic terms and concepts and
continuing with the development of source
materials for the guidelines. The first product
of this effort was the Risk Assessment Forum
report, Framework for Ecological Risk
Assessment (Framework Report; U.S. EPA,
1992), which proposed principles and
terminology for the ecological risk
assessment process. Since then, other
materials were developed, including
suggestions for guidelines structure,
ecological assessment case studies, and a set
of issue papers that highlighted important
principles and approaches for EPA scientists
to consider in preparing the guidelines. The
final guidelines are a product of all these
materials and were revised to reflect
comments received from peer reviewers, the
SAB, and the public.
Ecological risk assessment is a process for organizing and analyzing data, assumptions,
and uncertainties to evaluate the likelihood of adverse ecological effects that may occur or are
• ' ' '
occurring as a result of exposure to one or more stressors. Stressors can be chemical, physical
(e.g., habitat destruction), or biological (e.g., introduced species). Ecological risk assessment is
helpful for environmental decision making because it provides risk managers with an approach to
i i
consider available scientific information along with other important factors to select a course of
action. (Definitions of many of the terms used in this section are provided in Section 10, the
glossary.)
Ecological Risk Assessment and
Environmental Decision Making
Ecological risk assessment "is a useful risk
management tool feat:
* Highlights the greatest risks, which is
helpful for allocating limited, resources;
« Allows decision makers to ask 'what if
questionss regarding the consequences of
various potential management actions;
• Facilitates explicit identification of
environmental values of concern; and
» Identifies critical knowledge gaps,
thereby helping to-pdoritize future
research needs" (SETAC, 1997).
While there are many different applications
in which ecological risk assessmentsMare used
{e.g.,, regulation of hazardous waste sites,
Indttstrial chemicals* pesticides,,, or introduced
species), more iian the results of the risk
assessment is jjteeded to mate effltvir treatment
options, pollution, prevention), or legal
mandates.
1-2
-------
Ecological risks are estimated by integrating exposure (the interaction of stressors and
ecological receptors) and effects. All risk assessments involve some degree of uncertainty.
Some elements of uncertainty can be reduced by gathering additional data; others cannot, such as
the inherent variability in rainfall amounts or temperature fluctuations. Uncertainty analysis
describes the degree of confidence in the assessment and can help risk managers focus
future research on areas that will lead to the greatest reduction in uncertainty.
As shown in Figure 1-1, ecological risk assessment includes three primary phases:
problem formulation, analysis, and risk characterization. Problem formulation is the initial phase
of the process, which includes the development of assessment endpoints, conceptual models, and
an analysis plan. Assessment endpoints are explicit expressions of the actual environmental
value that is to be protected that link the risk assessment to management concerns. Assessment
endpoints include both a valued ecological entity and an attribute of that entity that is important
to protect and is potentially at risk (e.g., nesting and feeding success of piping plovers or areal
extent and patch size of eelgrass). Potential interactions between assessment endpoints and
stressors are explored by developing conceptual models that link anthropogenic activities with
stressors and evaluate interrelationships among exposure pathways, ecological effects, and
ecological receptors. The analysis plan justifies what will be done as well as what will not be
done in the assessment, describes the data and measures to be used in the risk assessment, and
indicates how risks will be characterized.
The analysis phase, which follows problem formulation, includes two principal activities:
characterization of exposure and characterization of ecological effects. The process is flexible,
and interaction between the ecological effects and exposure evaluations is critical. Both
activities include an evaluation of available data for scientific credibility and relevance to
assessment endpoints and the conceptual model. In exposure characterization, data analyses
describe the source(s) of stressors, the distribution of stressors in the environment, and the
contact or co-occurrence of stressors with ecological receptors. In ecological effects
characterization, data analyses may evaluate stressor-response relationships or evidence that
exposure to a stressor causes an observed response. The products of analysis are summary
profiles that describe exposure and the stressor-response relationships.
Risk characterization is the final phase. During risk characterization, risks are estimated
and interpreted and the strengths, limitations, assumptions, and major uncertainties are
summarized. Risks are estimated by integrating exposure and stressor-response profiles using a
wide range of techniques, such as comparisons of point estimates or distributions of exposure
and effects data, process models, or empirical approaches such as field observational data.
1-3
-------
Planning
(Risk Assessor/
Risk Manager/
Interested Parties
Dialogue)
Ecological Risk Assessment
Characterization
of
Ecological
Effects
Characterization
of
Exposure
Communicating Results
to the Risk Manager
Figure 1-1. Framework for ecological risk assessment (U.S. EPA, 1998),,
1-4
-------
Risk assessors describe risks by evaluating the evidence supporting or refuting the risk
estimate(s) and interpreting the adverse effects on the assessment endpoint. Criteria for
evaluating adversity include the nature and intensity of effects, spatial and temporal scales, and
the potential for recovery. Agreement among different lines of evidence of risk increases
confidence in the conclusions of a risk assessment.
Several important activities are shown
outside the risk assessment process in Figure
1-1, including discussions between risk
assessors and risk managers. Interactions
between risk assessors and risk managers at
the beginning and end of the risk assessment
are critical for ensuring that the results of the
assessment can be used to support a
management decision. Planning activities at
the outset of a risk assessment foster
agreements between risk assessors and risk
managers concerning the management goals,
risk assessment purpose, and resources
available to conduct the assessment. Other
interested parties also may be involved with
planning (see text box). The box following
risk characterization represents
communication of the risk assessment results
from assessors to managers.
The bar along the right side of Figure
1-1 highlights data acquisition, iteration, and
monitoring. Monitoring data can provide
important input to all phases of the risk
assessment process. Monitoring data can also
provide the impetus for a risk assessment by
identifying changes hi ecological condition,
or can be used to evaluate risk assessment
predictions, such as the success of mitigation
or source reduction efforts or the extent and nature of any ecological recovery that may occur.
The ecological risk assessment process is frequently iterative, and new data or information may
The Role of InterestedJParties if PtaaHg
In some risk assessments* interested parties -
also take an active role in planning»
particularly in goal development Interested
parties (commonly calledStakeholders'*)
may include Federal, State* tribal; and
municipal governments,, mdusMal leaders,,
environmental groupst smal»business
owners, landowners., and other segments of
society concerned, about an environmental
issue at hand: or attempting t
-------
require revisiting a part of the process or conducting a new assessment. Some assessments are
i • i
designed in tiers, which are preplanned sets of assessments of progressive data and resource
intensity. The outcome of each tier is either a management decision or the initiation of the next
tier.
1.2. CASE STUDY OVERVIEW
To prepare this report, task groups of Federal scientists selected and identified case
studies and other examples representing the diversity of ecological assessments commonly
conducted hi the Federal Government. The eight different types of assessments included hi this
i
report are summarized in Table 1-1. The assessments involve numerous Federal and State
agencies as well as many nongovernmental organizations and were done in response to a range of
statutory and nonstatutory requirements. Chemical, physical, and biological stressors were
included hi one or more of the assessments, as were a wide range of ecological systems.
The case studies are divided into three categories: established and potential applications
of risk assessment and related scientific assessments. The established case studies follow most
of the major elements of EPA's ecological risk assessment guidelines (U.S. EPA, 1998), while
the potential applications present varying degrees of implementation of ecological risk
assessment, ranging from ecological assessments that could benefit from ecological risk
assessment methods to those that have begun to apply the ecological risk assessment framework.
Three cases in the established category are primarily concerned with chemical stressors
(chemical premanufacturing notification, pesticide registration, and hazardous waste sites) and
t [
incorporate tiered assessments approaches that proceed from simple, relatively inexpensive
: ! i
assessments to more complex and costly assessments as necessary to provide: a level of certainty
sufficient to support a management decision. Initial screening assessments often involve using a
hazard quotient, which is the ratio between an exposure concentration and an effects
concentration. Quotients are most useful for categorizing risks as high or low.
Hazardous waste site assessments at U.S. Department of Energy facilities (Section 5) use
the data quality objective (DQO) process in conjunction with ecological risk assessment. The
DQO process is similar to the planning and problem formulation stages of an ecological risk
assessment and emphasizes determining the boundaries of a study as well as evaluating the
quality and quantity of the data necessary for the study. Another variation used at hazardous
i i
waste sites involves natural resource damage assessments, where emphasis is on demonstrating
actual rather than potential ecological damage. In this case, ecological risk assessment is
important for establishing a causal link between site contaminants and adverse effects.
1-6
-------
I
cu
Ecosystem t
ill
SS*
« „
I s
J&j 01 bD
(2 ia &,
a '" ^
So
'^ bu >•
<="•< « .S
Case study
(U <~v
•^"S.
Assessment
(report cha
*•»
§
a
en
EW
(»
03
Wl
Ui
03
CJ
'oil
o
8
u
0
S
•a
«^
en
J=
C3
en
§
1
a
"3
a
S
•s
^
^
H
§
New chemical
d
& S3
(U *T3
C 3
Ecological risks of a
industrial chemical v
(Chapter 2)
s
•3
u
1
§
-3
1
fc
^
P"t
h
1
Synthetic pyrethroids
1
1
§
Ecological risk asses
FIFRA (Chapter 3)
1
H
H
•3
S
1
_
1
*
i
fe
Black carp
•3
u
o
$
H
1
Recombinant rhizobia
"3
•5b
(^
<
PH
1
D
Pine shoot beetle
.S
Nonindigenous Spec
(Chanter 41
v-' — i — - •;
^
1
1
•3
S
•3
o
•J
s
Pi
w
o
1
??
•«
c§
EC
Linden Chemicals and
United Heckathorn
1
Metal Bank of America
/_s
CERCLA (Chapter f
1-7
-------
i
B „
P
.off
CO
^
« ° ^
a a 5
§ -S F
|||
P^€
Primary
agencies
involved"
Case study
t:
o
0,
&
f*c*
** os1
g-S
1
S
S3
eg
1
%
ationsof ecologi
.g
ex
ex
C3
^^
.2
B
I
H"
||
°
•a
1
' ^i7
§11 ^
1
Envkonmental Quality
Incentives Program
Conservative Research
Progam
1
•a
o
1
o
CQ
1
&
§1
en
0>
s
l^
*^
Ecological Impacts of
Nonindigenous Shrimp
•
1
S
§
"s o
It
™ 5
^
.8
1
00
1
1
!
Extinction models
S
I
n
I
t^.
It
la n
W v_
-ffi
S
en
w
ed scientific asst
«
5
It
&. 3
•a
§
'So
,2
as
PQ
CO
ID
.S
c9
OQ
Interior Columbia Rivei
Scientific Assessment
H
n
*
s
If!
d, S o
1
12 Federal
and 3 State
agencies
Southern Appalachian
Assessment
i
1
s^
J-3 oo
O j=
W v_
1
"B
>§
u
*a
-H
Patricia Sheridan releas
>u
i;i^
i Jl t
i 'I g
'C a S"
111
"3 J§ -S
w 42 *S
1-8
-------
M
,Q
CtS
H
1-9
-------
Biological stressors, which include nonindigenous species that are introduced into an
area, intentionally or unintentionally, are unique because of their ability to reproduce, adapt, and
evolve (Section 4). The cases that focused on nonindigenous species introductions applied an
approach consistent with ecological risk assessment. One unique aspect was the incorporation of
some management considerations such as perceived impacts (social and political influences) into
the risk assessment model. Although uncertainty in predicting risks associated with biological
stressors can be very high, management decisions must be made, and it is important to convey
these uncertainties to decision makers. As noted in Section 4, the strength of using risk
assessment to evaluate nonindigenous species is that it provides a framework for taking the
available information and placing it hi a format that can be used and understood for making risk
management decisions.
Sections 6 to 9 describe potential uses of ecological risk assessment and related scientific
assessments. Applications include agricultural ecosystems, endangered/threatened species,
i
ecosystem management, and oil spills. Many of the cases reported in these sections are not
ecological risk assessments and use varying terminology and different approaches. However, as
discussed below, many apply portions of the ecological risk assessment process and could
potentially benefit from increased use of ecological risk assessment methods and approaches.
Ecosystem management is increasingly being used hi assessments involving multiple
stressors and multiple spatial and temporal scales. As described in Section 8.2.1.1, the Forest
Service and Bureau of Land Management have developed a four-step framework for ecosystem
management that includes monitoring, assessment, decision making, and implementation. While
! I
ecosystem management is not a form of ecological risk assessment, assessments done as a part of
ecosystem management frequently share several common elements with ecological risk
assessments. For example, both recognize the importance of preassessment planning, the need to
link data gathering to assessment issues, and the importance of involving interested parties
(stakeholders) in the process. The most direct attempt to incorporate ecological risk assessments
in ecosystem management is illustrated by EPA's five watershed case studies (Section 8.2.3).
Ecosystem management frequently express its goals using terms such as ecological
sustainabttity, integrity, or health. While these terms are useful as guiding principles, they must
be explicitly interpreted to support an assessment. Some key questions (U.S. EPA, 1998) that
need to be addressed include the following:
i
• What does sustainability or integrity or health mean for a particular system?
• What must be protected to meet these goals?
1-10
-------
• Which ecological resources and processes are to be sustained and why?
• How will we know when we have achieved the goals?
Some believe that ecological risk assessment has only limited applicability to ecosystem
management (Lackey, 1994). The perception is that ecological risk assessment is hampered by
the difficulty in defining what constitutes adverse ecological effects hi complex situations and
that applying ecological risk assessments requires inappropriate simplifications to allow
application of quantitative risk methods. In fact, ecological risk assessment is simply a tool for
capturing scientific information and uncertainties in a way that can support decision making.
Ecological risk assessment does not always use quantitative tools (e.g., see Section 4 on
nonindigenous species), nor does it have to simplify information more than any other approach to
a highly complex problem. Further, both ecosystem management and ecological risk assessment
recognize the need for initial planning and discussions between risk assessors and risk managers
(including stakeholders) to define management goals that reflect societal concerns and to
communicate assessment results and decisions with the stakeholders, including the public.
Difficulties in defining adverse ecological effects and resolving conflicting societal values are
common to ecosystem management in general, not to any one type of decision support approach.
As with ecosystem management, application of ecological risk assessment to agricultural
ecosystems has been varied hi scope and extent. The shrimp virus case study (Section 6.2.2)
illustrated a preliminary problem formulation relevant to on area of agricultural concern
(aquaculture) that closely follows the ecological risk assessment process. Applying the process
to the multiple stressors, multiple receptors, and larger geographic scales found hi the
Environmental Quality Incentives and Conservation Reserve Programs was more difficult. In
these cases, the analysis focused heavily on the problem formulation phase, and quantitative
analyses were not possible given the scope of the assessment and the limited resources available.
The concept of assessment endpoints was modified slightly to accommodate program needs.
Further adaption of the ecological risk assessment process to agricultural ecosystems is
suggested, as are establishing an iterative process between risk assessors and risk managers,
identifying when risk assessments are required, and clearly stating the risk management
objectives.
The section on endangered species focuses on the use of specific modeling tools for
estimating the risks of extinction of small populations. Some: of the parameters in the models
include characteristics that influence the probability of extinction (e.g., random demographic or
environmental changes, loss of adaptive variation, environmental catastrophes, accumulation of
1-11
-------
deleterious genetic factors, or habitat fragmentation). This information could contribute to an
ecological risk assessment that might also include considerations of exposure to anthropogenic
stressors (e.g., habitat loss or introduced species), explicit descriptions of assessment endpoints
and conceptual models, and a more complete risk characterization.
The last section of this report (accidental release of chemicals) is more typical of the
established uses of ecological risk assessment for chemical stressors. What is different is the
very short tune frame required for decision maldng following a chemical spill. The ecological
risk assessment process can be useful in helping to structure the problem-solving process and
involving stakeholders, but ecological risk principles need to be incorporated in advance into the
strategies that determine how an agency will respond to an accidental release. For example, area
contingency plans could be restructured following ecological risk assessment principles.
Together, the established and potential uses of ecological risk assessment and related
scientific assessments described in this report illustrate the broad usefulness of the ecological risk
assessment process. The inherent flexibility of the paradigm provides the means to address a
wide range of stressors, ecological systems, and biological, temporal, and spatial scales.
Nevertheless, much remains to be done to further incorporate ecological risk assessment into the
environmental decision-making process.
1.3. REFERENCES
Lackey, RL. (1994) Ecological risk assessment. Fisheries 19(9): 14-18.
National Research Council. (1993) A paradigm for ecological risk assessment. In: Issues in risk assessment.
Washington, DC: National Academy Press.
National Research Council. (1996) Understanding risk: informing decisions hi a democratic society. Washington,
DC: National Academy Press.
SETAC. (1997) Ecological risk assessment. Technical Issue Paper. Pensacola, FL: SETAC.
U.S. Environmental Protection Agency. (1998, May 14) Guidelines for ecological risk assessment. Federal Register
63(93):26846-26924.
1-12
-------
ESTABLISHED USES
-------
-------
2. ECOLOGICAL RISKS OF A NEW INDUSTRIAL CHEMICAL UNDER TSCA
2.1. SUMMARY
This chapter illustrates how useful ecological risk assessments can be conducted even
when resources are severely constrained. EPA's Office of Pollution Prevention and Toxics
(OPPT) conducts ecological risk assessments for new chemical substances regulated by the
Toxic Substances Control Act (TSCA). Under TSCA, manufacturers and importers of new
chemicals are required to submit a premanufacture notification (PMN) to EPA before they intend
to begin manufacturing or importing. OPPT has only 90 days to complete the risk assessment
and has very limited exposure and effects data. In addition, OPPT receives more than 2,000
PMN submissions every year, which limits the amount of resources available for each case.
For PMN evaluations, the ecological risk assessment process (problem formulation,
analysis, and risk characterization) is applied in a tiered fashion. The initial planning and
problem formulation stage is quite similar for most assessments, because the assessments are
usually not site specific and similar models and endpoints are used for different chemicals.
Assessment endpoints and measures of effect (measurement eftdpoints) are identified, and the
analysis and risk characterization phases are conducted sequentially using additional data and
fewer worst case assumptions with each successive tier. The overall approach is to compare
potential ecological effect concentrations that have been adjusted for uncertainty with potential
exposure concentrations. If a risk is ascertained, more detailed analyses are performed.
Because of the paucity of datei, there is a heavy reliance on the use of structure-activity
relationships (S ARs) to predict ecotoxic effects and exposure/fate characteristics (such as
physical/chemical properties and biodegradation), and uncertainty (assessment) factors are used
to compensate for a lack of definitive data when comparing effects concentrations with exposure
levels. Given the constraints on the assessments, it is not possible to quantify effects on the
assessment endpoint: populations and communities of aquatic organisms and aquatic
ecosystems. Nevertheless, the risk assessment approach provides a useful way of applying
scientific information to environmental decisionmaking.
The case study in this chapter focuses on the assessment of a PMN substance, i.e., an
alkylated diphenyl, that is a neutral organic compound. The PMN substance was likely to be
discharged into freshwater aquatic systems, so the assessment focused on aquatic organisms
(e.g., fish, aquatic invertebrates, and algae). Beginning with SAR toxicity predictions and simple
dilution models for exposure, the risk assessment proceeded through five iterations. During risk
characterization at the end of each iteration, a quotient method was used to compare exposure
concentrations with the ecological effect concentrations.. A ratio of 1 or greater indicated a
2-1
-------
potential risk. The first four iterations identified an ecological risk and resulted in the collection
of additional and more specific ecological effects test data and more detailed information on
potential exposures to the PMN substance. The final outcome was that the PMN substance could
be used only at specific sites because there was uncertainty as to whether the concern level (1
i i
ug/L) might be exceeded at sites not identified and characterized by the submitter.
OPPT risk assessment results are used as the basis for any of several risk management
options, including a variety of regulatory enforcement actions such as banning discharges to
water or requiring pretreatment. This case study illustrates how an efficient and pragmatic
ecological risk assessment process can assist in eliciting reasonable risk management decisions.
i
2.2. INTRODUCTION
The prospective evaluation (and risk assessment) of new industrial chemicals and the
i |
retrospective assessment of an inventory of existing chemicals are within the purview of EPA's
Office of Pollution Prevention and Toxics (OPPT, formerly the Office of Toxic Substances).
This office and its mission were established when the Toxic Substances Control Act (TSCA) was
passed in 1976 to regulate the chemicals in commerce that were not covered by other legislation;
that is, TSCA covers only industrial chemicals (e.g., solvents, polymers, adhesives, coatings,
plastics, pigments, detergents) and not pesticides, Pharmaceuticals, etc.
In 1979, almost 62,000 chemical substances were reported to be in commerce, and these
were "grandfathered" as the TSCA inventory of existing industrial chemicals. Chemicals not hi
this inventory were to be considered new industrial chemicals, and more than 32,000 of these
have been submitted by industry for assessment since July 1979. Via the inclusion of about
13,000 new industrial chemicals that have been assessed for risk and are now in commerce, the
TSCA inventory has now increased to more than 75,000 chemical substances with a total
production/import volume that was about 6 trillion Ibs/year (2.7 trillion kg/year) in 1989
(INFORM, 1995). However, the total made and/or imported into the United States in 1989 was
larger than this estimate; approximately 25,000 existing chemicals were not reported because
they did not reach the 10,000 Ibs/site/year reporting threshold or because they were inorganic
chemicals. In addition, from 1989 through 1995, the production of just the top 50 organic and
inorganic chemicals in the United States increased 33% and 15%, respectively (Zeeman, 1996).
2.2.1. EPA/OPPT Risk Assessment Approach
This overview is based on OPPT's ecological risk assessment of a new chemical
premanufacture notification (PMN). The PMN case study (U.S. EPA, 1994) originally was
prepared to illustrate the consistency between the OPPT ecological risk assessment approach and
2-2
-------
EPA's Framework for Ecological Risk Assessment (U.S. EPA, 1992) (Figure 2-1); that is, they
are both composed of three phases: problem formulation, analysis, and risk characterization.
This framework approach and several other such case studies were then used as a basis for the
development of, and are therefore consistent with, EPA's Guidelines for Ecological Risk
Assessment (U.S. EPA, 1998). Portions of the OPPT/TSCA PMN case study were even used to
illustrate certain features of ecological risk assessment in the guidelines report.
This OPPT approach to ecological risk assessment of new industrial chemicals has been
in place for more than a decade (Zeeman and Gilford, 1993). The specific example presented
here (Section 2.3) is abbreviated from the original PMN case study. The PMN case study also
was used as the basis for a more comprehensive publication of OPPT's methods for ecological
risk assessment in the peer-reviewed literature (Nabholz et al., 1998).
OPPT's overall approach to assessing the risks of new chemicals is to compare potential
ecological effect concentrations that have been adjusted for uncertainty (i.e., concern
concentrations) with potential exposure concentrations. The process often begins with predicting
toxicity, adjusting these effect concentrations for uncertainty, and contrasting one or more of
these concern concentrations with one or more predicted environmental concentrations from
simple stream flow dilution models that typically result in reasonable worst-case exposure
scenarios. If a risk is ascertained, more detailed analyses are performed (Figure 2-2). Because of
the paucity of data typically associated with new chemical PMN submissions (see discussion in
Section 2.2.2, Statutory and Regulatory Background), there is a heavy reliance on the use of
quantitative structure-activity relationships (QSARs) to predict ecotoxic effects to develop a
stressor-response profile or an ecotoxicity profile (Nabholz, 1;991; Zeeman et al., 1993,1995).
Figure 2-2 does not include all likely risk management options. In addition to obtaining
additional exposure and ecological effects information, risk management options can include a
variety of regulatory enforcement actions, such as requiring pretreatment or even banning
discharges to water. In any event, OPPT risk assessors must ascertain that a risk exists before
OPPT risk managers need to exercise these risk management options.
The approach taken in this PltfN evaluation has the following strengths: (1) it relates
measurement endpoints to an assessment endpoint; (2) it demonstrates that ecological risk
assessments can be conducted with minimal toxicity data and exposure data; (3) it demonstrates
the usefulness of SARs in establishing a toxicity or stressor-response profile; and (4) it
demonstrates that regulatory decisions can be made quickly Using only the best data available at
the time.
2-3
-------
PROBLEM FORMULATION
Stressors: Neutral organic compound.
Ecological Components: Aquatic life (fish, invertebrates, algae) in rivers,
streams, and lakes.
Endpoints: Assessment endpoint is protection of aquatic life from unreasonable
adverse effects due to exposure to industrial chemicals. Measurement
endpoints are effects on mortality, growth, development, and reproduction using
surrogate species.
ANALYSIS
Characterization
of Exposure
Concentrations of the PMN
substance in the water column were
estimated with a simple dilution
model and PDM3. EXAMS II was
used to estimate concentrations in
the water column and sediments.
Characterization of
Ecological Effects
QSAR and test data for algae,
fish, daphnids, and chironomids
were used to establish a stressor
response profile.
RISK CHARACTERIZATION
The Quotient Method of ecological risk assessment was used. To establish
ecological effect concentrations of concern, an uncertainly factor of 10 was
applied to the most sensitive measurement endpoint concentration.
Figure 2-1. Structure of assessment for effects of a PMN substance.
2-4
-------
Stepl. FOCUS MEETING
• Determine the most sensitive species and endpoint using actual
test data or QSAR. Estimate a chronic value whenever possible.
• Apply an Uncertainty Factor to obtain a concern concentration (CC).
• Calculate a Predicted Environmental Concentration (PEC) using a
simple stream flow dilution model as a worst case scenario for
concentrations in the water column.
Step 2. STANDARD REVIEW
• Obtain more information about Production, Use, and Disposal of the
PMN substance.
• Obtain additional ecotoxicological data (testing, analogs, QSAR).
• Estimate a chronic value (ChV) for the most sensitive species.
• Adjust the ChV with a margin of exposure (typically 10) to obtain a
new CC. i
• Use additional release data and the Probabilistic Dilution Model
(PDM3) to estimate the number of days in one year that the CC is
exceeded. Further analyses could employ EXAMS II.
Additional
ecotoxicity
or fate
tests
Is the CC exceeded
more than 20 times
in one
Step 3. RISK MANAGEMENT OPTIONS
• Control releases of the PMN substance pending additional testing.
• Ban manufacture or use under Section 5f of TSCA.
Figure 2-2. Flow chart and decision criteria for the ecological risk assessment of a PMN
substance.
2-5
-------
2.2.2. Statutory and Regulatory Background
TSCA provides for the regulation of chemicals not covered by other statutes (e.g., Food,
Drug, and Cosmetic Act; Federal Insecticide, Fungicide, and Rodenticide Act). TSCA requires
the assessment and, if necessary, regulation of all phases of the life cycle of industrial chemicals:
manufacturing., processing, use, and disposal.
TSCA regulates two categories of industrial chemicals: (1) existing chemicals in
commerce on the TSCA Chemical Substances Inventory, and (2) new chemicals that are not on
this inventory. The inventory includes both chemicals in commercial production between 1975
and 1979 and the chemicals reviewed under the new chemical PMN program that went into
commercial production after 1979. Section 5 of TSCA requires manufacturers and importers of
new chemicals to submit a PMN to EPA before they intend to begin manufacturing or importing.
EPA has up to 90 days to evaluate whether the substance will present an unreasonable risk of
injury to human health or the environment. With good cause, EPA can allow an extension for
another 90 days for the evaluation of the chemical.
In addition to the short review time allowed, three major difficulties are associated with
evaluating PMNs. The first is the confidential business information (CBI) protection afforded by
TSCA. Under this protection, manufacturers and importers can designate as CBI many
characteristics of the PMN substance, such as chemical name, structure, intended uses, and sites
of manufacture and use. This information is not available to the public, and only personnel with
TSCA CBI security clearance and members of Congress can access the information. There are
strict safeguards against disclosure of the CBI. The second difficulty is that, on average,
manufacturers and importers currently submit more than 2,200 new chemical notices to EPA
annually (Zeeman et al., 1995; Zeeman, 1997). The third difficulty is that only the following
information must be submitted: chemical identity; molecular structure; trade name; production
volume, use, and amount for each use; by-products and impurities; human exposure estimates;
disposal methods; and any test data that the submitter may have. The manufacturer does not
have to initiate any ecological or human health testing before submitting a PMN. Only about 5%
of the PMNs reviewed to date contain ecological effects data, and most of those data consist of
acute toxicity tests performed on fish (Nabholz, 1991; Nabholz et al., 1993a; Zeeman, 1995;
Zeeman etal, 1993,1995).
2.3. CASE STUDY: DESCRIPTION OF A NEW CHEMICAL ASSESSMENT UNDER
TSCA
This case study describes how OPPT evaluates the ecological risks of a PMN substance.
The risk assessment begins with a reasonable worst-case analysis using a stream flow dilution
2-6
-------
model to estimate environmental concentrations. This is the typical approach taken by OPPT,
and it results in conservative estimates of aquatic exposures. OPPT risk assessors initially use
any measured ecotoxicity data and SARs available to evaluate effect concentrations and dose-
response curves for a new substance. The assessors then adjust these effect concentrations with
assessment factors (or uncertainty factors) to set concern concentrations in the environment for
the chemical. The quotient method is used to integrate these exposures and effects into a
quantitative estimate of risk. That level of assessment typically suffices to show little or no risk
for the majority of new chemicals assessed by OPPT.
Because the initial worst-case assessment identified a risk potential for this specific PMN
chemical, additional iterative analyses were performed using actual test data and a more refined
exposure analysis using a probabilistic dilution model (PDM3). The second risk characterization
indicated risks to pelagic and benthic aquatic life; therefore, OPPT risk assessors used the
exposure analysis modeling system (EXAMS) II model and generic site data to predict
concentrations hi both the water column and sediments. OPPT assessors estimated toxicity to
benthic organisms using the chronic test data for fish and daphnids and assumed that the
sediments would decrease toxicity through adsorption of the chemical to the organic matter in
sediments. The results of these analyses still identified a potential risk.
The submitter then supplied OPPT with more precise data on the use and disposal of the
PMN substance, that is, a list of specific use sites. OPPT assessors input data for each of these
sites into EXAMS II and the results indicated little potential risk to benthic organisms at many of
the identified sites. OPPT was ready to issue a consent order to restrict use of the PMN
substance to the identified sites that posed low risk to the aquatic environment; however, the
submitter chose to perform OPPT's recommended test with contaminated/spiked sediments using
chironomids as the surrogate species for benthic organisms. The results of the test indicated
moderate toxicity and little potential risk to benthic organisms at the identified sites after 1 year's
release. The final outcome was that EPA restricted the use of the PMN substance to the
identified sites because there was uncertainty as to whether the concern level of 1 ug/L might be
exceeded at sites not identified and characterized by the submitter.
2.3.1. Background Information and Objective
OPPT performs the analyses listed below in assessing the human and ecological risks of
PMN substances. For a more detailed discussion of the process, see U.S. EPA 1986, Auer et al.,
1990; Moss et al., 1996; Nabholz, 1991; Nabholz et al., 1993a; and Wagner et al., 1995.
2-7
-------
2.3.1.1. Chemistry Report
The Industrial Chemistry Branch of OPPT's Economics, Exposure, and Technology
Division (EETD) evaluates PMNs to ensure that: (1) the chemical name matches the structure,
(2) the chemical/physical properties are accurate, (3) the information about the manufacture and
processing is accurate, and (4) the uses are consistent with the chemical.
i i
2.3.1.2. Engineering Report
The Chemical Engineering Branch of EETD estimates worker exposure during the life
cycle of the chemical (manufacturing, processing, use, and disposal) and estimates releases of the
chemical to the environment. The sites of release can be generic or specific using standard
industrial codes.
2.3.1.3. Exposure Assessment
The Exposure Assessment Branch of EETD evaluates (1) available fate, transport,
abiotic, and biotic fate parameters, and (2) consumer exposure. This is analogous to the exposure
profile discussed in EPA's framework report (U.S. EPA, 1992) and the guidelines report (U.S.
EPA, 1998). The exposure assessment estimates the environmental concentrations likely to
occur during the life cycle of the PMN substance. This includes an evaluation of potential
exposure from releases to surface waters, landfills, and land spray, as well as nonoccupational
(consumer) exposures. Environmental concentrations can be generic or site specific. PMN
i I I
substances frequently are discharged to water; therefore, more than 80% of PMN exposure
assessments address aquatic environments, chiefly rivers arid streams.
2.3.1.4. Ecological Hazard Assessment
Also known as a toxicity profile in OPPT, the ecological hazard assessment is analogous
to the stressor-response profile discussed in the framework report (U.S. EPA, 1992) and
guidelines report (U.S. EPA, 1998); this assessment was performed by the Environmental Effects
Branch of the Health and Environmental Review Division (now the Risk Assessment Division).
The initial ecological hazard assessment predicts and evaluates the potential adverse ecological
effects of a PMN substance and relies primarily on SARs. For many classes of discrete organic
chemicals reviewed by OPPT (about 50% of which are neutral organic chemicals), SARs are
available that permit a prediction of acute and chronic toxicity to surrogate species, such as fish,
aquatic invertebrates, and algae (Clements, 1988,1994; Auer et al., 1990; Nabholz, 1991;
Nabholz et al., 1993a, 1993b; Zeeman et al., 1993, 1995; Clements and Nabholz, 1994; Zeeman,
2-8
-------
1995). The Risk Assessment Division also reviews the results of submitted test data and, if valid
and adequate for risk assessment, incoirporates them into the ecological hazard assessment.
2.3.1.5. Ecological Risk Assessment
In practice, staff of the Risk Assessment Division develop the ecological risk assessment
for new chemicals and support them. Ecological risk assessments are conducted in a tiered
fashion (see Figure 2-2). Initial hazard and exposure assessments are evaluated at the first risk
assessment meeting, that is, FOCUS meeting, to ascertain if there are any potential risks. If risks
are not identified at the FOCUS meeting, the chemical is typically dropped from further review.
If a risk is identified, which happens about 5% of the time, the'PMN substance undergoes a more
detailed assessment, called a standard review (Wagner et al., 1995; Moss et al., 1996).
Alternatively, additional information may be requested from the manufacturer or importer
immediately following the FOCUS meeting. If a risk is still identified after all additional
information has been submitted, then risk management options are considered. Possible risk
management options include but are not limited to (1) control options (such as no releases to
water) pending further tests of the PMN substance, (2) issuance of a TSCA significant new use
rule, and (3) direct control under Section 5f (e.g., banning the manufacture or use of the PMN
substance).
2.3.2. Problem Formulation
2.3.2.1. Stressor Characteristics
Table 2-1, which appears in Section 2.3.3.1.1, lists the ,physical/chemical properties of the
subject PMN substance. The manufacturer declared the chemical identity, structure, intended
uses, and sites of use as CBI. This particular example evaluated only the parent compound
because OPPT risk assessors did not expect the PMN substance to readily degrade or be
transformed into more toxic metabolites.
2.3.2.2. Ecosystem Potentially at Risk
The processing, use, and disposal sites were adjacent to rivers and streams. OPPT
assessors expected the PMN substance to be discharged to such rivers and streams. Thus,
pelagic and benthic aquatic populations and communities were determined to be potentially at
risk.
2-9
-------
2.3.2.3. Ecological Effects
\ i
The PMN substance was an alkylated diphenyl, and it belongs to a class of chemicals
known as neutral organic compounds. These chemicals are nonelectrolyte zmd nonreactive and
exert toxicity through a narcotic or nonspecific mode of action (Lipnick, 1985; Auer et al., 1990;
Veith and Broderius, 1990). Neutral organic compounds can exert both acute and chronic
effects. The toxicity of neutral organic compounds has been correlated with molecular weight
and the logarithm of the KOW. Experimental data have shown that neutral organics with a log KQW
of 5.0 or more do not exert pronounced acute effects (toxic effects such as mortality or
immobilization within 2 to 4 days). This is due mainly to the low water solubility of such
compounds, which results in decreased bioavailability to aquatic organisms. Because of this
decreased bioavailability, exposure durations of 4 days or fewer are typically insufficient to elicit
marked acute effects (e.g., as measured by a 96~h LC50 test). Because of the high K,,w of this
PMN substance, OPPT risk assessors expected only chronic effects to be able to occur at or
below the chemical's aqueous solubility limit.
OPPT typically assesses ecological effects for three trophic levels of food webs: primary
producers (algae), primary consumers (aquatic invertebrates), and forage/predator fish. OPPT
assessors use the most sensitive species and toxicological effect for the initial risk assessment.
Unless only chronic effects are expected, such as for the PMN substance in this study, OPPT
usually assesses both acute and chronic effects. The ecological effects characterization is based
on effects on mortality, growth and development, and reproduction. The SARs used for neutral
organic chemicals are:
Fish acute toxicity (Veith et al., 1983)
Daphnid acute toxicity (Hermens et al., 1984)
Green algal toxicity (Clements, 1988,1994)
Fish chronic value (Broderius and Russom, 1989)
Daphnid chronic value (Hermens et al., 1984)
Green algal chronic value (Clements, 1988,1994).
The rationale and approach used to assess these effects are presented under measurement
endpoints (Section 2.3.2.5).
2.3.2.4. Assessment Endpoints
TSCA was intended to prevent unreasonable risks to health and the environment as a
result of the manufacture, processing, use, and disposal of industrial chemicals. The assessment
endpoint (Suter, 1990) used in this study was the protection of aquatic organisms (algae, aquatic
2-10
-------
invertebrates, and fish). OPPT assessors assumed that any effects from the PMN substance
would be exhibited at least up to the population level of organization.
2.3.2.5. Measurement Endpoints
OPPT assessors used the following measurement endpoints (Suter, 1990) to assess the
risks to the assessment endpoint:
• Mortality
• Growth and development
• Reproduction.
Clements (1983) and EPA (1983) present the rationale,for selecting these endpoints. In
summary, documented evidence indicates that xenobiotics cari adversely affect these endpoints
both directly and indirectly. Since populations are governed by mortality, growth and
development, and reproduction, OPPT assessors presume that adverse effects to these
measurement endpoints would manifest themselves at least up to the population level of
ecological organization. Thus, there is a logical connection between the assessment endpoint
(i.e., the protection of aquatic life, at least up to the population level) and the measurement
endpoints.
OPPT uses a tiered approach when testing the toxicity of a given industrial chemical
(U.S. EPA, 1983; Smrchek et al., 1993; Zeeman and Gilford, 1993). The first tier consists of
relatively inexpensive short-term tests that measure acute effects chiefly on the three trophic
levels discussed in Section 2.3.2.3, Ecological Effects, that is, mortality to fish and aquatic
invertebrates and population growth for green algae. The first tier or "base set" for aquatic
toxicity consists of a 96-h fish acute test, a 48-h daphnid test, land a 96-h algal test. Because the
algal test represents exposure across about eight generations of algal cells, OPPT considers the
algal test to be representative of chrome toxicity to algal populations.
Additional tiers consist of chronic tests, such as the fish early life-stage toxicity test,
which measures effects on mortality and growth and development, and the daphnid chronic test,
which measures effects on survival and reproduction. OPPT assessors typically must ascertain a
potential risk before proceeding to request any acute or chronic toxicity testing. For high-KoW
chemicals, such as the subject PMN, OPPT assessors usually expect little or no acute toxicity to
be seen from such short-duration tests, and most everyone agrees to save time and money by
going directly to the tier of chronic toxicity testing.
2-11
-------
2.3.2.6. Conceptual Model
On the basis of experience with neutral organic compounds and available SARs, it was
clear that the high K^ for the PMN substance indicated a risk of chronic toxicity only to pelagic
and benthic aquatic organisms. Principal concerns were for effects on mortality, growth and
development, and reproduction. OPPT assessors presumed that these effects would be
i
manifested at least up to the population level of organization (Clements, 1983).
A preliminary exposure profile was developed through the use of simple stream flow
models. To characterize ecological effects, SARs (Clements, 1988,1994; Clements and
Nabholz, 1994) were used to develop an initial toxicity profile or stressor-response profile (see
Table 2-3 in Section 2.3.3.2). There was low concern for acute or short-term exposures but high
concern for chronic or long-term exposures.
The SARs, which were developed from actual testing of neutral organic compounds using
surrogate species (U.S. EPA, 1982) that represented aquatic organisms in rivers and streams,
predicted that fish would be the most sensitive group of aquatic species, with a predicted chronic
value (ChV) of 0.002 mg/L (2 ug/L= 2 ppb). However, aquatic invertebrates (i.e., daphnids) also
were predicted to be sensitive, with a ChV of 0.004 mg/L (4 |J.g/L= 4 ppb).
Assessment factors (U.S. EPA, 1984; Nabholz, 1991; Nabholz et al., 1993a; Zeeman and
Gilford, 1993; Zeeman, 1995) were used to address uncertainties in extrapolating from laboratory
to field effects. Investigators used a quotient method of ecological risk characterization to assess
risk (Bamthouse et al., 1986; Nabholz, 1991; Rodier and Mauriello, 1993). If the results of the
; |
risk characterization predicted an unreasonable risk, OPPT assessors planned to perform a more
in-depth analysis, including fate and transport modeling and ecological effects testing in
accordance with ecological effects test guidelines (U.S. EPA, 1985). The PDM3 and EXAMS II
exposure models would further characterize and refine exposure, and additional ecological
effects testing of the PMN substance would be based on the criteria established by OPPT (U.S.
EPA, 1983). Assessors would continue to use the quotient method to characterize risks.
2.3.3. Analysis, Risk Characterization, and Risk Management—First Iteration
2.3.3.1. Characterization of Exposure
Because the use of the PMN substance is claimed as CBI, only the terms manufacturing,
' !
processing, use, and disposal are used to describe the life cycle of the alkylated diphenyl. The
sites of manufacture, use, and disposal and the actual releases (i.e., kg/day) that were used to
calculate concentrations of the PMN substance in receiving rivers and streams also are
considered as CBI. The production volume was estimated at more than 100,000 kg/year.
2-12
-------
2.3.3.1.1. Stressor characterization. The alkylated diphenyl has low water solubility (<1 ppm)
and is not expected to volatilize from water because of the low vapor pressure (Table 2-1).
Photodegradation is negligible, and the compound is expected to sorb strongly to sediments. The
half-life for aerobic degradation could be weeks; anaerobic degradation could require months or
longer.
2.3.3.1.2. Exposure analysis. In the first iteration, OPPT assessors used a simple stream flow
dilution model to calculate predicted environmental concentrations (PECs). The calculation was
based on the following algorithm:
Concentration = releases (kg/day)/stream flow (millions of L/day)
The PEC calculations use both stream mean and low flow rates. In addition, the initial OPPT
exposure analysis typically ranks stream flow rates and uses the 10% and 50% flow rates. The
measured solubility limit of 0.300 mg/L was used. '
OPPT assessors determined that there would be no significant releases during the
manufacture of this PMN substance. The most significant routes of exposure would result from
the use and disposal of the chemical. Effluents containing the PMN substance would first be
treated in publicly owned treatment works (POTWs), which are wastewater treatment plants that
Table 2-1. Physical/chemicsil properties of PMN substance
Property
Measured or estimated value
Chemical class
Chemical name
Generic name
Chemical structure
Physical state
Molecular weight
Log Kow
LogKoc
Water solubility
Vapor pressure
Neutral organic
CBI
Alkylated diphenyls
CBI
Liquid :
232
6.7a
6.6"
0.051 mg/L (estimated)0
0.300 mg/L (measured)
O.001 Torr @ 20 °Cd
"Estimated using CLOGP program (Leo and Weininger, 1985).
•"Estimated by a regression equation developed by Karickhoff et al. (1979). The average method
error for the log KOC was 0.2 log KOC units over a log K,,,. range of 2 to 6.6;
"Estimated by a regression equation developed by Banerjee et al. (1980). ••
""Estimated by a regression equation cited in Grain (1982).
2-13
-------
include primary and biological treatment of the incoming waste stream. POTWs normally are
located off-site or between the processing plant and the receiving river. To assess the extent of
removal of the PMN substance by POTWs, OPPT assessors used data from laboratory-scale
wastewater treatment experiments and their output from mathematical wastewater treatment
1 i
simulations. The results indicated that removal would be due largely to adsorption to sludge;
however, the analysis assumed approximately 10% of the PMN substance released from
treatment was sorbed to solids in the effluent. This assumption was based on typical solids
removal for secondary wastewater treatment systems.
This study did not consider the fate and ecological effects of the PMN substance in sludges.
2.3.3.1.3. Exposure profile. Table 2-2 lists the PECs estimated at mean and low stream flows
for the PMN substance during manufacture, use, and disposal.
2.3.3.2. Characterization of Ecological Effects—Stressor-Response Profile
OPPT assessors initially used SARs to estimate the ecological effects of the PMN
substance as the result of a prenotice communication from the submitter. The potential submitter
contacted EPA before submitting the PMN and was informed about OPPT's concerns for chronic
toxicity. As a result, the submitter conducted and included the results of a fish acute toxicity test
and a fish early life stage toxicity test of this alkylated diphenyl in its PMN submission. Table
2-3 summarizes the SAR-derived effect concentrations and the results of the fish acute and fish
early life-stage toxicity tests.
2.33.3. Risk Characterization
Five risk characterizations were performed in this case study. Table 2-4 provides a brief
summary of the assumptions, estimations, and types of uncertainty for each of the five iterations.
Table 2-2. Predicted environmental concentrations (PECs) for
PMN substance Qig/L or ppb)
Process
Manufacture
Use
Disposal
Mean flow
10%a
0.0
9.0
52.0
50%
0.0
0.5
0.7
Low flow
10%
0.0
68.0
90.0
50%
0.0
4.0
6.1
"Percent of streams having flows equal to or less than the value used to calculate the PECs.
2-14
-------
Table 2-3. PMN substance initial stressor-response profile
SAR estimated toxicity8
Endpoint
Effect concentration
Reference
Fish 96-h LC50
Daphnid 48-h LC50
Green algae 96-h EC50b
Fish ChV°
Daphnid ChV
Algal ChV
No effect at saturation
No effect at saturation
No effect at saturation
0.002 mg/L
0.004 mg/L
No effect at saturation
Actual measured toxicity submitted with PMN
Veith et al. (1983)
Hermensetal. (1984)
Clements (1988, 1994)
Broderius and Russom (1989)
Hermens et al. (1984)
Clements (1988, 1994)
Fathead minnow
(Pimephales promelas)
96-h acute test
P. promelas early life-stage
test 31-day ChV (growth,
mean wet weight)
P. promelas early life-stage
test 31-day ChV (survival,
growth flengthl)
No effect at saturation
0.013 mg/L
0.061 mg/L
U.S. EPA (1993)
U.S. EPA (1993)
U.S. EPA (1993)
"Based on molecular weight an
"Median effect concentration.
°The ChV is the geometric mean of the highest treatment concentration for which no statistically
significant effects were observed and lowest treatment concentration for which
statistically significant toxic effects were observed. The ChV is the geometric mean
of the maximum acceptable toxicant concentration and is also known as the chronic
no-effect-concentration.
Table 2-4. Summary of five risk characterization iterations
Iteration Estimates/assumptions
Uncertainty
Fish are the most sensitive species. CC set at 1 (ig/L.
PMN substance mixes instantaneously in water. No
losses.
Actual test data for daphrnids still indicate a CC of 1,
Hg/L. Determine how often this concentration is
exceeded using PDM3. ;
Estimate risk to benthic organisms using daphnid \
ChV and mitigation by organic matter. EXAMS II :
used to estimate concentrations.
Site-specific data obtained on use and disposal.
EXAMS II rerun with new data.
Actual test data for benthic organisms obtained.
Worst-case analysis.
Worst-case analysis. Other
species may be more sensitive.
Generic production sites.
Actual data for benthic
organisms not available.
Estimated toxicity for benthic
invertebrates.
Best estimates for identified
sites. May not hold for other
sites or uses.
2-15
-------
2.3.3.3.1. Risk estimation: integration and uncertainty analysis via the use of assessment
\ I ' ' I
factors. OPPT assessors use the quotient method to estimate ecological risks. A quotient of 1 or
greater indicates a risk. The algorithm is given below:
Risk quotient = PEC/CC
! i
OPPT calculates the concern level or concern concentration (CC) by identifying the most
sensitive species and effect from the stressor-response profile and dividing by an appropriate
assessment factor (U.S. EPA, 1984). These assessment, factors are akin to uncertainty factors but
originally were designed to provide a risk-based rationale for requesting information and testing.
The assessment factors developed and used by OPPT (see Table 2-5) are as follows: (1) 1,000 if
only one acute value is available; (2) 100 applied to the most sensitive species when the
environmental base set of toxicity data (i.e., fish acute toxicity, daphnid acute toxicity, and green
algal toxicity) are available; (3) 10 applied to the lowest ChV (see Table 2-3, footnote c) for fish,
daphnids, and algae; and (4) 1 applied to the ChV from a field study (e.g., pond) or from a
i ' i
microcosm study. Note that these assessment factors are designed to decrease in magnitude as
more definitive toxicity data are made available to adequately assess the hazard profile of a new
chemical.
In this case, OPPT assessors used the measured ChV of 0.013 mg/L from the fathead
minnow early life-stage test rather than the estimated ChV of 0.004 mg/L for the daphnids that
was based on a SAR (Table 2-3). To account for the uncertainty between chronic toxicity noted
Table 2-5. OPPT assessment factors rased in setting "concern levels" for
new chemicals
Available data on chemical
or analogue
Assessment factor
Limited (e.g., only one acute LC50
viaSAR/QSAR)
Base set acute toxicity (e.g., fish
and daphnid LC50s and algal EC50)
Chronic toxicity MATCsa
Field test data for chemical
1,000
100
10
1
"MATC - maximum acceptable toxicant concentration.
Source: EPA (1984); Nabholz (1991); and Zeeman and Gilford (1993).
2-16
-------
in the laboratory and those that might occur in the field, an assessment (uncertainty) factor of 10
was used. The ChV was divided by this assessment factor to yield a CC of 0.0013 mg/L, which
was rounded off to 0.001 mg/L or 1 ug/L (ppb). ;
In estimating risk, the CC of 1 |ig/L was compared with the PECs (Table 2-2). As can be
seen, the CC was exceeded at both the low and mean flow rates for 10% of the streams and at
low flow for 50% of the streams. A chronic risk for the use and disposal of the chemical was
inferred based on these stream flows.
It should be noted that the initial risk assessment evaluates risks to aquatic species in the
water column only.
2.3.3.4. Risk Management \
Because the results of the initial risk characterization identified a potential unreasonable
risk, OPPT assessors recommended requesting a chronic daphnid test to complete the chronic tier
tests. EPA also informed the submitter that a benthic test with contaminated sediments could be
required if there was a potential unreasonable risk to sediment-dwelling organisms. The concern
for benthic organisms was based on the high K^,, low vapor pressure, and low water solubility,
which indicate that this alkylated diphenyl was likely to partition to the sediments of rivers and
streams, resulting in exposures of benthic organisms. EPA also requested a test that simulates
the effectiveness of a POTW in removing the PMN substance from the waste stream.
2.3.4. Analysis, Risk Characterization, and Risk Management—Second Iteration
2.3.4.1. Characterization of Exposure
The coupled units test is a measure of the POTW removal of the PMN substance under
conditions that simulate treatment in activated sludge. The POTW simulation conducted by the
manufacturer indicated that a POTW would remove from 95% to 99% of the PMN substance.
2.3.4.2. Characterization of Ecological Effects
A 21-day daphnid chronic toxicity test of the chemical was conducted and was found to
be valid and adequate for risk assessment purposes. The daphnid ChV for survival, growth, and
reproduction was determined to be 0.007 mg/L (ppm) or 7.0 jj.g/L (ppb). This was found to be in
excellent agreement with the SAR determined ChV of 4 ppb.
2.3.4.3. Risk Characterization
OPPT assessors then used these new data in a probabilistic dilution model (PDM3) (U.S.
EPA, 1988) to estimate the number of days out of 1 year that the CC will be exceeded. OPPT
2-17 :
-------
assessors continued to use the CC of 1 |ig/L or 1 ppb, since the daphnid ChV of 0.007 mg/L
divided by the assessment factor of 10 also rounds off to 0.001 mg/L (ppm) or 1 ug/L (ppb).
Like the simple stream flow model, PDM3 assumes that 1 day's release of the chemical will mix
instantaneously with 1 day's flow of stream water in the receiving stream reach, and no losses
will occur through any physical, chemical, or biological transformations after release. Stream
flow rates for the proposed sites of use and disposal of this chemical were obtained from the U.S.
1 I
Geological Survey stream reach database. Table 2-6 presents the results of PDM3.
2.3.4.3.1. Interpretation of ecological significance. As a matter of policy, OPPT infers a
potential unreasonable risk to aquatic organisms if a CC based on chronic effects exceeds 20
days or more. The greater the number of days the CC is exceeded, the greater the potential risk.
The 20-day criterion is derived from partial life-cycle tests (daphnid chronic and fish early life-
stage tests) that typically range from 21 to 28 days in duration. OPPT infers low potential risk or
no unreasonable risk if the CC is exceeded on fewer than 20 days. It is important to remember
that the PDM3 model estimates only the total number of days out of 1 year that the CC is
exceeded. The days are not necessarily consecutive, and thus the 20-day criterion is a
conservative one. However, in practice, many low-flow days occur together during the same
season for many stream reaches in the United States. The second iteration continued to show an
unreasonable risk to aquatic organisms from the PMN substance because the CC of 1 ppb was
exceeded on 20 days for use and 39 days for disposal (Table 2-6)
'
2.3.4.4. Risk Management
EPA notified the submitter that a potentially unreasonable risk to aquatic organisms still
existed. A meeting was held to discuss possible benthic toxicity tests and to clarify unanswered
questions regarding releases of the PMN substance through use and disposal. It also was decided
to evaluate exposure further through the use of EXAMS II (Burns, 1989).
i i
Table 2-6. PDM3 analysis8
Process
Exceedance (days/year)
Manufacture
Use
Disposal
0
20
39
•Releases to water in actual kg/d considered CBI. PMN substance was expected to be released 350 days/year, and a
95% removal by POTW was assumed.
2-18
-------
2.3.5. Analysis, Risk Characterization, and Risk Management—Third Iteration
2.3.5.1. Characterization of Exposure
A preliminary EXAMS II analysis at the site expected to be at greatest risk indicated
sediment concentrations ranging from 11.0 to 22.0 mg/kg (ppm) dry weight sediment after 1 year
of releases of the PMN substance.
2.3.5.2. Characterization of Ecological Effects
Currently, there are no SARs for aquatic benthic organisms; however, SARs do exist for
neutral organics with earthworms hi artificial soil. To estimate the ecological effects of the PMN
substance to aquatic benthic organisms, predictions from, a fish 14-day LC50 SAR (Konemann,
1981) were compared with the earthworm 14-day LC50 SAR developed by OPPT assessors. The
earthworm 14-day LC50 was about 10' tunes higher than the fish 14-day LCSO for the alkylated
diphenyl. OPPT assessors concluded, that the organic matter (i.e., ground peat) in the artificial
soil could mitigate the toxicity of neutral organic chemicals by about 10 times.
OPPT assessors similarly expected that the organic matter found in natural sediments
would mitigate the toxicity of the PMN substance by about another factor of 10, because natural
organic matter in natural sediments should be more efficient at binding neutral organic chemicals
than freshly ground peat in artificial soil. That is, sediment organic matter is likely to have a
larger surface area-to-volume ratio than ground peat and, therefore, have more sites to bind
hydrophobic compounds. Proceeding on the above assumption, the effective concentrations in
the chronic toxicity profile for fish and daphnids were multiplied by 20 to produce the stressor-
response profile for benthic organisms (Table 2-7). This scenario used the best data available at
the time for neutral organic compounds, and the PMN submitter accepted the rationale for
mitigation because it had no better data.
2.3.5.3. Risk Characterization: Risk Estimation and Uncertainty Analysis
The most sensitive endpoint was the proposed invertebrate 21-day ChV of 0.1 mg/kg. An
assessment factor of 10 was applied to derive a benthic CC of 0.010 mg/kg (ppm) or 10 ug/kg
(ppb). The quotient method was used. As can be seen from the above preliminary EXAMS II
analysis, the exposure concentrations that were predicted exceeded the CC by factors of 1,000 to
2,000, and a risk to benthic organisms was inferred.
2-19
-------
Table 2-7. Predicted stressor-response profile for benthic organisms
Organism
Endpoint
Effect level
(mg/kg dry weight)
Invertebrate
Invertebrate
Vertebrate
14-day LC50
21-day ChV
31-dayChV
0.300
0.100
0.300 to 1.0
2.3.5.4. Risk Management
As a result of these assessments of exposure and risk, the submitter initiated an extensive
site-specific evaluation of the releases of the PMN substance during uses and disposal and
forwarded this new exposure information to OPPT for evaluation. The report is CBI.
I !
' " I '
2.3.6. Analysis, Risk Characterization, and Risk Management—Fourth Iteration
23.6.1. Characterization of Exposure
OPPT used the additional information that was submitted to conduct a more
comprehensive EXAMS II analysis. Table 2-8 summarizes the results for three specific use and
disposal sites.
2.3.6.2. Risk Characterization
\ \
As can be seen from the information in Table 2-8, the sediment concentrations predicted
from this refined exposure assessment were several orders of magnitude less than the preliminary
EXAMS II analysis. As a result, there was not enough of a risk to benthic organisms to warrant
a ban pending a testing decision by OPPT.
Table 2-8. EXAMS O analysis
Site
1
2
3
Water column (ug/L)
0.004
0.001
0.008
Sediments (mg/kg)
0.019
0.014
0.038
2-20
-------
2.3.6.3. Risk Management
A decision was made by OPPT Division Directors to offer the submitter a consent order
to allow manufacturing but require a benthic/sediment toxicity test to confirm the toxicity profile
and thus the risk assessment. Before offering the consent order, the submitter volunteered to test
with a benthic organism using clean natural sediment contaminated with known amounts of the
PMN alkylated diphenyl. The submitter and OPPT agreed to a 28-day chironomid toxicity test
using Chironomus tentans.
2.3.7. Analysis, Risk Characterization, and Risk Management—Fifth Iteration
2.3.7.1. Characterization of Ecological Effects .
Table 2-9 presents the results of the chironomid toxicity test.
2.3.7.2. Risk Characterization—Risk Estimation
Using an assessment factor of 10, a CC of 2.0 mg/kg dry weight sediment was set for the
benthic community based on the most sensitive effect, a ChVof 23 mg/kg for survival and
emergence of chironomids. The sediment CC was 50 times higher than the highest PEC for
sediments, and the ChV was an order of magnitude higher. Thus, there did not appear to be an
unreasonable risk to benthic organisms as a result of the use and disposal of the PMN substance
over a 1-year period.
As can be seen from Table 2-8, concentrations of the PMN substance at the specific sites
of use and disposal were estimated to be two to three orders of magnitude lower than the CC of 1
|ag/L that had been set for water column organisms.
Table 2-9. Stressor-response profile for Chironomus tentans
Endpoint
Effect concentration
(mg/kg dry weight sediment)
14-day ChV
21-day EC50 emergence
25-day EC50 emergence
28-day EC50 emergence
28-day LC50 survival
ChV survival
ChV emergence
32
23
25
24
22
23
23
2-21
-------
2.3.7.2.1. Uncertainty and assessment factors. In this study, the three main types of uncertainty
with regard to ecological effects are variations in species-to-species sensitivity, uncertainty
regarding acute versus chronic effects, and uncertainty regarding extrapolating laboratory-
observed effects to those that might occur in the natural environment (Table 2-5). EPA (1984)
developed these assessment factors specifically for establishing concern levels or concentrations
for PMN substances. Their use was not intended to establish a "safe" level for a particular
substance, but rather to identify a concentration that, if equaled or exceeded, could result in some
: I
adverse ecological effects. As can be seen above, such a finding provides the rationale for
requesting either actual testing of the PMN substance and/or more specific information about fate
and exposure. Naturally, there are other types of uncertainty, such as the effects of the PMN
substance on adult rather than juvenile fish. Such types of uncertainty are considered research
issues.
In the case of the exposure profile, an important aspect of uncertainty has to do with the
actual duration of exposure. The PDM3 model predicts only the number of days out of one year
j
the CC will be exceeded (Table 2-4). These days are not necessarily consecutive days. Thus,
only flow rates could be used to account for seasonal variation. The presence or absence of
critical life stages of aquatic organisms cannot be accounted for with this type of analysis. In
addition, the generic nature of the assessment precludes identification of specific biota.
I
2.3.7.2.2. Risk description—ecological risk summary. This study demonstrates the utility of
SARs hi establishing toxicity profiles for aquatic organisms (fish, invertebrates, and algae). In
this case, the chemical structures of the alkylated diphenyl indicated that the PMN substance was
analogous to chemicals known to behave like neutral organic compounds. The high K,,w
indicated that the substance would not be acutely toxic, and this was confirmed by an actual test
with a surrogate fish species. Actual chronic toxicity testing in fish and daphnids confirmed the
SAR-predicted chronic toxicity (well within an order of magnitude). EPA experience with other
high KQW compounds such as hexachlorobenzene and chloroparaffins further confirms the
: I
chronically toxic nature of such compounds. The predictions for chironomid toxicity did not
agree with the actual test data. SARs have not been developed for benthic organisms simply
because not enough test data are available to permit such analyses.
The use of SAR is not limited to neutral organic compounds. Currently, SARs are
available for compounds that show more specific modes of toxicity or excess toxicity over the
neutral organics. These SARs are organized by chemical class and examples include acrylates,
methacrylates, aldehydes, anilines, benzotriazoles, esters, phenols, and epoxides (Clements,
1988,1994; Auer et al., 1990; Clements and Nabholz, 1994).
2-22
-------
Because the CCs were exceeded enough times out. of 1 ;year, the PDM3 model indicated a
risk to aquatic organisms. When actual sites were analyzed using EXAMS II, no unreasonable
risks were identified.
2.3.7.2.3. Ecological significance. There appears to be no unreasonable risks to pelagic and
benthic organisms at the identified use sites after 1 year's release. The potential risk posed by
the PMN substance bioaccumulating through the aquatic food: web was not thought to be
significant during the first several years of use (see Section 2.3.7.2.5, Recovery Potential).
2.3.7.2.4. Spatial and temporal patterns of the effects. CBI restrictions preclude revealing the
uses and specific sites for the PMN substance. However, OPPT assessors identified important
river systems that could be impacted by this PMN substance. Thus, if there was a risk, the
effects were not likely to be localized. However, given the restrictions in the consent order, any
future risk will be localized to known sites.
2.3.7.2.5. Recovery potential. The PMN substance is a neutral hydrophobic chemical. This
mode of toxicity is akin to a simple narcosis type of action (Auer et al., 1990; Veith and
Broderius, 1990) that is reversible if exposure to the toxicant is terminated before lethality or
death occurs. ;
The recovery potential was not evaluated. Exposures over a year were predicted to have a
low potential to cause adverse effects. However, continued exposure at the same site for a
number of years may cause some impact to benthic organisms, but OPPT does not regulate
multiyear exposures to the aquatic environment because of the greater degree of uncertainty
about future production volume and uses. OPPT assessors warned the submitter that continued
release of this alkylated diphenyl at one site could cause environmental problems in the future.
Since the alkylated diphenyl was predicted to be persistent in sediments and was expected to
continue to accumulate in sediments, the submitter could be liable for cleaning up any sediments
contaminated with this alkylated diphenyl after a decade or so of continuous use at the same
site(s).1 !
'The final regulatory decision was a significant new use restriction (SNUR, see Risk Management - Final
Decision) that limited the releases to surface water to 1 ug/L (ppb). OPPT technical staff advised the submitter's
contractors and technical contact that release of this chemical from one site over an extended period could lead to
contaminated sediments and that the company might be liable for cleanup if monitoring determined that the
sediments were sufficiently contaminated. Use of this PMN chemical substance began in 1993. As a SNUR was
attached to the PMN the submitter had to inform the users and local regulatory authorities about the SNUR and its
limitations. This resulted in the environmental protection agency of amidwestern State monitoring the PMN
2-23 :
-------
2.3.7.3. Risk Management—Final Decision
i i
The OPPT Division Directors and other risk managers agreed that the PMN substance
posed no unreasonable risks to pelagic aquatic organisms at the specific sites of use and disposal.
However, there could be risks at other sites through the use and disposal of the PMN substance.
Therefore, the final disposition was a significant new use restriction (SNUR), including a
restriction against releasing concentrations higher than 1 ug/L (the concern level for the PMN
substance). The submitter must also submit a significant new use notice if it wants to use the
PMN substance at sites other than the ones identified in its submission.
i
2.3.8. Discussion of Case Study
As outlined in Figure 2-1, this is an example of how the Framework for Ecological Risk
Assessment (U.S. EPA, 1992) and the Guidelines for Ecological Risk Assessment (U.S. EPA,
1998) are consistent with the underlying structure of a new chemical assessment by OPPT. It is
also a real-world demonstration of the iterative manner in which the ecological risk assessment
of new industrial chemicals can be evaluated by EPA.
A large majority of the new chemical evaluations performed by OPPT do not make such
a risk-based case and therefore do not need to undergo this level of assessment. Even though this
evaluation may not be typical of a new chemical, it proved useful in illustrating (1) the depth of
ecological risk assessment that is feasible in OPPT, (2) the routine and pragmatic use that has
been made of SAR, and (3) the routine and pragmatic use of the assessment (uncertainty) factors
that were developed by OPPT for new chemical evaluations (U.S. EPA, 1984).
Discussion of the empirical basis for the development of these assessment factors is to be
found in the OPPT report on how concern levels (i.e., concern concentrations) in the
environment are to be determined by the use of these assessment factors (U.S. EPA, 1984). This
simple approach also is mentioned and elaborated on in other, more recent publications (Auer et
al., 1990; Nabholz, 1991; Zeeman and Gilford, 1993; Zeeman et al. 1995).
Indeed, the simple assessment factor method developed by OPPT in the early 1980s
remains a very pragmatic and effective tool for estimating the levels of concern (risk) for
industrial chemicals released into the aquatic environment. Its use has been supported by
analyses comparing it with more complex statistical methods (Calabrese and Baldwin, 1993;
Forbes and Forbes, 1993,1994; Zeeman, 1995).
substance in the sediments and the fish of a stream receiving effluent containing the PMN substance. In 1997 the
State EPA monitoring efforts found measurable concentrations of the PMN substance in sediments and in fish
fillets. Hie concentrations in fish fillets were determined to range as high at 1.65 mg/kg (ppm) fresh weight.
2-24
-------
2.3.9. Summary of Case Study
This is a relatively comprehensive example of OPPT's capabilities in conducting
ecological risk assessments for new chemical substances. It illustrates the consistency among
OPPT's approach, EPA's 1992 Framework document, and EPA's 1998 Guidelines for
Ecological Risk Assessment.
The essential features in this case study reflect several practical considerations. TSCA
requires the manufacturer or importer of new industrial chemicals to submit a PMN to EPA 90
days before it intends to begin manufacturing or importing. Because actual test data are not
required to be developed as part of a PMN submission, OPPT must frequently use SARto
estimate both ecological effects and exposure/fate characteristics (such as physical/chemical
properties and biodegradation). Because test data for new chemicals are seldom available, an
empirical set of assessment factors (or uncertainty factors) was developed by OPPT and are
routinely used in the ecological risk assessment of PMNs.
This study focuses on the assessment of a PMN substance, i.e., an alkylated diphenyl, that
is a neutral organic compound. Chemicals belonging to this class of compounds elicit a
nonspecific and simple form of toxicity known as narcosis. The toxicity of neutral organic
compounds can be estimated through. S ARs that correlate toxicity with the octanol-water
partition coefficient (Kow) and molecular weight. The subject PMN substance had a predicted log
K^ of 6.7. Compounds with such a Mgh log Kow are not expected to be acutely toxic (i.e., no
acute effects at saturation over short-term exposure durations), but are expected to elicit chronic
effects following long-term exposures. Actual testing of the PMN substance confirmed these
predictions of ecotoxicity.
The PMN submitter identified processing, use, and disposal sites adjacent to rivers and
streams since the chemical was to be imported into the United States. Because it was expected
that the PMN substance would be discharged into such environments, pelagic and benthic aquatic
populations, communities, and ecosystems were considered to be at risk. Therefore, the
assessment endpoint used in this study was the protection of aquatic organisms (e.g., fish, aquatic
invertebrates, and algae). Measurement endpoints used to evaluate the risks to aquatic organisms
(the assessment endpoint) were mortality, growth and development, and reproduction.
Initial exposure concentrations were estimated using a simple dilution model that divided
releases (kg/day) by stream flow (millions of liters/day). Subsequent exposure analyses used a
probabilistic dilution model and the exposure analysis modeling system. PDM3 was used to
estimate the number of days a particular effect concentration would be exceeded in 1 year, and
EXAMS II was used to estimate concentrations in the water column and in sediments using site-
specific data. |
2-25
-------
Toxicity initially was predicted by the use of S AR. Aquatic toxicity test data
I
accompanying the PMN submission and later receipt of additional test data confirmed the
accuracy of these SAR predictions. Assessment (uncertainty) factors were used to determine the
concern level or concern concentration (CC) in the receiving stream. This stream water-column
CC was set at 1 ug/L (ppb). When the OPPT risk assessment determined that this CC was
exceeded for more than 20 days, a potential unreasonable risk was assumed to be expected if this
PMN chemical substance was allowed to be used.
In risk characterization, the quotient method was used to compare exposure
concentrations with the ecological effect concentrations. A ratio of 1 or greater indicated a
potential risk. The PMN evaluation resulted in five iterations of analysis and risk
characterization. The first four iterations identified an ecological risk and resulted in the
collection of additional and more specific ecological effects test data and more detailed
information on potential exposures to the PMN substance. The final outcome was that the PMN
substance could be used only at the identified sites because there was uncertainty as to whether
the concern level (1 pg/L) might be exceeded at sites not identified and characterized by the
submitter.
2.4. RISK ASSESSMENT METHODOLOGY DEVELOPMENT
The OPPT methodology for the ecological risk assessment of new chemicals was
developed more than a decade ago (Zeeman and Gilford, 1993) and reflects several regulatory
constraints within which OPPT had to operate. There was a need to assess large numbers of new
chemicals, typically in a short tune frame and typically with a minimal level of data provided
with which to perform this risk assessment (i.e., seldom were physical/chemical properties,
environmental fate, or ecotoxicity data provided). From these restrictions it was obvious that the
methods of ecological risk assessment used by OPPT had to be very pragmatic. Ecological risk
assessors from OPPT were involved in and played a major role hi the development of the EPA
framework and guidelines documents. Therefore, the extant OPPT methodology for ecological
risk assessment proved to be both useful and illustrative in the development of many of the
principles and practices espoused in each of these documents.
2.5. RISK MANAGEMENT
As is evident from the new chemical assessment case study above, there were five
iterations in characterizing the risk of this chemical to organisms in the environment (Table 2-5).
It is also plain that the risk management decisions made here played a key role in deciding on the
next steps for each of these iterations. This case study is illustrative of how an efficient and
2-26
-------
pragmatic ecological risk assessment process can assist in eliciting reasonable risk management
decisions. These risk management decisions helped to develop the kinds of information needed
to perform an adequate risk assessment and to come to closure on the regulatory actions
determined to address and/or mitigate the ecological risks expected from allowing the use of this
chemical.
2.6. REFERENCES
Auer, CM; Nabholz, JV; Baetcke, KP. (1990) Mode of action and the assessment of chemical hazards in the
presence of limited data: use of structure activity relationships (SAR) under TSCA, section 5. Environ Health
Perspect 87:183-197.
Banerjee, S; Yalkowsky, SH; Valvani, SC. (1980) Water solubility and octanol/water partition coefficients of
organics. Limitations of the solubility-partition coefficient correlation. Environ Sci Technol 14:1227-1229.
Barnthouse, LW; Suter, GW; Bartell, SM; et al. (1986) User's manual for geological risk assessment. Oak Ridge
TN: Oak Ridge National Laboratory. ORNL pub. no. 2679. :
Broderius, SJ; Russom, CL. (1989) Mode of action-specific QSAR models for predicting acute and chronic toxicity
of industrial chemicals to aquatic organisms. Prepared for the Environmental Research Laboratory, U.S. EPA,
Duluth MN. Deliverable No. 81421.
Burns, LA. (1989) Exposure analysis modeling system: user's guide for EXAMS II version 2.94. Prepared for the
Environmental Research Laboratory, U.S. EPA, Athens, GA. :
Calabrese, EJ; Baldwin, LA. (1993) Performing ecological risk assessments. Boca Raton, FL: Lewis Publishers.
Clements, RG. (1983) Environmental effects of regulatory concern under'TSCA—a position paper. Prepared for the
Environmental Effects Branch, Health and Environmental Review Division (7403), EPA Office Of Toxic
Substances, U.S. Environmental Protection Agency, Washington, DC.
Clements, RG, ed. (1988) Estimating toxicity of industrial chemicals to aquatic organisms using structure activity
relationships. Prepared for the Environmental Effects Branch, Health and Environmental Review Division (7403),
Office Of Toxic Substances, U.S. Environmental Protection Agency, Washington, DC. EPA/560/6-88-001.
Clements, RG, ed. (1994) Estimating toxicity of industrial chemicals to aquatic organisms using structure activity
relationships: 2nd edition. Prepared for the Environmental Effects Branch, Health and Environmental Review
Division (7403), Office Of Pollution Prevention and Toxics, U.S. Environmental Protection Agency, Washington,
DC.EPA/748/R-93/001.
Clements, RG; Nabholz, JV. (1994) ECOSAR: a computer program for estimating ecotoxicity of industrial
chemicals based on structure activity relationships—User's guide. Prepared for the Environmental Effects Branch,
Health and Environmental Review Division (7403), Office Of Pollution Prevention and Toxics, U.S. Environmental
Protection Agency, Washington, DC. EPA/748/R-93/002.
Forbes, TL; Forbes, VE. (1993) A critique of the use of distribution-based extrapolation models in ecotoxicology.
Funct Ecol 7:249-254.
Forbes, VE; Forbes, TL. (1994) Ecotoxicology in theory and practice. New York: Chapman and Hall.
2-27
-------
Grain, CF. (1982) Vapor pressure. In: Handbook of chemical property estimation methods, environmental behavior
of organic compounds. Lyman, WJ; Reehl, W; Rosenblatt, DH, eds. New York: McGraw-Hill Co, 14:1-20.
: ' I
Hermens, J; Canton, H; Janssen, P; et al. (1984) Quantitative structure-activity relationships and toxicity studies of
mixtures of chemicals with anesthetic potency: acute lethal and sublethal toxicity to Daphnia magnet. Aquat Toxicol
5:143-154.
INFORM. (1995) Toxics watch 1995. New York: INFORM, Inc., 816 pp.
Karickhoff, SW; Brown, DS; Scott, TA. (1979) Sorption of hydrophobic pollutants on natural sediments. Water Res
13:241-248.
Konemaim, H. (1981) Quantitative structure-activity relationships in fish toxicity studies. Part 1: relationship for 50
industrial pollutants. Toxicology 19:209-221.
Leo, A; Weininger, D. (1985) CLOGP version 3.3. Estimation of the n-octanol/water partition coefficient for
organics hi the TSCA industrial inventory. Claremont, CA: Pomona College.
Lipnick, RL. (1985) Validation and extension offish toxicity QSARs and interspecies comparisons for certain
classes of organic chemicals. In: QSAR in toxicology and xenobiochemistry. Tichy M, ed, Amsterdam: Elsevier
Press, pp. 39-52.
Moss, K; Locke, D; Auer, C. (1996) EPA's new chemicals program. Chem Health Safety 3(l):29-33.
Nabholz, JV. (1991) Environmental hazard and risk assessment under the United States Toxic Substances Control
Act. Sci Total Environ 109/110:649-665.
Nabholz, JV; Zeeman, M; Rodier, D. (1998) Case study No. 1: assessing the ecological risks of a new chemical.
Case study No. 1: In: Uncertainty analysis in ecological risk assessment. Warren-Hicks, W; Moore, D, eds.
Pensacola, FL: SETAC Press, pp. 207-225.
Nabholz, JV; Miller, P; Zeeman, M. (1993a) Environmental and risk assessment of new chemicals under the Toxic
Substances Control Act (TSCA) section five. In: Environmental toxicology and risk assessment. Landis, WG;
Hughes, JS; Lewis, MA, eds. ASTM STP 1179. Philadelphia, PA: American Society for testing and Materials, pp.
40-55.
Nabholz, JV; Clements, RG; Zeeman, MG; et al. (1993b) Validation of structure activity relationships used by
EPA's Office of Pollution Prevention and Toxics for the environmental hazard assessment of industrial chemicals.
In: Environmental Toxicology and Risk Assessment. Gorsuch, JW; Dwyer, FJ; Ingersoll, CG; et al., eds. ASTM
STP 1216. Philadelphia, PA: American Society for Testing and Materials, pp. 571-590.
i i
Rodier, DJ; Mauriello, D. (1993) The quotient method of ecological risk assessment and modeling under TSCA: a
review. In: Environmental toxicology and risk assessment. Vol. 2. Landis, WG; Hughes, JS; Lewis, MA, eds.
ASTM STP 1179. Philadelphia, PA: American Society for Testing and Materials, pp. 80-91.
Smrchek, J; Clements, R; Morcock, R; et al. (1993) Assessing ecological hazard under TSCA: methods and
evaluation of data. In: Environmental toxicology and risk assessment. Landis, WG; Hughes, JS; Lewis, MA, eds.
ASTM STP 1179. Philadelphia, PA: American Society for Testing and Materials, pp. 22-39.
Suter, GW, II. (1990) Endpoints for regional ecological risk assessment. Environ Manage; 14:9-23.
! !
U.S. Environmental Protection Agency. (1982) Surrogate species workshop: workshop report. Environmental
Effects Branch, Health and Environmental Review Division (7403), Office of Toxic Substances, Washington, DC:
EPA Contract No. 68-01-6554.
2-28
-------
U.S. Environmental Protection Agency. (1983) Testing for environmental effects under the Toxic Substances
Control Act. Environmental Effects Branch, Health and Environmental Review Division (7403), Office of Toxic
Substances, Washington, DC.
U.S. Environmental Protection Agency. (1984) Estimating concern levels for concentrations of chemical substances
in the environment. Environmental Effects Branch, Health and Environmental Review Division (7403), Office of
Toxic Substances, Washington, DC. :
U.S. Environmental Protection Agency. (1985) Toxic Substances Control Act test guidelines; final rules. Federal
Register 50(188):39252-39516. \
U.S. Environmental Protection Agency. (1986) New chemical review process manual. Chemical Control Division
(7405), Office of Toxic Substances, Washington, DC. EPA/560/3-86/002.
U.S. Environmental Protection Agency. (1988) User's guide to PDM3: final report. Exposure Assessment Branch,
Exposure Evaluation Division (7406), Office of Toxic Substances, Washington, DC, under EPA contract no. 68-02-
4254, task no. 117.
U.S. Environmental Protection Agency. (1992) Framework for ecological risk assessment. Risk Assessment Forum,
Office of Research and Development, Washington, DC. EPA/630/R-92/001.
U.S. Environmental Protection Agency. (199'3) PMN ecotox database: a data base of environmental toxicity studies
which are protected by confidential business information (CBI). Environmental Effects Branch, Health and
Environmental Review Division (7403), Office of Pollution Prevention and Toxics, Washington, DC.
U.S. Environmental Protection Agency. (1994) Ecological risk assessment case study: assessing the ecological risks
of a new chemical under the Toxic Substances Control Act. In: A review of ecological assessment case studies from
a risk assessment perspective. Vol. II. Risk Assessment Forum, Office of Research and Development, Washington,
DC. EPA/630/R-94/003, pp. 1-1 to 1-35.
U.S. Environmental Protection Agency. (199'8, May 14) Guidelines for ecological risk assessment. Federal Register
63(93):26846-26924. I
Veith, GD; Broderius, SJ. (1990) Rules for distinguishing toxicants that cause type I and type II narcosis syndromes.
Environ Health Perspect 87:207-211.
Veith, GD; Call, DJ; Brooke, LT. (1983) Structure activity relationships for the fathead minnow, Pimiphales
promelas: narcotic industrial chemicals. Can J Fish Aquat Sci 40:743-748.
Wagner, PM; Nabholz, JV; Kent, RJ. (1995) The new chemicals process at the Environmental Protection Agency
(EPA): structure-activity relationships for hazard identification and risk assessment. Toxicol Lett 79:67-73.
Zeeman, M. (1995) Ecotoxicity testing and estimation methods developed under section 5 of the Toxic Substances
Control Act (TSCA). In: Fundamentals of aquatic toxicology: effects, environmental fate, and risk assessment. 2nd
ed. Rand, G, ed. Washington, DC: Taylor and Francis, pp. 703-715.
Zeeman, M. (1996) Our fate is connected with the animals (book review of Our stolen future). BioScience 46:542-
546. '•
Zeeman, M. (1997) Aquatic toxicology and ecological risk assessment: US-EPA/OPPT perspective and OECD
interactions. In: Ecotoxicology: responses, biomarkers, and risk assessment. Zelikoff, JT; Lynch, J; Schepers, J, eds.
Organization for Economic Cooperation and Development (OECD), Paris. Published for the OECD by SOS
Publications, Fair Haven, NJ, pp. 89-108.
2-29
-------
Zeeman, M; Gilford, J. (1993) Ecological hazard evaluation and risk assessment under EF'A's Toxic Substances
Control Act (TSCA): an introduction. In: Environmental toxicology and risk assessment. Landis, WG; Hughes, JS;
Lewis, MA, eds. ASTM STP 1179. Philadelphia, PA: American Society for Testing and Materials, pp. 7-21.
Zeeman, M; Nabholz, JV; Clements, RG. (1993) The development of SAR/QSAR for the use under EPA's Toxic
Substances Control Act (TSCA): an introduction. In: Environmental toxicology and risk assessment. Vol. 2.
Gorsuch, JW; Dwyer, FJ; Ingersoll, CG; et al., eds. ASTM STP 1216. Philadelphia, PA: American Society for
Testing and Materials, pp. 523-539.
Zeeman, M; Auer, CM; Clements, RG; et al. (1995) U.S. EPA regulatory perspectives on the use of QSAR for new
and existing chemical evaluations. SAR and QSAR Environ Res 3(3): 179-202.
2-30
-------
3. ECOLOGICAL RISK ASSESSMENT UNDER FIFRA
3.1. SUMMARY
The ecological risk assessment methodologies under the Federal Insecticide, Fungicide,
and Rodenticide Act (FIFRA) are consistent with current ecological risk assessment guidelines
(EPA, 1998). FIFRA requires prospective assessments of pesticides in a tiered framework.
Typically, the industry generates environmental fate and effects data and submits it to the EPA
Office of Pesticide Programs (OPP). (DPP evaluates the data and conducts the risk assessment.
This chapter discusses the generalized process for effects and exposure analyses and
assessment methods. For effects analysis, the tiers move from acute toxicity test data to
subchronic and chronic toxicity data to field, farm, pond, or mesocosm studies. In exposure
analysis, level 1 uses conservative assumptions in exposure models. These are refined with site-
specific data, pesticide use information, use of more complex exposure models, and the
application of probability modeling in higher levels. In risk characterization, the quotients
(exposure/effects) are used at lower tiers, with more complex methods often used at higher tiers.
The registration and reregistration of pesticides under FIFRA is a cost-benefit statute that
balances no unreasonable adverse effects to human health or the environment with economic
issues, societal concerns, and political and legal factors. Ecological effects are often mitigated
through reduction in application frequency, dose, specific crop or area use, or other restrictive
requirements.
3.2. INTRODUCTION
This chapter discusses ecological risk assessment methods and approaches used by EPA's
Office of Pesticide Programs (OPP). The chapter's specific objectives are to:
• Provide a regulatory context for ecological risk assessment under the Federal Insecticide,
Fungicide, and Rodenticide Act (FIFRA)
• Discuss ecological risk assessment in pesticide regulatory operations as part of a broader
risk management and decision-making context
• Summarize the application of file ecological risk assessment in risk management decision
making.
3-1
-------
3.3. REGULATORY CONTEXT FOR PESTICIDE REGISTRATION AND
REREGISTRATION
FIFRA gives EPA the authority to register pesticides to ensure no unreasonable adverse
effects to human health or the environment, taking into account the economic, social, and
i j , |
environmental costs and benefits of the pesticide use. As such, FIFRA is a cost-benefit statute,
and an "unreasonable adverse effect" on the environment is a regulatory determination that must
account for scientific as well as economic, social, and governmental cost and benefit factors.
Under FIFRA authority, EPA regulates insecticides, herbicides, fungicides, rodenticides,
disinfectants, plant growth regulators, biological agents, and other pest control agents. The
primary regulatory vehicle under FIFRA is the pesticide label. Every registered pesticide
i
product must bear a label that includes the producer number, product registration number, active
ingredient statement, warning or precautionary statements, and directions for use.
EPA's Office of Pesticide Programs currently reviews about 5,000 pesticide registration
submissions annually. The scope of the submissions ranges from simple label amendments to
registration of new active ingredients. Since 1947, thousands of pesticide products have been
registered. Perhaps not surprisingly, standards for approval and test data requirements reflect
changes in science and pesticide regulatory policy over tune. To ensure compliance with current
scientific and regulatory standards, FIFRA also requires the review and reregistration of existing
pesticides. During reregistration, registrants may delete pesticide uses or voluntarily withdraw
products or uses. Further, EPA has the authority to cancel registrations for pesticide products
that do not meet the requirements for reregistration. Since 1988, the registered products subject
to reregistration have declined from approximately 50,000 to about 20,000.
Following registration or reregistration, problems that arise during the use of a particular
!
pesticide may be investigated under the special review process. Special review consists of
scientific and legal analysis before a major regulatory decision is made on a registered pesticide.
Special review is conducted by notice-and-comment rulemaking. Science issues are developed
and presented to the FIFRA Scientific Advisory Panel for review. Additionally, the U.S. Food
and Drug Administration and congressional committees are invited to provide formal comments.
i
Once a decision is made, the registrant may appeal the decision through administrative procedure
i
or judicial review.
3.4. RISK MANAGEMENT
Ecological risk assessment hi pesticide regulatory operations is best viewed as the
application of regulatory science in a risk management context. This view is supported by
emerging risk-based approaches to environmental regulations (Thomas, 1987; Science Advisory
3-2
-------
Board, 1990a,b,c,d), which promote increased integration, of societal values, science, and risk
mitigation practices. The integrated decision-making process involves the following three
interactive phases:
1. Risk assessment is a science-based activity that consists of hazard characterization and
exposure characterization and ultimately integrates the;two into a risk characterization.
2. Risk mitigation involves remediation or mitigation measures to reduce or eliminate
source contamination and adverse environmental impacts.
3. Risk management is a policy-based activity that defines risk assessment questions and
endpoints to protect human health and the environment. It takes the scientific risk
assessment; incorporates social, economic, political, and legal factors that impinge on or
influence the final decision; and selects regulatory actions.
The underlying principles behind risk reduction and integrated decision making are detailed in
the strategic initiatives and guiding principles recently released by EPA (1994) and include
ecosystem protection, pollution prevention, strong science and data, partnerships, and
environmental accountability. In essence, the emerging policies are directed toward greater
participation hi environmental problem solving and decision making, including risk assessors,
risk managers, and parties affected by the decision (regulated community, user groups,
environmental interest groups, general public, and scientists).
Improved understanding of the different perspectives of risk assessors and risk managers
is crucial to the ultimate success of integrated decision-making processes. Risk assessors are
generally concerned with performing risk assessments in the most scientifically credible manner
and identifying additional data or research to better characterize risk.
In contrast, risk managers have little interest in the scientific nuances or technical details
surrounding an ecological risk assessment. Rather, they may be primarily concerned with
integrating ecological risk conclusions into a broader risk or risk-benefit framework to finalize
regulatory decisions. The decision may include imposing risk reduction or risk mitigation
control practices rather than undergoing successive iterations of the original risk assessment.
Risk reduction or risk mitigation activities are becoming increasingly important risk management
tools. For pesticides, such activities frequently include changes in or restrictions for specific
uses, manifested as label changes.
Risk assessors must be aware of risk management needs in the problem formulation stage
to ensure that the assessment endpoints and resolving power that the decision maker requires are
understood. Ideally, discussions should occur during a forma:! a priori problem formulation step
3-3
-------
in the assessment process. Once risk assessors and risk managers have agreed on assessment
goals and objectives, it falls to the risk assessor to design and conduct the risk assessment.
Routine problem formulation that engages both risk assessors and risk managers is increasing,
but has not been commonly practiced in the past. This has sometimes led to differing
expectations between risk assessors and risk managers regarding the objectives, scope, and
i
application of a risk assessment. The importance of promoting formal problem formulation
cannot be overstated.
3.5. RISK ASSESSMENT METHODS IN PESTICIDE REGULATORY OPERATIONS
Registration and reregistration decisions are based in part on the evaluation, synthesis,
and integration of pesticide studies conducted by registrants and submitted to the Agency.
Studies are routinely conducted in mammalian toxicology, occupational and residential exposure,
environmental fate and transport, and ecological effects. Individual studies are evaluated by EPA
scientists and subsequently used hi human health and ecological risk assessments. The risk
assessments are then used by regulatory decision makers, who make the final risk management
decisions. Only ecological risk assessment will be further considered here.
Ecological risk assessment methods and procedures under FIFRA are detailed elsewhere
(40 CFR 158.130; 40 CFR 158.145; Urban and Cook, 1986; Fite et al., 1988; Touart, 1988;
SETAC, 1994; Touart, 1995; Touart and Maciorowski, 1997) and only briefly described here.
Existing methods predate EPA's ecological risk assessment framework (U.S. EPA, 1992) and
guidelines (U.S. EPA, 1998). However, two pesticide case studies (carbofuran, synthetic
pyrethroids) were used in the Agency's state of the practice for ecological risk assessment
prepared during the guidelines development process (U.S. EPA, 1993). Further, the traditional
FIFRA ecological risk assessment approach is consistent with the ecological risk assessment
framework process and includes problem formulation, exposure characterization, effects
characterization, and risk characterization.
Generally, ecological risk assessments for pesticide registration are prospective estimates
based on single active ingredients and use sites (e.g., corn, wheat, ornamental plants, etc.). The
scope and complexity of any specific pesticide risk assessment will vary with the specific
chemical and use, but a tiered iterative approach is generally used. The tiers progress from
simple risk quotients derived from laboratory fate, transport, and toxicity data in early tiers to a
weight-of-evidence approach in later tiers (Tables 3-1 and 3-2).
Exposure analysis may consist of a preliminary or comprehensive fate and transport
i . i
assessment (Table 3-1) based on registrant-submitted data. The exposure analysis provides
exposure profiles and estimated environmental concentrations (EEC) for the pesticide use (e.g.,
3-4
-------
Table 3-1. Generalized exposure analysis and assessment methods and
procedures used in prospective ecological risk screens of pesticides3
Preliminary exposure analysis includes simple laboratory tests and models to provide an
initial fate profile for a pesticide (hydrolysis and photolysis in soil and water, aerobic and
anaerobic soil metabolism, and mobility).
Fate and transport assessment provides a comprehensive profile of the chemical
(persistence, mobility, leachability, binding capacity, degradates) and may include field
dissipation studies, published literature, other field monitoring data, ground-water studies, and
modeled surface water estimates.
Estimated environmental concentrations (EEC) are derived during the exposure analysis or
comprehensive fate and transport assessment. There are four EEC estimation procedures:
Level 1: A direct-application, high-exposure model designed to estimate direct
exposure to a nonflowing, shallow-water (<15 cm) system.
Level 2: Adds simple drift or runoff exposure variables such as drainage basin size,
surface area of receiving water, average depth, pesticide solubility, surface runoff, or
spray drift loss, which attenuate the Level 1 direct application model estimate.
Level 3: Computer runoff and aquatic exposure simulation models. A loading model
(SWRBB-WQb, PRZM°, etc,,) is used to estimate field losses of pesticide associated
with surface runoff and erosion; the model then serves as input to a partitioning model
(EXAMS IId) to estimate sorbed and dissolved residue concentrations. Simulations are
based on either reference environment scenarios or environmental scenarios derived
from typical pesticide use circumstances.
Level 4: Stochastic modeling where EECs are expressed as exceedance probabilities
for the environment, field, and cropping conditions.
Tor additional details regarding environmental fate data requirements, see 40 CFR § 158.130,
SETAC (1994), and Touart (1995).
•"Simulator for Water Resources in Rural Basins-Water Quality. j
Testicide Root Zone Model.
dExposure Analysis Modeling System. !
3-5
-------
Table 3-2. Generalized ecological effects analysis and risk quotient
methods and procedures used in prospective risk screens of pesticides
Tier I effects analysis provides acute toxicity values and dose-response information
(mammalian and avian acute oral LD50; avian dietary LC50; seedling emergence and vegetative
vigor EC2s; honeybee acute contact LD50; and additional wild mammal, esituarine, and plant
tests depending on pesticide use category).
Tier BE effects analysis provides subchronic and chronic toxicity values (NOEC) including
avian reproduction studies; special avian or mammal studies; fish early life stage studies;
invertebrate life cycle studies; and a fish bioaccumulation factor.
Tier IDE effects analysis provides refined NOEC estimates for chronic toxicity that may
include a fish full life cycle test, aquatic organism accumulation, or food chain transfer tests
The quotient method is used to provide a set of acute and chronic risk quotients (RQ) for
fish, birds, invertebrates, plants, and endangered species. The RQs are calculated by dividing
exposure (EEC) by hazard (LD50 or LC50 or NOEC). Risk quotients are men compared to
regulatory risk criteria as follows:
Presumption of
acceptable risk
Presumption of risk
that may be mitigated
by restricted use
Presumption of
unacceptable risk
Nonendangered species Endangered
species
Acute toxicity
EECO.l LCso
Chronic toxicity
EECs chronic NOEC
0.1LC50<;EEC:>0.1LC50
N/A
EEC <; 0.50 LC50
EEC > NOEC
EEC<: 0.05 LC50 or
EC^0.10LC10
EEC > NOEC
Tier IV effects analysis allows registrants to rebut a presumption of risk derived from
laboratory studies by performing field or simulated field studies, including qualitative
terrestrial field studies, farm pond studies, mesocosm studies, or other special studies.
'For additional details regarding ecological effects data requirements, see 40 CFR § 158.145 Subdivision E; Urban
and Cook, 1986; SETAC, 1994; and Touart, 1995.
3-6
-------
corn, cotton, wheat, etc.). Note that EECs may be derived from four estimation procedures
ranging from simple to complex. The ecological effects analysis (Table 3-2) is also tiered. Tier I
provides an acute toxicity profile for birds, fish, mammals, and invertebrates. Tier II provides a
subchronic and chronic toxicity (no-observed-effects concentration, or NOEC) profile and
bioaccumulation potential for the same test species. Depending on the hazard and exposure
characteristics of a particular pesticide and use pattern, Tier II analyses may be conducted for all
representative taxa, or may focus on either aquatic or terrestrial species. When warranted, Tier
III effects analysis is used to refine NOEC and bioaccumulation estimates.
Following exposure and effects analysis, ecological risk is estimated as a function of
ecotoxicological effects and environmental exposure using the quotient method (Table 3-2). A
number of risk quotients are calculated (e.g., acute avian, acute fish, acute invertebrate, chronic
avian, chronic fish, chronic invertebrate, etc.) and compared with regulatory risk criteria (e.g.,
presumption of acceptable risk, presumption of unacceptable risk, etc.). Traditionally, if
regulatory criteria are exceeded, a high-risk potential is assumed to exist for the pesticide-use
combination. If a registrant wishes to refute a presurnption-of-risk finding, a Tier IV effects
analysis, consisting of field studies, simulated field studies, or other special studies, may be
conducted (Fite et al.,1988; Touart, 1988).
3.5.1. Application of Ecological Risk Assessments in Pesticide Regulatory Decision Making
The application of ecological risk assessments in pesticide regulatory decisions is subject
to practical constraints imposed by law, regulatory policy, and precedent. In October 1992,
EPA's Office of Prevention, Pesticides, and Toxic Substances (OPPTS) released a set of policy
decisions following a comprehensive review of ecological and environmental fate data
requirements for registraticti and reregistration of pesticides. Issues considered in the review
included resource requirements necessary to review data, the utility of data in assisting regulatory
decision making by risk managers, and the impact of data requirements on meeting
congressionally mandated deadlines for the reregistration of pesticides already in use. Major
points of the policy decisions are paraphrased below.2
2In response to the OPPTS policy decisions, a number of actions were initiated by OPP to develop a
strategy for implementation. Foremost was the development of the Ecological Fate and Effects Implementation
Work Group. This group developed an implementation strategy that subsequently led to the formation of the
Aquatic Risk Assessment and Mitigation Dialogue Group. The mission of the Dialogue Group was to discuss
pesticide risk assessment and risk reduction for aquatic systems and recommend methods and use of risk mitigation
measures in regulatory decision making. The background material leading to the Dialogue Group and its
deliberations and final recommendations is detailed in SETAC (1994) and summarized briefly here.
3-7
-------
• Establish risk-based priorities to allow protective decisions in a timely fashion.
i -
• I
• Base decisions primarily on laboratory studies, with less dependence on terrestrial and
aquatic field studies.
i i
• Provide better integration of risk assessment and risk management processes.
• Use risk mitigation to the extent feasible to achieve acceptable risk reduction.
i i
• Develop the concept of continuous improvement and develop strategies to characterize
I ! I
long-term ecological risk with less uncertainty.
Aquatic field studies or simulated field studies were conducted with the objective of
rebutting the presumption of risk identified when regulatory criteria as described in Tables 3-1
and 3-2 were not met. The policy recommendations., using the same methods and procedures,
promote envkonmentally protective decisions through early application of mitigation actions to
reduce the off-field movement of pesticides, and therefore reduce risk to nontarget organisms.
There are also provisions for more sophisticated assessment procedures, which allow for
i |
probabalistic estimates of levels of concern. The use of mitigation and monitoring also shifts the
assessment from a solely a priori process to one of a posteriori monitoring and mitigation. The
sections of the document that follow represent a proposed set of procedures to implement
mitigation and to assess the efficacy of mitigation and the adequacy of the ri.sk assessment
procedure.
I
3.5.2. The Risk Identification and Mitigation Process
Pesticide registration and reregistration processes are considered to be iterative, as
presented in Tables 3-1 and 3-2. That is, the database supporting the current registration status of
a pesticide will be periodically reviewed and evaluated to ensure that it meets current scientific
requirements, standards, and regulatory policies. The verification of risk mitigation steps is
based on this cyclic evaluation of the pesticide in light of new or additional information. For the
purpose of discussion, the risk identification/mitigation process begins any time a pesticide is
reviewed for the purpose of registration or reregistration.
When a pesticide undergoes evaluation for registration or reregistration, the scientific
experts review and evaluate the data available in a comprehensive manner to ensure it meets the
standards established for carrying out risk assessments. The database is evaluated and integrated
3-8
-------
in such a manner that routes of dissipation, significant environmental degradates, residue levels,
and tune of persistence of degradates in the various environmental compartments are elucidated.
This information, along with the hazards of the pesticide as determined in the required studies
and available incident data, is used to determine what level of concern exists in each of several
compartments in the environment If a level of concern is unacceptable, then risk
mitigation/verification procedures are initiated.
In the registration and reregistration processes, a conclusion that an unacceptable risk will
result from the proposed or registered use(s) of pesticides is immediately passed to the
appropriate divisions within OPP. The Registration Division and Special Review and
Reregistration Division pass this infoimation on to the registrant(s). The purpose of notification
is to ensure that everyone is actively involved in the process of identifying appropriate risk
reduction measures. Once OPP and ti'ae registrant(s) have concluded their work on appropriate
risk mitigation steps, negotiations between OPP and the registrants) on the label changes
necessary to reduce the risk begin. j
The product of the negotiation is a set of mitigation actions, to which OPP agrees, that
effectively reduce the risk to an acceptable level. As this process begins, data to support the
effectiveness of the mitigation steps will be nonexistent or limited in scope. To ensure the
effectiveness of the mitigation steps, the Agency may require some sort of verification data.
Once mitigation measures have been identified and implemented, post-registration or post-
reregistration monitoring may be required to verify the efficacy of the risk mitigation measures.
Quantifiable verification of effectiveness of the mitigation may take several years. The
verification data would then be reviewed to evaluate the effectiveness of the mitigation measures.
3.6. RISK ASSESSOR AND RISK MANAGER COMMUNICATION
Once an ecological risk characterization is passed to a risk manager, additional
communication and discussion is necessary. Presented with a scientific evaluation of risk, the
risk manager may want additional inirbrmation or study, or may need to act on the information in
hand regardless of its scientific strengths or shortcomings. Rather than refine the risk
assessment, a risk manager may opt to impose mitigation to reduce the risk, even in the face of
uncertainty that the mitigation will be effective. When such situations occur, risk assessors must
clearly and succinctly summarize risk, uncertainties, and options for the benefit of risk managers,
stakeholders, and the public at large. Further, risk assessors must be willing to discuss the
relative merits of risk mitigation even in the absence of data.
Although there is general agreement that risk assessors need to be involved in risk
management decisions, their involvement is also important to ensure the scientific integrity of
3-9
-------
the risk assessment process. Once a risk characterization is used to reach a decision, the risk
assessor rarely has an opportunity to request more data or information on which to base opinions
or recommendations. More important, the risk assessor has now moved into the risk
management arena. In the risk management decision process, the risk assessor may be asked to
analyze or judge the effect of proposed risk mitigation on the original risk aissessment. This does
not change the original risk assessment, which serves as a baseline estimate, but the analysis may
begin here as to whether management actions such as mitigation will reduce risk to acceptable
levels.
Until the overall integrated decision-making process is better defined and understood by
both risk assessors and risk managers, there undoubtedly will be some controversy regarding the
application of ecological risk assessments in regulatory operations. However, recognizing and
i i . i
understanding that risk assessors and risk managers have different roles and responsibilities
should go a long way toward improving the decision process.
3.7. NEXT STEPS
Although the process described above has been partially implemented in decision
making, full implementation requires action on the following recommendations promoted by the
Dialogue Group (SETAC, 1994). EPA-OPP is actively pursuing these recommendations through
technical committees.
i
• Integrated probabilistic risk assessments that include both the probability of exposure and
effects should be implemented within OPP.
• Improved capabilities for predictive risk assessments through tiered modeling and
focused laboratory studies should be encouraged, and when conducted should be included
as part of the risk assessment.
• Mitigation must provide meaningful ecological risk reduction, be pragmatic and
achievable, and consider the need for timely decisions and cost-effective utilization of
financial and human resources.
• The effectiveness of the paradigm in improving the ability of EPA to implement timely
protective environmental decisions must be monitored and evaluated on a routine basis.
3.8. REFERENCES
Fite, EC; Turner, LW; Cook, NJ; Stunkard, C. (1988) Guidance document for conducting terrestrial field studies.
Hazard Evaluation Division Technical Guidance Document. Office of Pesticide Programs, U.S. EPA, Washington
DC. EPA 540/09-88-109.
3-10
-------
National Research Council. (1983) Risk assessment in the Federal Government. Washington, DC: National
Academy Press.
National Research Council. (1993) Issues in risk assessment. Washington, DC: National Academy Press.
Science Advisory Board, U.S. Environmental Protection Agency. (1990a) Reducing risk: setting priorities and
strategies for environmental protection. Washington, DC. SAB-EC-90-021.
Science Advisory Board, U.S. Environmental Protection Agency. (1990b) The report of the Ecology and Welfare
Subcommittee: Relative Risk Reduction Project, reducing risk. Appendix A. Washington, DC. SAB-EC-90-021A.
Science Advisory Board, U.S. Environmental Protection Agency. (1990c) The report of the Human Health
Subcommittee: Relative Risk Reduction Project, reducing risk. Appendix B. Washington, DC. SAB-EC-90-021B.
Science Advisory Board, U.S. Environmental Protection Agency. (1990d) The report of the Strategic Options
Subcommittee: Relative Risk Reduction Project, reducing risk. Appendix C. Washington, DC. SAB-EC-90-021C.
SETAC. (1994) Final report: Aquatic Risk Assessment and Mitigation Dialogue Group. Pensacola, FL: Society of
Environmental Toxicology and Chemistry, SETAC Foundation for Environmental Education, 220 p.
Thomas, LM. (1987) Environmental decision-making today. Environ Prot Agency J 13:2-5.
Touart, LW. (1988) Aquatic mesocosm tests to support pesticide registration. Hazard Evaluation Division Technical
Guidance Document. Office of Pesticide Programs, U.S. EPA, Washington, DC. EPA/540/09-88-035.
Touart, LW. (1995) The Federal Insecticide, Fungicide and Rodenticide Act. In: Fundamentals of aquatic
toxicology. Rand, GM, ed. Washington, DC: Taylor and Francis, pp. 657-668.
Touart, LW; Maciorbwski, AF. (1997) Information needs for pesticide registration in the United States. Ecol Appl
74:1086-1093.
Urban, DJ; Cook, JN. (1986) Hazard Evaluation Division standard evaluation procedure. Office of Pesticide
Programs, U.S. Environmental Protection Agency, Washington, DC. EPA/540/19-83-001.
U.S. Environmental Protection Agency. (1992) Framework for ecological risk assessment. Risk Assessment Forum,
Office of Research and Development, Washington, DC. EPA/600/R-92/001.
U.S. Environmental Protection Agency. (1993) A review of ecological assessment case studies from a risk
assessment perspective. Risk Assessment Forum, Office of Research and Development, U.S. EPA, Washington, DC.
EPA/630/R-92/005.
U.S. Environmental Protection Agency. (1994) The new generation of environmental protection. Washington, DC.
EPA/200/B-94-002.
U.S. Environmental Protection Agency. (1998, May 14) Guidelines for ecological risk assessment. Federal Register
63(93):26846-26924.
3-11
-------
-------
4. NOPrtNDIGENOUS SPECIES
4.1. SUMMARY
Traditionally, ecological risk assessment has been applied to chemical stressors. This
chapter shows how ecological risk assessment principles can be used to evaluate biological
stressors, including nonindigenous species and genetically engineered organisms. Biological
stressors are unique in their ability to reproduce, adapt to new environments, and evolve over
time. The concept of exposure for biological stressors includes evaluating potential entry sources
and pathways as well as describing their potential for colonization and spread. In general, the
case studies in this chapter follow the same logical sequences and steps outlined in EPA's
ecological risk assessment guidelines. However, in contrast to EPA's guidelines, which keep the
risk management and risk assessment processes separate (but coordinated), two of the case
studies in this chapter (black carp and pine shoot beetle) include risk management considerations
such as socioeconomic impacts as part of their risk assessment approach.
Black carp are native to eastern Asia and have been proposed for introduction into the
United States. The black carp case study illustrates a cost-benefit issue in which the potential for
positive gain from intentional introductions (biological control of yellow grub parasites in fish
ponds and of the zebra mussel in the wild) needs to be balanced with the potential for economic
and/or environmental damage resulting from establishment of the black carp in the wild. Risk
was estimated by an expert judgment process that combined estimation of the probability of
establishment (organism in entry pathway, entry potential, colonization potential, and spread
potential) with the consequences of establishment (environmental, economic, and perceived—
social and political). Qualitative risk rankings (high, medium, or low risk) were accompanied by
detailed descriptions of the rationale for each rating. An overall judgment of unacceptable risk
was assigned to uncontrolled releases of black carp.
The second case study evaluates risks associated with release of recombinant rhizobia at a
small-scale agricultural field site. This case was written at the request of EPA's Risk Assessment
Forum to test the utility of EPA's Framework for Ecological Risk Assessment (U.S. EPA, 1992)
with genetically engineered microorganisms. This case study was developed from a submission
received by EPA under the Toxic Substances Control Act (TSCA) premanufacture notification
(PMN) provision (see Chapter 2). In this case, there was concern over the possible effects that
might result from field testing of recombinant rhizobia (symbiotic bacteria) intended to increase
yields of alfalfa. Risks were characterized as low, primarily because off-site movement of the
recombinant rhizobia was considered very unlikely. One positive aspect of the case was the use
of postassessment monitoring to verify risk predictions.
4-1
-------
The third case study involves concerns over introduction of the pine shoot beetle into the
United States on imported logs. The pine shoot beetle can cause serious damage to the new
growth of healthy trees as well as to weak and dying ones. Scenario analysis was offered as a
methodology to evaluate the pathways and variables contributing to the pest risk and to identify
the options to mitigate the risk. An interesting facet of this case was the application of expert
i
opinion to quantify risks of different scenarios. The product of this effort provides a basis for
evaluating and adjusting risk management options and quarantine regulations.
Using risk assessment to evaluate nonindigenous species provides a framework for
placing available information into a format that can be used and understood by policy makers for
making risk management decisions. The major difficulty is the high uncertainty associated with
predicting the outcome of a nonindigenous species in a new environment, given the lack of
information on specific organisms and our current state of understanding on how an ecosystem
functions. Nevertheless, the degree of uncertainty surrounding the introduction of nonindigenous
!
organisms only increases the need for careful, unbiased risk assessments before making a
decision for or against an introduction.
4.2. INTRODUCTION
4.2.1. Definition and Scope of Risk Analyses
Humans have moved plants and animals from one ecosystem to another throughout
recorded history. Organisms moved outside their historic or natural geographic range are
i i • i
considered nonindigenous species. Within the United States, this includes species imported into
the country as well as those moved from one bioregion to another. In the United States,
nonindigenous species have been referred to as "exotics," "transplants," "nonnatives," or
"introduced species." In foreign countries, they often are called "alien" species.
In the United States alone, humans have intentionally or unintentionally introduced more
than 4,500 foreign species that have established and spread (OTA, 1993). Many introductions
have been viewed as providing economic and social benefits. However, the economic and
i ' i
environmental consequences of some introductions have been harmful, and in a few cases
catastrophic.
The definitions given below are generally consistent throughout risk analyses involving
nonindigenous species:
i • I '
• Nonindigenous species—the condition of a species being beyond its natural range or
natural zone of potential dispersal; includes all domesticated and feral species and all
hybrids except for naturally occurring crosses between indigenous species (OTA, 1993).
4-2
-------
• Pathway—any means by which nonindigenous species are transported.
Evaluations of nonindigenous species, independent of the method or process used,
generally contain one or more of the following components:
• Identification of one or more nonindigenous species of concern or the identification of a
pathway transporting or vectoring nonindigenous species of concern
• Determination of the likelihood that these nonindigenous species could become
established
• Determination of the impact if the nonindigenous species became established
• Determination of the available actions to reduce the risk that the nonindigenous species
will cause unacceptable damage.
4.2.2. Relationship to EPA's Ecological Risk Assessment Framework
This chapter discusses risk analyses that are directed toward evaluating and reducing the
negative impact from the establishment of "new" nonindigenous species. Risk analyses triggered
by nonindigenous organisms already established falls beyond the purview of this chapter and
more into the specific methods of pest control or the more general realm of ecosystem
management.
This chapter presents three case studies to illustrate the applicability and efficacy of risk
analysis as it applies to a range of nonindigenous species problems and issues. These case
studies were chosen because each represents different types of nonindigenous species (a fish
introduction, a genetically engineered bacteria, and a forestry beetle pest) and because each study
explores different types of risk evaluations (risk assessment focus, risk management focus, and
qualitative and quantitative evaluations). Only a summary of the risk analysis for each of the
case studies is presented in this chapter. Details on the risk processes and methodologies used
can be found in the original risk documents.
The main difference between physical/chemical ecological stressors and biological
stressors is that biological stressors are capable of reproducing. Equally important is the
characteristic of a biological organism to control its behavior so that it can adjust to or modify
the environment to fit its needs. In addition, a newly established population can, over successive
generations, change (evolve) to better adapt themselves to the new environment. These basic
characteristics of life add anew dimension of complexity and uncertainty that has little parallel
with risk analyses on nonliving ecological stressors. EPA's Guidelines for Ecological Risk
4-3
-------
Assessment (U.S. EPA, 1998) attempt to incorporate these biological characteristics and provide
guidelines for conducting risk assessments on nonindigenous species.
The case studies presented in this chapter did not intentionally follow the EPA process
(except for the EPA recombinant rhizobia assessment). Yet each case study follows the same
logical sequences and steps outlined in the 1998 EPA document and provides examples of how
analyses can be done on nonindigenous species.
i
4.2.3. Federal Agencies Involved in Nonindigenous Species Risk Issues
A number of Federal agencies are involved in issues surrounding nonindigenous species.
These include, but are not limited to, the U.S. Department of Agriculture, Animal and Plant
Health Inspection Service; U.S. Department of the Interior, Fish and Wildlife Service; Biological
Resource Division, U.S. Department of Commerce (NOAA); U.S. Department of Defense; EPA;
and NASA. A number of Federal and State agencies periodically or continually conduct
nonindigenous species risk assessments of varying levels of detail and sophistication for various
reasons in support of their primary missions.
Federal and State governments presently share responsibilities for issues concerning the
introductions of plants, animals, and their diseases. At present the Federal effort is primarily a
i
patchwork of laws, regulations, and policies scattered among several agencies. Most of these
policies address nonindigenous species peripherally; others focus more narrowly on specific
problems such as the introduction of crop pests. The need for a unifying national policy on
j
nonindigenous species is generally acknowledged. However, the development of such a policy is
impeded by historical divisions within and among government agencies arid pressure from
outside user groups and constituencies.
4.3. DISCUSSION ON THE STATE OF THE PRACTICE
The strength of using risk assessment to evaluate nonindigenous species is that it provides
a framework for taking the available information and placing it into a format that can be used and
understood by policy makers for making risk management decisions. The weakness of risk
analysis for nonindigenous species rests with the specific problems associated with predicting the
outcome of a newly established species in a new environment. The most serious problem is the
lack of information on specific organisms and our current state of understanding on how an
... : |
ecosystem functions.
Even complete life-history studies of a nonindigenous species do not guarantee that
managers can predict the impact that the species will have when introduced (although,
admittedly, good scientific information helps). The reason is that the complexity of the
4-4
-------
interaction between the organism and a new environment is so great that current predictive
models do not work with enough reasonable regularity to help decision makers. Indeed, there is
mounting evidence that normal linear predictive models rarely capture what occurs in a self-
actualized criticality or chaotic-based ecosystem.
It is important to note that the difficulty surrounding the evaluation of an exotic
biological stressor does not negate the need for management decisions to be made. It also is
important to realize that because information derived from scientific methods is probabilistic and
provisional, not absolute, we will never be free of uncertainty. The risk assessment, if properly
designed, should allow new and innovative predictive models to be incorporated. The degree of
uncertainty surrounding the introduction of nonindigenous organisms only increases the need for
careful, unbiased risk assessments before making a decision for or against an introduction. It is
imperative that a risk assessment honestly communicate its predictive limitations along with its
strengths to policy makers.
The connection between risk assessment and risk management must be present for the
risk assessment to be relevant to the needs of the risk managers. All three case studies in this
chapter showed how the risk assessment (assessors) can be connected to the risk managers. The
need for this type of initial bond (communication) between the assessors and the managers is
recommended in the final report of the Presidential/Congressional Commission on Risk
Assessment and Risk Management (1997).
4.4. CASE STUDIES
4.4.1. Risk Assessment on Black Carp (Pisces: Cyprinidae)
The black carp risk assessment (Nico and Williams, 1996) was initiated to test the
Generic Nonindigenous Aquatic Organisms Risk Analysis Review Process (RAM, 1996). This
"review process" was developed by the Risk Assessment and Management (RAM) Committee to
meet the risk analysis needs of the Aquatic Nuisance Prevention and Control Act of 1990. The
committee represented a number of government agencies, potentially impacted industries, and
special interest groups. The review process as a risk assessment tool was designed to evaluate
recently established nonindigenous organisms, evaluate nonindigenous organisms proposed for
deliberate introduction, and evaluate the risk associated 'with individual pathways (e.g., ballast
water, aquaculture, aquarium trade, fish stocking). As a risk management tool, the review
process was designed to reduce the probability of unintentional introductions and reduce the risk
associated with intentional introductions (RAM, 1996).
The nonindigenous species risk assessment process used by the review process is outlined
in Figure 4-1. The assessment is divided into probability of establishment and consequences of
4-5
-------
^>
*§
I
>«ft
R
^
• ^^
^
Sf5
^
§s
^
•^
•a
03
a*
r—(
3
O
ft,
^4
CO
•r-l
P^
TJ
^
T3
3
4-»
00
•a
5
1
§
O
a
S
ti
S
"CJ O
.S o
"
t»
08
& M
•S §
<
I §
II
13
••S a
o S
« TS
1
E
e
to
09
ett
fc
o
I
o
S
o
"&
o
a
I
w
S
2
.1
VO
S
13
4-6
-------
establishment. Biologic, economic, and other pertinent information is organized under the seven
elements. Each of the elements is assigned a risk rating and an uncertainty rating based on the
information gathered under the element. The ratings for each element are then combined to
provide an overall rating for the nonindigenous species being evaluated.
The black carp (Mylopharyngodon piceus) was chosen as a test organism for the review
process because it demonstrated: (1) a real issue in which the potential for positive gain
(biological control of yellow grub and zebra mussel) has to be balanced with the potential of
becoming established and causing economic and/or environmental damage in a new
environment; (2) a real issue in which political, economic, and environmental concerns were
already present (an assessment process must be able to withstand issues that are controversial);
and (3) a situation in which there still exists time to correctly manage this issue to the benefit of
the American people (the assessment would not have been done in vain).
The black carp is native to eastern Asia. Although it is one of several commercially
important carp species in China, some aspects of its natural history, such as details of its
reproduction in natural conditions, are poorly known. Most of the data on black carp natural
history are studies originally published in Russia and China.
Sections 4.3.1.1 and 4.3.1.2 give a summary of the black carp assessment, which follows
the risk model provided in Figure 4-1.
4.4.1.1. Probability of Establishment,
1. Estimate probability of the exotic organism being on, with, or in the pathway:
High—very certain. This species is already present in the United States. The pathway is
dependent on human transport.
2. Estimate probability of the organism surviving in transit: High—very certain. The
black carp is present, and survival in transit has been proved on at least several occasions.
3. Estimate probability of the organism successfully colonizing and maintaining a
population where introduced: Medium—reasonably certain. Appropriate habitats and climate
are found throughout most of the United States (i.e., large rivers and canals). Preferred food (i.e.,
aquatic snails and mussels) is locally abundant. The black carp became established after it was
introduced to several localities in Asia (e.g., Japan, possibly northern Vietnam), including at least
one water body in the former Soviet Union (Kara Kum Canal),. In addition, the grass carp, a
closely related species from Asia with similar spawning habitat requirements, has naturally
reproducing populations in open waters of the United States.
4. Estimate probability of the organism to spread beyond the colonized area:
High—reasonably certain. Appropriate habitats (i.e., large lowland rivers and canals) and
4-7
-------
climate are available throughout most of the United States. Preferred foods (i.e., aquatic snails
and mussels) are locally abundant in most U.S. rivers. The black carp is closely related to
another east Asian cyprinid, the grass carp (Ctenopharyngodon idella); the native distributions of
these two species are nearly identical, and their reproductive requirements appear to be very
similar. As such, if the black carp colonizes open water sites within the United States, the
species would likely spread beyond colonized areas, as has been the case with the grass carp.
The grass carp was first introduced into the United States (Alabama and Arkansas) in 1963 and
now occurs in more than 45 States.
Unless intentionally or incidentally spread into other areas by humans, black carp spread
in the United States would be expected to be limited to those river basins where introduced.
Major river basins in the United States that appear to provide appropriate habitat include the
Mississippi, the Snake, the Sacramento-San Joaquin, and the Colorado, among others. However,
if the black carp is salt tolerant, there is a risk that carp could spread along coastal waters into
adjacent basins or drainages. In a laboratory setting, the closely related grass carp has been
shown to survive up to 24 days in 10.5 parts per thousand salinities. Additionally, because of
their similarity in appearance to grass carp, there is potential that black carp will be incorrectly
identified as grass carp and be unintentionally introduced to some areas. Based on climate, black
! ! • I
carp might be expected to occur over at least most of the continental United States as well as
Hawaii.
4.4.1.2. Consequences of Establishment
5. Estimate economic impact if established: Low-^noderately certain. Possible costs
incurred from introducing black carp include: (1) reduction in the numbers and kinds of native
mussels (many of which at<3 important to the freshwater mussel industry); (2) competition with
native fishes; (3) competition with waterfowl and other vertebrates that utilize mollusks for food;
and (4) introduction of a probable carrier of parasites and diseases that utilize mussels as an
intermediate host (while the carrier frequently remains immune from the effects of the disease
itself).
The low rating is justified because none of the negative impacts described above would
strongly affect the U.S. economy. The domestic freshwater mussel indusuy is likely to be most
affected, but the extent of the damage is unclear. It is unlikely that black carp would be capable
j " ; j
of feeding on the adults of the majority of the species utilized by the mussel industry; however,
black carp would probably be able to take juveniles of these species.
6. Estimate environmental impact if established: High—very certain. It is highly likely
that the black carp would negatively impact native aquatic communities by feeding on and
4-8
-------
reducing populations of native mussels and snails. The black carp is known to feed on mussels
that are similar in shape and size to some native mussels of the United States. The United States
has a high diversity of gastropods and bivalves, and many of these are endemic to relatively
small regions of the country. For instance, the black carp would potentially threaten many of the
imperiled mussels currently on the brink of extinction. Of the 297 native freshwater mussels,
213 taxa (71.7%) are considered endangered, threatened., or of special concern. There also exists
potential that the black carp would directly compete for food (i.e., snails) with several other fish
species (e.g., certain catfishes, sunfishes, and suckers, freshwater drum), as well as certain birds
and mammals, including some native species listed as threatened or endangered. Because the
black carp shows a preference for snails as food, there is potential for impacting stream
communities where snails play an important role as a grazer of attached algae. Black carp may
directly and indirectly reduce aquatic insects.
7. Estimate impact from social and/or political influences: Medium—moderately certain.
Certain groups and industries in the United States support the introduction of the black carp,
including many fish farmers and also industries that have a problem with zebra mussels and
perceive black carp as a potential solution.
Those against introducing the black carp include various environmental groups and
persons involved with the mussel and freshwater pearl industries. The American Fisheries
Society recently passed a resolution asking governmental agencies to strictly prohibit the sale,
possession, and distribution of black carp, largely in part because of the potential to harm native
mussel fauna.
The overall rating of unacceptable risk was recommended by the assessor and agreed to
by the managers for uncontrolled releases of black carp. The assessors recommended that its
establishment in North American native aquatic environments should be prevented. However,
the assessors did conclude that sterile (triploid) black carp kept under a strict quality assurance
program could be used (introduced) for specific uses.
In this study, the risk communication between the risk managers and the risk assessors
was provided (as outlined in the review process) as specific management questions that were
submitted to the risk assessors before the assessment was started. These questions contained the
specific problems that the managers hoped the assessment would answer, but not what they
wished the outcome of the assessment to show. In this way the assessment was kept policy
relevant without becoming policy driven. Suggestions and processes for reducing the risk of
nonindigenous species (risk management) are covered in the review process but were beyond the
purview of the black carp test case.
4-9
-------
4.4.2. Risk Assessment for the Release of Recombmant Rhizobia a* a Small-Scale
Agricultural Field Site
The case study "Ecological Risk Assessment Case Study: Risk Assessment for the
Rdease of Recombinant Rhizobia at a Small-Scale Agricultural Field Site" (McClung and Sayre
1 993) was not Vfdttea in respome to ft ^.^ ^^ ^ ^ ^.^ ^ ^ ^^ ^ >
BPA s Risk Assessment Forum to test the utility of EPA's Framework for Ecological Risk
Assessment (U.S. EPA, 1992) with biological stxessors, specifically, genetically engineered
nucroorganism, Although the framework was written for chemical and physical stressors, the
Risk Assessment Forum was interested in identifying the shortcomings of the framework for
biological stressors.
Rhizobia, a general term for various species of the genus Rhizobium (and because of a
recent taxonomic revision, also Sinorhizobium), are gram-negative, motile, rod-shaped, aerobic
bacteria that infect legume roots, forming a symbiotic relationship with the plant. The bacteria
fix atmospheric nitrogen, providing an inorganic nitrogen form, ammonium, usable by the plant
^exchange for energy from the plant in the form of photosynthate, specifically dicarboxylates
The parental strains were modified by insertion of various genes, including antibiotic resistance
markers to allow for detection of these recombinants in the environment from indigenous
rhizobia, and nif genes to enhance the nitrogen fixation capability of the microorganism
In 1988 and 1989, EPA received voluntary premanufactoe notifications (PMNs) for
proposed small-scale field testing in 1989 of various recombinant strains afOMton meliloti
These mtergeneric" microorganisms were reviewed using typical procedures within the Office
of Pollution Prevention and Toxics' Biotechnology Program. For a PMN submission under
Section 5 of the Toxic Substances Control Act (TSCA) Section 5, an integrated risk assessment
is developed. Various assessments are written in support of a final risk assessment, including a
human health hazard assessment, an ecological hazard assessment, a construct analysis a
chemistry report, an engineering/worker exposure report, and an environmental exposure
assessment .The risk is evaluated using the typical EPA paradigm Risk = Hazard x Exposure
Since TSCA is a risk-versus-benefit statute, the benefits to society of use of a microorganism are
weighed into the final risk management decision. If a finding of "no unreasonable risk to hZ
health orthe environment" is made, then the Agencytakes no regulatory action. However if
there is sufficient infoimation to show an unreasonable risk, or if there is insufficient information
to determine that the risks are reasonable, then the Agency can prohibit or restrict the use of a
microorganism.
Since the case study followed the Risk Assessment Forum framework (U.S EPA 1992)
the summary that follows is given in the same general format.
4-10
-------
4.4.2.1. Problem Formulation
The case study focused on determining potential adverse effects of conducting small-
scale field trials in 1989 with recombinant rhizobia. The stressor was genetically engineered
rhizobial strains that could result hi two types of effects: biological effects (e.g., altered alfalfa
growth, altered growth of other nontarget legumes, displacement of indigenous species, gene
transfer) and chemical effects (production of toxins, detrimental metabolites, overproduction of
nitrate). Characterization of tihe recipient and donor microorganisms was a critical component
for the risk assessment.
The four recombinant rbizobia reviewed were strains made by the insertion of a gene
encoding for resistance to the antibiotics streptomycin and spectinomycin, which allowed the
recombinants to be differentiated from their parental strains hi both the laboratory and the
environment. Additional nif genes, which encode for the enzyme nitrogenase, also were inserted
into one strain to enhance the nitrogen fixation capacity of the, rhizobial strain.
The ecosystem potentially at risk was the surrounding agroecosystem in Dane County,
WI. Potential biotic components of the agroecosystem of concern were target and nontarget
legumes (including weedy crop legumes and noncrop legumes), native rhizobia, and bacterial
pathogens that could acquire the antibiotic resistance genes from the recombinant rhizobia. The
primary concern was the area surrounding the field plots, with lessening concern for areas farther
removed from the field site.
A number of assessment endpoints were identified: (1) decreased alfalfa growth, (2)
decreased growth of legumes outside the typical nodulation/cross-inoculation group, (3)
decreased growth of nonlegume crops, (4) unanticipated effects of introduced DNA sequences,
(5) effects of introduced DNA on recipient DNA at the insertion site, (6) unanticipated effects of
recipient microorganisms, (7) effects of antibiotic resistance genes, (8) competitive displacement
of native rhizobia if coupled with hazards listed in 1-3 or 9-10, (9) increased/decreased growth of
sweet clover, (10) increased/decreased growth of fenugreek, (11) effects of coumarin on cattle,
and (12) effects on the nitrogen cycle.
Predictive information on assessment endpoints 1 through 7 was used in the risk
assessment. The only assessment endpoints to be monitored during the field trials were 1 and
8—the effects on alfalfa yield and the competitiveness of the introduced recombinant strains for
alfalfa nodulation, respectively.
4.4.2.2. Analysis: Characterization of Exposure
The number of microorganisms to be released in the 1989 field trials was on the order of
1012 cells for each strain applied through in-furrow spraying. However, there was uncertainty
4-11
-------
regarding exposure over time because of the potential for microbial reproduction and transport.
Survival in the soil and root nodules, vertical and horizontal movement through soil, and aerial
dispersion warranted consideration. The ability to detect the recombinant microorganisms in the
environment created uncertainty because of the lack of sophisticated techniques available at the
time of the field studies. Laboratory studies showed a 1-log decline in numbers over a 4-week
; j i , j
period. However, literature data had shown extended survival of rhizobia in, soils, particularly in
the presence of a suitable host plant. The fact that alfalfa is a perennial crop suggested that the
recombinant rhizobia could potentially survive for years once released in the environment.
The field trials were intended for a 2-year period. Although literature data indicated
limited movement of rhizobia in soils, there was potential for aerial dispersion during inoculation
or through wind-blown soil particles, and there was also the potential for runoff from the field
test site given heavy precipitation. The presence of nontarget legumes in the 14-acre test site
area was evaluated before initiation of the field tests as another facet of the exposure
characterization. It was assumed that there would be limited off-site migration of the
recombinant rhizobia.
4.4.2.3. Analysis: Characterization of Ecological Effects
The primary effects data reviewed before the field test were greenhouse data on alfalfa
yield resulting from nodulation with the recombinant rhizobia. With one exception of increased
yield, greenhouse studies did not demonstrate any significant differences in nitrogen fixation as
measured through alfalfa top growth. However, the greenhouse studies were of questionable
utility. The data provided did not address ecological effects that would be of concern if there
were substantial movement of the microorganisms off-site, such as:
i
• Increased competitiveness
• Increased nitrogen and, therefore, nitrate production in soils
• Alteration of host range
• Effects on nonleguminous plants
• Effects on other legumes, sweet clover, and fenugreek, which are also known to be
nodulated by R. melilottal
• Spread of antibiotic resistance genes to other microbial populations.
4-12
-------
4.4.2.4. Risk Characterization
The risk of conducting the small-scale field testing with the recombinant rhizobia was
considered low. The case study did not evaluate several assessment endpoints because of the
limited likelihood of off-site dispersal of the microorganisms. Even if dispersal had occurred, the
small number of microorganisms applied may have precluded effective nodulation of other
legumes in the ecosystem of interest. The assessment did not address various large-scale effects,
such as effects on the nitrogen cycle or the spread of clinically important antibiotics, because this
was a small-scale field test site that was expected to remain small scale. Both effects and fate
data had elements of uncertainty resulting from the protocols used and extrapolation from
laboratory and greenhouse studies to field situations. There was no information on the effects of
the recombinant rhizobia on other legumes. Likewise, there were no data on the competitive
ability of these rhizobial strains compared with native rhizobia. The ability of the monitoring
techniques for the recombinant rhizobia also led to uncertainty.
4.4.2.5. Risk Verification
Data obtained from the small-scale field tests verified the risk assessment conducted for
this PMN submission. As expected from knowledge of rhizobial behavior and greenhouse data,
the recombinant rhizobia survived well in the rhizosphere of alfalfa plants. Field results did not
show any increased competitiveness of the recombinant rhizobia as evaluated by nodule
occupancy. As expected, there was little off-site movement of the strains observed in the various
dispersal studies. In addition, as predicted from laboratory and greenhouse studies, the construct
analysis, and the literature, there were no adverse effects on alfalfa growth with any of the
rhizobial strains tested, nor were there any significant differences in alfalfa growth between the
recombinant strains and their unmodified parental strains.
This case study followed the format of the ecological risk assessment framework, which
has been extensively peer reviewed and tested with chemical and physical stressors. Although
there were various shortcomings of the framework hi relation to biological stressors—such as the
lack of provision for survival, multiplication, and dispersal of both the microorganisms and the
introduced genetic material—this case study should serve as a useful model for assessing the risks
of future releases of genetically modified microorganisms. One strength of the case study was
the existing body of knowledge on the effects of previous rhizobial inoculations with naturally
occurring rhizobial strains. The practice of using rhizobial inoculants has a long history (nearly a
century) of safe use. Another attribute of this case study is the risk verification portion, whereby
it was possible to compare the outcome of the ecological risk assessment conducted using the
framework with the in-house risk assessment done for the PMN submission, and to have the risk
4-13
-------
assessments validated by the data and results obtained during the field trials. The field data
confirmed the predictions of the framework ecological risk assessment and the EPA PMN
submission risk assessment.
4.4.3. Scenario Analysis for the Risk of Pine Shoot Beetle Outbreaks
The pine shoot beetle was discovered in North America near Cleveland, OH, in July
1992. At the time of the analysis (January 1995), it was known to be established in six States:
Illinois, Indiana, Michigan, New York, Ohio, and Pennsylvania. It was expected to continue
spreading naturally.
The pine shoot beetle is the most destructive bark beetle (Scolytidae) of pines in Eurasia,
where it is a native pest. It can cause serious damage to the new growth of healthy trees as well
as to weak and dying ones. Healthy trees are at risk when populations of the beetle are high.
The beetle also may be an important vector of several diseases of pine. The current season's
growth (shoots) of many species of pine serve as the primary hosts for feeding by adult beetles,
while felled logs and downed trees are the primary breeding sites.
The pine shoot beetle has great potential to spread. Adults can fly 1 km, and the logs,
rough-cut lumber, nursery stock, Christmas trees, and decorative foliage they infest are often
transported long distances.
Infested counties were regulated under Federal and State quarantines. Logs of pine trees
i ' i
could be transported from infested counties to noninfested areas from July 1 through October 31
with no restriction (most beetles are assumed to be in the shoots during this time—normal
logging practice would remove all branches from the logs before moving to the sawmill). From
November 1 through June 30, logs must be fumigated or processed at the destination within 24 h
of harvest (beetles are overwintering at the base of the tree during this time; the assumption is
that beetles are destroyed during debarking/processing at the sawmill).
Early in 1994, the Michigan Department of Agriculture (MDA) proposed modifications
to the current regulatory regime. The APHIS of USDA rejected the proposal, resulting in a
continuing discourse with MDA, APHIS, and the USDA Forest Service about technical aspects
of the proposal, options, and risk.
An analysis document was developed by APHIS to provide an assessment of the risks so
that any decision regarding regulations will be sound. Scenario analysis was offered as a
methodology to evaluate the variables contributing to the pest risk and to identify the options to
mitigate the risk.
The product of this effort will provide the basis for evaluating and adjusting quarantine
regulations.
4-14
-------
4.4.3.1. Assessment Summary
A single expert meeting was organized to discuss scenarios and the evidence surrounding
each event component. Five outside experts representing a range of experience and perspectives
met with several APHIS staff members for 2Vz days of discussion. Technical background
information was provided to experts in advance of the meeting. The results (USD A, 1994) are
summarized herein.
Figure 4-2 describes the combination of scenarios determined to be the pathways for
possible new outbreaks. Each pathway was demonstrated individually with its respective data hi
the section of the document devoted to the analysis of the probability data. Probability estimates
were developed by the expert group for each event for each scenario. The products of point
estimates for each scenario have been calculated and added to the summary table. Evidence and
reference materials used or provided as the basis for estimates were listed in an appendix.
In each scenario, the most likely probability hi the sequence of events occurring was
represented by a point estimate (mode value) and surrounded by an estimate of the lowest and
highest probability in a triangular distribution. Experts were encouraged to estimate a range to
ensure that the actual probability lay within the area of the curve defined by the estimates. A
point estimate alone was used when the evidence indicated a very high degree of certainty.
Estimates and continuing calculations of probability were terminated when any event resulted hi
the elimination of the pest risk. The estimates were based solely on Michigan data; however, the
probability estimates developed from the data are believed to be generally representative of
locations in the North Central and Northeast United States above 40° north latitude.
By combining the curves for each event in a scenario pathway, an overall estimate of the
risk and associated uncertainty was developed for scenarios describing the situations) as they
would be without the addition of mitigation. This facilitated the identification of high-risk
scenarios and events. It also provided the background for evaluating the application and value of
mitigation schemes applied to specific scenarios and events.
Each scenario (A, B, C, and E>) was analyzed according to seasons corresponding with the
insect's activities (summer, fall, whiter, early spring, and late spring). This creates a total of 20
subscenarios. However, the summer subscenarios were determined by experts to have a
negligible risk after the first event because insects would be feeding hi shoots and therefore
would not be associated with delimbed logs during this period. Eliminating the summer
subscenarios brings the total number of subscenarios to 16.
4-15
-------
frequency
in
units per
season*
Combined Scenarios for New Outbreaks
of Pine Shoot Beetle
Due to the Movement of Logs
one unit is
Infested
I beetles
disperse In
transit
beetles are
-M able to
I colonize I
Scenario A
beetles disperse
at mill before
processing
unit is chipped
or debarked
no
beetles
survive
slabbing
r
beetles
->| disperse from
I products
yes
I beetles survive
debarking or
chipping
beetles
disperse from
mill products
*one unit = 100 trees
beetles are
able to
colonize
Sconario B
beetles are
able to
colonize
Scenario C
beetles are
-M able to
I colonize J
Scenario D
Figure 4-2. Combined scenarios for new outbreaks of pine shoot beetle due to the
movement of logs.
4-16
-------
Computer simulation using specially designed software (@ Risk by Palisade Software,
Inc.) was used to graphically represent the distributions for each event and to calculate the
product of all events for a scenario. Two types of curves were generated using Latin hypercube
sampling and 3,000 to 9,000 iterations (trials with random numbers). One curve is roughly "bell-
shaped" and demonstrates the distribution of probability across the range of values defined by the
experts. The other curve is S-shaped and demonstrates the cumulative probability from 0% to
100%. The endpoints—the frequency of outbreaks by season and year—are given in Table 4-1 for
the 16 subscenarios.
4.4.3.2. Risk Management Summary
After the expert group completed the assessment, a list of 10 potential risk management
options was developed. The list included options such as having no restrictions on logs without
bark to allowing movement from regulated areas after fumigation to allowing the movement after
certain dates. Then the group discussed a list of potential treatment measures. These included
fumigation, debarking, insecticide spray, total tree utilization, ;butt-cutting, high stumping, and
others. Finally, fumigation options were identified that would allow the safe movement of logs
and lumber with bark from the regulated area for four time periods. The options included
fumigation, movement to an approved facility, and harvesting tune limitations. Restrictions on
logs moved between February 16 and June 30 (early and late spring periods) were the most
restrictive. Restrictions on logs moved between July 1 and September 30 were the least
restrictive. This corresponded with the risk identified in the assessment section. The following
advantages were identified in the process used for the pine shoot beetle assessment:
• The use of scenario trees aids the assessor and readers hi identifying and understanding
the events within a pathway mat lead to an unwaated consequence.
• Convening an expert group meeting for the development of estimates for determining risk
was beneficial. The method used here was a modification of the expert information
approach developed by Stan Kaplan (1992). Statistical and nonstatistical information
relevant to the parameter was reviewed and discussedjby the participants. Then low,
high, and point (most likely) estimates were established for each event. Thus, the
uncertainties of the estimates are captured in the curve (probability distribution)
developed by the group. A high narrow curve indicates a large degree of certainty
(confidence); a wide low curve indicates a lack of confidence.
4-17
-------
Table 4-1. Frequency of outbreaks of the pine shoot beetle and years
between outbreaks
Dispersal from:
Scenario A
Transit
Scenario B
Mill
Scenario C
Slabwood
Scenario D
Mill by-products
All scenarios by
season
mean
mode
95% limit
mean
mode
95% limit
mean
mode
95% limit
mean
mode
95% limit
mean
mode
95% limit
New outbreaks per year
0.00369
0.000905
0.00949
0.00379
0.000698
0.00995
4.63
0.884
11.8
0.0000103
0.00000125
0.0000302
4.64
0.885
11.8
Years between outbreaks
271
1,110
105
264
1,430
101
0.22
1.13
0.08
97,300
802,000
33,100
0.216
1.13
0.0846
I II
The resulting probability distribution from the calculations documented the amount of
certainty in a process matter. In addition, the quantitative process allowed the risk
managers to understand the great differences in risk between a low-risk and high-risk
scenario. For example, 1 of the 16 scenarios represented more than 65% of the risk while
the 12 lowest risk scenarios combined represented only 0.2% of the risk. Giving the one
scenario a qualitative value of high risk and the 12 others a value of low risk would not
convey the differences in magnitude.
4.5. NEXT STEPS
These recommendations are intended for the Federal and State agencies that periodically
or continually conduct nonindigenous species risk assessments to support their primary missions.
• Improve the science surrounding nonindigenous species to achieve a better understanding
of why some species are more likely to establish and become pests.
4-18
-------
Conduct retrospective evaluations of nonindigenous organism risk assessments to identify
opportunities for improvement, of existing processes and methods and to help determine
which approaches work the best under what circumstances.
Improve Federal interagency cooperation to help reduce redundancy and focus limited
resources. :
Enhance international cooperation. Global management strategies will be necessary to
address many nonindigenous species problems. Participation in existing international
organizations for plant and animal protection, environmental protection, and ballast water
management should be encouraged, and new opportunities for cooperation should be
pursued.
Improve awareness of nonindlgenous species issiies by the public and potential
stakeholders through communication and education. Ensure that interested parties and
concerned individuals are involved in risk assessment planning and in the management of
nonindigenous species.
4.6. REFERENCES
Kaplan, S. (1992) "Expert information" versus "expert opinions." Another approach to the problem of
eliciting/combining/using expert opinion in PRA. J Reliability Syst Saf 35:61-72.
McClung, G; Sayre, PG. (1993) Ecological risk assessment case study: risk assessment for the release of
recombinant rhizobia at a small-scale agricultural field site. In: A review of ecological assessment case studies from
a risk assessment perspective: Risk Assessment Forum, U.S. Environmental Protection Agency, 1993, Washington,
DC. EPA 630/R-92/005.
Nico, LG; Williams, JD. (1996) Risk assessment on black carp (Pisces: Cyrinidae). Final draft. A report to the Risk
Assessment and Management Committee of the Aquatic Nuisance Species Task Force. 59 pp.
Office of Technology Assessment. (1993) Harmful non-indigenous species in the United States. OTA-F-565, U.S.
Congress, Office of Technology Assessment Washington, DC: U.S. Government Printing Office.
The Presidential/Congressional Commission on Risk Assessment and Risk Management. (1997) Risk assessment
and risk management in regulatory decision-making. Vol. II, final report.; The Presidential/Congressional
Commission on Risk Assessment and Risk Management, Washington, DC. GPO #055-000-00568-1.
Risk Assessment and Management Committee. (1996) Generic nonindigenous aquatic organisms risk analysis
review process. Report to the Aquatic Nuisance Species Task Force.
U.S. Department of Agriculture. (1994) Scenario analysis for the risk of pine shoot beetle outbreaks resulting from
the movement of pine logs from regulated areas. Unpublished report. Animal and Plant Health Inspection Service,
Beltsville, MD.
4-19
-------
U.S. Environmental Protection Agency. (1992) Framework for ecological risk assessment. Risk Assessment Forum,
Office of Research and Development, Washington, DC. EPA/630/R-92/001.
U.S. Environmental Protection Agency. (1998, May 14) Guidelines for ecological risk assessment. Federal Register
63(93):26846-26924.
4-20
-------
5. CERCLA
5.1. SUMMARY
Ecological risk assessments under the Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA) are retrospective evaluations of the effects of
contamination in a given area. They provide baseline information on whether a clean-up should
be considered for ecological reasons, and risk assessments are used in the evaluation of remedial
alternatives. Each of the three case studies follows the EPA framework, and they present and
discuss the three phases of ecological risk assessment.
Three case studies are presented in this chapter. The Linden Chemicals and Plastics site
in Georgia was contaminated with mercury and polychlorinated biphenyls (PCBs). The site was
evaluated for its impact on selected mammals and birds exposed to the contaminants through a
saltmarsh. Tissue samples were collected from selected specimens at the contaminated site and a
reference site. A food web approach was used and compared with toxicity values found in the
literature. Substantial threat concentrations and potential risk concentrations were identified.
The United Heckathorn site presents an assessment of DDT contamination of a section of
San Francisco Bay. Benthic community structure, fish tissue levels, sediment toxicity tests, and
food web models were conducted in the analysis phase of the assessment. Areas of sediment
were identified for remediation to reduce the risk to birds and fish to acceptable levels (i.e., bulk
sediment concentrations of 1.9 mg/kg DDT at 1.9% total organic carbon).
The Metal Bank of America site was located on the Delaware River, and PCBs were the
primary contaminant. The protection of the shortnose sturgeon, a designated endangered species,
and other fish species were assessment endpoints. Tissue levels of representative fish collected
at the site were compared with literature values. The risk characterization determined that there
was potential risk to fish reproduction.
Risk managers are required to protect human health and the environment and to comply
with applicable, relevant, and appropriate requirements. They also balance the risk and proposed
mitigation methods with various economic, societal, technical;, and political concerns discussed
in this chapter.
5.2. INTRODUCTION
The fundamental purpose of performing an ecological risk assessment at Superfund sites
is to determine if releases or potential releases of hazardous substances from the site have
resulted in or are likely to result in unacceptable adverse effects on ecological receptors. The
goal of Superfund response actions is to prevent effects from Occurring or, if effects have
5-1
-------
occurred, to implement a remedy that will provide adequate protection of the ecological
resources in a cost-effective manner that also meets any appropriate Federal or State laws.
Ecological risk assessment data are used for:
• Characterizing baseline risk to determine whether a cleanup should be considered,
:
• Deriving specific contaminant concentrations that provide adequate protection from
unacceptable risks,
• Evaluating the remedial alternatives for potential effectiveness and potential risks, and
• Providing baseline information that can be followed with monitoring to document that the
I
remedy is effective at reducing risk.
I
Table 5-1 presents the use of ecological data in Records of Decision (RODs) in 1995.
•i • ! I . < • I
According to Section 104(a)(l) of the Comprehensive Environmental Response, Compensation,
and Liability Act (CERCLA), whenever there is a release or a substantial threat of a release of a
hazardous substance into the environment, EPA is authorized to take whatever action is deemed
appropriate to protect the environment, as long as the action is consistent with the National
Contingency Plan (NCP). EPA uses the information from the risk assessment in its decision-
making process.
The NCP requires that a baseline risk assessment be conducted by the lead agency during
the remedial investigation/feasibility study in order to "characterize the current and potential
threats to human health and the environment" (Section 300.430[d][4]). Any remedy selected by
EPA must be protective of the environment and human health. It must also comply with any
enforceable Federal or State standards or criteria that apply to the site. In addition, the NCP
requires that certain balancing criteria be considered: (1) long-term effectiveness and
permanence of the response; (2) reduction of toxicity, mobility, or volume of the waste through
treatment; (3) short-term effectiveness; (4) implementability; and (5) cost. Two modifying
criteria also must be considered: State acceptance and community acceptance.
Section 300.430(e)(9)(iii)(A) of the NCP states, "Alternatives shall be assessed to
determine whether they can adequately protect human health and the environment, in both the
: i i
short- and long-term, from unacceptable risks posed by hazardous substances."
5-2
-------
Table 5-1. Use of ecological data in Records of Decision (RODs) in 1995
Explanation
Number of RODs with ecological risk
assessments
Number of RODs where remedial action
based at least partially on ecological risk
Number of RODs where population/
community study performed
Number of RODs where modeling
performed
Number of RODs where literature values
used
Number of RODs where ambient water
criteria used
Number of RODs where NOAAa sediment
values used
Number of RODs where site-specific
toxicity tests performed
Number of RODs where tissue sampling
performed
Total number
113
52 :
24
25 .
50
19
11
14
1
10
Percentage of
total RODs
60%
46%
21%
22%
44%
17%
10%
12%
9%
*NOAA = National Oceanic and Atmospheric Administration.
Risk assessments should be designed to determine a threshold media concentration that
will provide adequate protection of important ecological resources. This requires substantial
discussion between the risk assessor and the risk manager before sampling to make sure the
information needed to make these decisions is collected. Many remedial alternatives have short-
term adverse consequences for the environment because of resulting physical disruption of the
ecosystem. .
5.3. RISK MANAGEMENT
The Superfund program currently has few written policies or guidances that explicitly
explain how to make reasoned ecological risk management decisions. Often the decision maker
5-3
-------
must rely on the guidance given in the NCP. Unlike human health risk assessments, which have
quantifiable risk goals that define levels of acceptable risk to one species (e.g., to reduce human
cancer risks to levels below 1 in 10,000), quantifiable ecological risk assessment, goals have not
been established by the Agency. The NCP states only, "Alternatives shall be assessed to
determine whether they can adequately protect human health and the environment, in both the
short- and long-term, from unacceptable risks posed by hazardous substances" (Section
300.430[e][9][iii][A]). This lack of a simple and easily articulated cleanup goal makes the
selection of an appropriate remedy that is protective of the environment and meets the other eight
NCP criteria problematic. In the Agency's recent Five-Year Strategic Plan, the Administrator
stated that one of EPA's goals is to achieve "healthy, sustainable ecosystems" (U.S. EPA, 1994).
Superfund risk managers and risk assessors should work together to translate this overarching
goal into site-specific goals and objectives.
Superfund decision makers must consider nine criteria when selecting a response action
that is appropriate for the site:
• Overall protection of human health and the environment
• Compliance with applicable, relevant, and appropriate requirements (ARARs)
• Long-term effectiveness and permanence
• Reduction of toxicity, mobility, or volume through treatment
• Short-term effectiveness
• Implementability
Cost
• State acceptance
• Community acceptance.
The first two criteria are thresholds that must be met at every site (though the ARARs can
be waived under certain circumstances), the next five are balancing criteria, and the last two are
modifying criteria. The three criteria usually most important to the ecological risk manager are
protection, long-term effectiveness, and short-term effectiveness.
5-4
-------
5.4. CASE STUDIES AND EXAMPLES
5.4.1. Linden Chemicals and Plastics Wildlife Assessment
The Linden Chemicals and Plastics (LCP) site is located in Brunswick, GA. Among
other operations, LCP operated a chlor-alkali plant from 1955 through the closing of the facility.
Graphite electrodes were impregnated with polychlormated biphenyls (PCBs) (specifically
Aroclor 1268). The site is adjacent to a saltmarsh system that encompasses 550 acres.
To support EPA Removal Program objectives, the LCP wildlife assessment conformed to
the EPA process for designing and conducting ecological risk assessments (U.S. EPA, 1997).
Bulk chemistry, toxicity tests, population and community evaluations, and contaminant
accumulation data were used in exposure models to evaluate ecological risks at this site.
Sampling locations were based on the ability to collect not only target organisms but also the
organisms at targeted contaminant exposure levels.
5.4.1.1. Problem Formulation
Information collected at the site indicated that PCBs and base, neutral, and acid-
extractable compounds and metals (particularly mercury and lead) were the contaminants of
concern (COCs). The concentrations of these compounds were compared with benchmark
criteria (i.e., no-effect levels) to determine if further investigation was necessary. This procedure
is defined as a preliminary risk assessment. Any contaminant in which the resultant quotient was
less than 1 was discontinued from further review. If the quotient was greater than 1 (indicating a
potential for risk), the contaminant was retained for further review and evaluated using food
chain accumulation models.
Although multiple COCs were identified, this case study addresses only mercury and
PCBs. To determine the effects of these contaminants on biota, it is necessary to understand the
mechanisms of toxicity of the chemicals and the systems that they affect.
5.4.1.2. Hazard Characterization
The objective of an exposure assessment is to determine the pathways and media through
which receptors may be exposed to site contaminants. Exposure pathways are dependent on the
habitats and receptors present on site, the extent and magnitude of contamination, and the
environmental fate and transport of the COCs.
COCs present in forage and prey species could cause toxicity via ingestion in higher
trophic level organisms. In addition to exposure via consumption of contaminated forage,
ecological receptors may be exposed through incidental water and soil/sediment ingestion or
5-5
-------
through direct contact. The exposure pathways that were evaluated in this risk assessment were
the ingestion of prey, the incidental ingestion of soil/sediment, and direct contact.
5.4.1.3. Assessment Endpoints and Testable Hypotheses
Three assessment endpoints selected for evaluation were:
» Protection of long-term health and reproductive capacity of omnivorous mammal species
that utilize the marsh.
• Protection of long-term health and reproductive capacity of piscivorous (marine and
terrestrial) mammal species that utilize the marsh/river system.
• Protection of long-term health and reproductive capacity of avian species that utilize the
marsh and Purvis Creek.
The specific risk questions based on the assessment endpoints were as follows:
• Are levels of site contaminants in water, sediment, and biota sufficient to result in a dose
that could cause adverse effects on the long-term health and/or recruitment of omnivorous
mammal species that utilize the marsh?
• Are levels of site contaminants in water, sediment, and biota sufficient to result in a dose
that could cause adverse effects on the long-term health and/or recruitment of marine or
terrestrial piscivorous mammal species that utilize the marsh/river system?
• Are levels of site contaminants in water, sediment, and biota sufficient to result in a dose
that could cause adverse effects on the long-term health and/or recruitment of passerine
birds that utilize the marsh and Purvis Creek?
• Are levels of site contaminants in water, sediment, and biota sufficient to result in a dose
that could cause adverse effects on the long-term health and/or recruitment of
piscivorous/benthic organism-feeding birds that utilize the marsh and Purvis Creek?
5.4.1.4. Conceptual Model
The conceptual model was designed to determine the following: source release to marsh
sediment and water; exposure of forage to contaminated water and sediment; and exposure of the
assessment endpoints through ingestion of contaminated forage, incidental ingestion of
contaminated sediment, and ingestion of contaminated water. The otter was selected as an
endpoint because it ingests aquatic biota, sediment, and water; the raccoon was selected because
it ingests aquatic biota and water; and the clapper rail (an avian species that utilizes the marsh
and Purvis Creek) was selected because it ingests aquatic biota, sediment, and water.
5-6
-------
The protection of long-term health and reproductive capacity of mammal species that use
the marsh was determined to be an assessment endpoint. Food chain accumulation models with
hazard quotient (HQ) evaluations were selected to evaluate risk to mammals that use the marsh.
The otter was selected as a measurement endpoint for piscivorous mammals and the raccoon as a
measurement endpoint for omnivorous mammals. Appropriate forage species were identified for
the above receptors, collected, and analyzed. Exposure of receptors to contamuiants was
quantified and compared with existing toxicity data for these species.
The protection of long-term health and reproductive capacity of avian species that utilize
the marsh and Purvis Creek was also determined to be an assessment endpoint. Food chain
accumulation models with HQ evaluations were selected to evaluate risk to avian species. The
clapper rail was selected as a measurement endpoint for wading waterfowl. Clapper rails and
appropriate forage species were collected for the exposure model and analyzed. Dietary
exposure of receptors to contaminants was calculated and compared with existing toxicity data
for avian species.
5.4.1.5. Food Chain Model Assumptions
This portion of the assessment concentrated on exposure to mercury and PCBs through
food ingestion. The body burden concentration of mercury and PCBs in prey items collected at
the site was used to evaluate exposures to receptor species.
The risk characterization was initiated by evaluating each of the measurement endpoints.
For the assessment endpoints that had multiple measurement endpoints, an overall risk
conclusion was determined by reviewing the multiple lines of evidence (referred to as a weight-
of-evidence approach).
Three ecotoxicological benchmarks were used:
• No-observed-apparent-effects level (NOAEL)
• Low-observed-adverse-effects level (LOAEL)
• Acute benchmark, which is used to evaluate imminent ecological threats.
Because mercury is a reproductive, behavioral, and developmental toxin, mortality can
occur depending on the form of mercury and the degree of exposure. The rate of mercury
speciation and chemical conversion also may determine the toxicity. Conservative assumptions
were made on the proportion of organomercury versus inorganic mercury.
5-7
-------
5.4.1.6. Sources of Uncertainty
Identification of the sources and nature of uncertainty for an ecological risk assessment is
critical for the appropriate utilization of the risk assessment in risk management decisions.
Identifying uncertainty allows for certain decisions to be made confidently; that is, the risk
assessment can confidently identify where there is no substantial ecological risk. However, there
may be uncertainty as to what level of contamination would actually result in an adverse
response. Risk calculations were based on conservative life-history values (e.g., the lowest body
weight and the highest ingestion rates). The benchmarks (NOAEL, LOAEL, and acute) used to
determine HQs were also the lowest values found in the literature. While there is uncertainty
associated with each benchmark, a consistent process for selection has been used in its selection.
5.4.1.7. Clapper Rail Tissue Evaluation
In July 1995, seven clapper rails were collected from the south marsh, and in August
1995, seven clapper rails were collected from the reference area. Table 5-2 presents the results of
the evaluations. Mean mercury concentrations in the liver tissue of birds collected on the site
were 15.7 mg/kg versus 3.5 mg/kg from birds from the reference area.
Acute mortality was found to be associated with liver mercury concentrations ranging
from 4.6 mg/kg to 91 mg/kg wet weight in white-tailed eagles (Haliaaetus albicilld) (Henriksson
et al., 1966; Koeman et al., 1972; Oehme, 1981; Falandysz, 1986; and Falandysz et al., 1988).
Captive-raised grackles (Quiscalus guiscula) displayed mortality at 54.5 mg/kg wet weight in
Table 5-2. Clapper rail mercury and PCB tissue levels
Mean from site, mg/kg
(dry weight)
Reference area,
mg/kg (dry weight)
Mercury tissue levels
Breast muscle
Liver
Remaining carcass
Feathers
PCB tissue levels
Breast muscle
Liver
Remaining carcass
5.1
15.7
5.1
11.3
9.2
212.0
27.8
1.6
3.5
1.1
3.6
0.8
0.8
1.8
5-8
,1 „:
-------
liver, whereas red-winged blackbirds displayed mortality at mercury concentrations in liver of
126.5 mg/kg wet weight (Finley et al., 1979).
5.4.1.8. Hazard Quotient Results
The HQ calculations incorporate the life-history mformation on the modeled species.
The species used for the HQ calculations were selected as conservative representatives of a
trophic level/food chain exposure pathway related to the assessment endpoints.
5.4.1.8.1. Raccoon. For mercury, the raccoon food web model predicts an acute threat at an
exposure point concentration of 90 mg/kg (HQ = 1.1). When a.LOAEL toxicity benchmark is
used, the model predicts a threat of adverse responses at a sediment concentration of 15 mg/kg
(HQ = 6.6). For PCBs, the raccoon model predicts an acute threat at a sediment exposure
concentration of 70 mg/kg (HQ = 1.0). When a LOAEL toxicity benchmark is used, the model
predicts the threat of an adverse response at a sediment concentration of 2.3 mg/kg (HQ = 1.0).
5.4.1.8.2. Otter. For mercury, the otter model predicts an. acute threat at an exposure point
concentration of 90 mg/kg (HQ = 1.4). When a LOAEL toxicity benchmark is used, the model
predicts a threat of an adverse response at a concentration of 15 mg/kg (HQ = 6.6). For PCBs,
the otter model predicts a threat of acute toxicity at an exposure point concentration of 70 mg/kg
(HQ = 2.5). When a LOAEL toxicity benchmark is used, the otter model predicts the threat of
adverse responses at a sediment concentration of 5.2 mg/kg (HQ = 1.1).
5.4.1.8.3. Clapper rail. For mercury, the clapper rail model does not suggest Hie threat of acute
adverse responses at an exposure point concentration of 150 mg/kg. When a LOAEL toxicity
benchmark is used, the model suggests that there is a threat of adverse responses above 15 mg/kg
(HQ = 3.2). For PCBs, the clapper rail model suggests that there is neither an acute threat nor a
LOAEL-based threat of adverse response at an exposure point concentration of 150 mg/kg.
5.4.1.9. Risk Assessment Conclusions
5.4.1.9.1. Protection of long-term health and reproductive capacity of omnivorous mammal
species that utilize the marsh. Based on HQ calculations and LOAEL benchmarks, there is
imminent and substantial threat at exposure point concentrations of 90 mg/kg mercury and/or
70 mg/kg PCBs. Potential risk exists at levels at least as low as 15 mg/kg mercury and 2 mg/kg
PCBs.
5-9
-------
5.4.1.9.2. Protection of long-term health and reproductive capacity of piscivorous mammal
species that utilize the marsh/river system (both marine mammals and terrestrial mammals).
Based on HQ concentrations and LOAEL benchmarks, there is imminent and substantial threat at
exposure point concentrations of 30 mg/kg mercury and/or 66 mg/kg PCBs. Potential risk exists
at levels as low as 2 mg/kg mercury and 5 mg/kg PCBs.
5.4.1.9.3. Protection of long-term health and reproductive capacity ofavian species that
utilize the marsh and Purvis Creek. Food chain exposure models using HQ calculations
indicate that there is a substantial and imminent threat due to sediment mercury concentrations of
34 mg/kg and sediment PCB concentrations of 56 mg/kg. LOAEL benchmarks indicate a risk at
90 mg/kg mercury and no potential risk based on PCB exposure. A comparison of body burden
levels in clapper rails to literature values indicates that there is no risk due to mercury; however,
there is substantial risk due to PCBs. In conclusion, based on the food chain accumulation
models for clapper rail, it appears that there is imminent and substantial threat to at least one
species at exposure point concentrations of 56 mg/kg PCBs and 34 mg/kg mercury (based on the
food chain accumulation models calculated for clapper rails). In addition, the LOAEL
benchmarks indicate that potential risk exists at 90 mg/kg mercury and that no potential risk is
associated with PCBs.
!
5.4.2. United Heckathorn Assessment
5.4.2.1. Site History and Background
The United Heckathorn site has been a major source of DDT in San Francisco Bay since
1947, when a pesticide blending and packaging plant began operations. Although the pesticide
blending and packaging operations ended in 1966, DDT accumulations in mussels near the site
remain among the highest detected in the California Mussel Watch program. The site is located
on the eastern shoreline of the central bay hi the city of Richmond and includes Lauritzen
Channel, Parr Canal, the Santa Fe Channel, and Richmond Inner Harbor. Sediments in the
channels, harbor, and soils around the facility are contaminated with DDT, dieldrin, and other
persistent chlorinated pesticides. An ecological risk assessment was completed for the site in
1994 by EPA's Environmental Research Laboratory in Newport, OR (Lee et al., 1994).
5.4.2.2. Problem Formulation and Conceptual Model
Central San Francisco Bay provides habitat for many birds, fishes, arid invertebrates.
Aquatic habitats closest to the site include areas of soft bottom with armored shoreline used by
anchovy, surfperch, starry flounder and English sole, herring, gobies, and other marine fish.
5-10
-------
Brooks Island lies at the southern end of the inner harbor, which is vegetated and surrounded by
mudflats and patches of eelgrass that are used by Pacific herring. The open water channels near
the site also are used by marine birds and harbor seals.
Contaminants of greatest concern include dieldrin and the DDT metabolites, which are
both readily adsorbed to sediment particles. The loading of pesticides into vessels adjacent to the
site resulted in direct discharge to the channels, where sediments are now highly contaminated.
The pesticides also are present in surface water and pore water and are accumulating hi biota at
the site. DDT is associated with reproductive impacts in fish-eating birds. DDT metabolites and
dieldrin also can be directly toxic to fish and invertebrates at low concentrations. DDT residues
in fish have been associated with reproductive problems such as early life-stage mortality.
Although not explicitly stated as such in the risk assessment, the assessment endpoints
evaluated included the following:
• Protection of the benthic community from direct toxic effects
• Protection of other aquatic species from direct toxic effects
• Protection of birds from reproductive effects after food chain transfer
• Protection of fish from reproductive effects
• Ensuring that concentrations in edible species do not exceed thresholds for human health
concerns.
The risk assessment utilized a thorough suite of measurements to evaluate the assessment
endpoints (Table 5-3). Sediment sampling formed the foundation for the assessment. Sediment
grab samples were collected from a total of 20 stations at the site. Samples for chemical
analysis, benthic community evaluation, toxicity testing, interstitial water chemistry, and
laboratory bioaccumulation testing all were taken from the same grab. Surface water, fish, crabs,
shrimp, and benthic invertebrates from the site also were analyzed for chemical residues.
5.4.2.3. Risk Characterization
A detailed exposure evaluation was conducted using measurements from the site and
equilibrium partitioning theory to support food web modeling and toxicity evaluations.
Contaminants associated with three phases of the sediment matrix were examined (particles,
freely dissolved, and associated with dissolved organic matter). Chronic and acute ambient
water quality criteria values were exceeded in interstitial water at many of the stations.
Organisms sampled near the site contained elevated concentrations of DDT metabolites and
5-11
-------
Table 5-3. Measurement endpoints and approach
Assessment endpoint
Protection of the benthic community
from direct toxic effects
Measurement endpoints and approach
Protection of other aquatic species from
direct toxic effects
Protection of birds from reproductive
effects after food chain transfer
Protection offish from reproductive
effects
Ensuring that concentrations in edible
species do not exceed thresholds for
human health concerns
Benthic community structure compared to reference site;
correlations with sediment and interstitial water
chemistry
Acute amphipod sediment toxicity test (survival);
correlations with sediment and interstitial water
chemistry; comparisons with reference site and historical
information
Chronic juvenile bivalve sediment toxicity test (growth);
correlations with sediment and interstitial water
chemistry
i
Chronic bivalve laboratory sediment toxicity and
bioaccumulation test
Water concentrations compared with literature effects
thresholds and AWQCa; modeling from sediment to
water
Prey concentrations; food web modeling from sediment
through prey to receptors
Fish tissue concentrations at the site compared with
literature effects thresholds; correlations with water,
sediment, or prey concentrations; food web modeling
from sediment through prey
Fish tissue concentrations at the site compared with FDA
and cancer thresholds; correlations with water, sediment,
or prey concentrations
"AWQC = ambient water quality criteria.
dieldrin in tissues; for sessile organisms, the concentrations correlated with sediment
concentrations. DDT concentrations in shiner surfperch and bay goby were especially elevated
and exceeded the Food and Drug Administration (FDA) action levels.
Benthic community evaluations indicated that increasing concentrations of DDT in
sediment are associated with a reduction in the number of amphipods (especially after excluding
one amphipod species that appeared to be more tolerant), and also was associated with an altered
Infaunal Index. Ten-day sediment toxicity tests using Eohaustorius estuarius indicated that
5-12
-------
sediments near the site are significantly toxic to amphipods and that there is a gradient of toxicity
away from the site. A toxic unit approach was used to evaluate the contribution of various
contaminants present in the samples.
Food web modeling indicated that sediments appear to be a significant source of
contaminants to birds and that the birds would be at risk, based on comparisons with literature
effects thresholds. Fish-eating birds would need to feed exclusively near the site for 2 months
each year to exceed risk standards. Risk was evaluated on a comparative basis between channels,
with the channel nearest the site posing the greatest risk.
5.4.2.4. Conclusions
The risk assessment report concluded that the greatest risk was due to DDT compounds
present in sediment nearest the site. The Lauritzen Channel was identified as a major
contamination source, with tidal action transporting contaminated sediment and water away from
the area. Organisms near the site are exposed to and accumulating high levels of DDT
compounds.
A food web model was used to evaluate which areas of the site would need to be
remediated to reduce risk to birds and fish to acceptable levels. The Lauritzen and Santa Fe
Channels, plus some stations at the end of Richmond Inner Harbor, would require remediation on
the basis offish tissue concentrations found there. To reach protective concentrations in fish and
benthic invertebrates, sediment concentrations would need to be between 200 and 500 g/g
organic carbon (OC). Sediment concentrations exceeding 300 g total DDT/g OC were toxic to
amphipods, and those exceeding 100 g/g OC had a reduced abundance of amphipods. This
minimum effects threshold (100 g/g OC) represents a bulk sediment concentration of 1.9 mg/kg
total DDT at 1.9% total organic carbon (TOG). The record of decision for the United
Heckathorn site was signed on October 26,1994, requiring the dredging of all soft bay mud from
the Lauritzen Channel and Parr Canal, with monitoring to document that remediation goals for
the site are achieved. The final remediation goals for the site include that the average sediment
concentration be below 0.59 mg/kg total DDT, which should be protective of humans and fish-
eating birds.
5.4.3. Metal Bank of America
5.4.3.1. Site History and Background
The Metal Bank of America site is located on the Delaware River in Philadelphia, PA.
Between 1968 and 1973, transformer salvage operations were conducted at the site. The waste
oil from the transformers was stored in an underground storage tank adjacent to the river. PCB
5-13
-------
oil from the tank and other operations at the site formed a light nonaqueous phase layer
(LNAPL), which seeped into the mudflat and Delaware River adjacent to the site and resulted in
an emergency removal action. As part of this action, an oil recovery system operated between
1983 and 1989. PCBs are the major contaminant at the site; however, polyaromatic
hydrocarbons (PAHs), phthalate esters, and trace elements also have been detected in
groundwater at the site. These contaminants may be present from the burning of electrical wire
as part of metal recovery operations.
The Delaware River in the vicinity of the Metal Bank site provides habitat for Federal-
and State-designated endangered shortnose sturgeon. Shad, herring, white perch, and catfish
spawn near the site. Fishing advisories have been implemented in the river because of PCB
contamination. Because the National Oceanic and Atmospheric Administration (NOAA) has
technical expertise in aquatic ecological risk assessment and the site has the potential to
adversely affect aquatic habitats and species for which NOAA serves as a natural resource
trustee, NOAA was asked to conduct the aquatic ecological risk assessment for the site in support
of EPA Region 3. The risk assessment report was finalized in March 1994.
5.4.3.2. Problem Formulation and Conceptual Model
Aquatic habitats of concern at the Metal Bank of America site include the surface waters,
tideflats, and bottom substrates of the Delaware River, a freshwater tidal system. The shortnose
sturgeon spend their entire life cycle in the Delaware River, and some of these fish may remain
in the section of the river near the site following spawning. The river also provides habitat for a
wide variety of other freshwater, estuarine, and anadromous fish species and benthic .
invertebrates such as blue crab.
PCBs were the primary contaminant evaluated for the risk assessment because of their
elevated concentrations in groundwater, nonaqueous phase layer (NAPL), and sediment. PAHs,
phthalates, DDT, and cadmium were secondary contaminants evaluated because of elevated
concentrations in NAPL and/or sediment.
'i i
Exposure pathways were considered from surface water, NAPL, and sediment through
ingestion (including food chain accumulation) and direct contact. Accumulation in biota was
considered as the integrating pathway for PCB exposure. PCBs (and DDT) are known to elicit
theur most severe effects through bioaccumulation. The effect of PCBs of greatest concern to
NOAA is the potential for disruption of reproduction and toxicity to early life stages offish after
maternal transfer of PCBs to eggs. The other contaminants considered in the risk assessment
(PAHs, phthalates, DDT, and metals) are known to have the potential for direct toxicity to
sensitive benthic invertebrates and sensitive life stages offish and other aquatic biota. Although
5-14
-------
not explicitly stated in the risk assessment report, the assessment endpoints considered by NOAA
included:
• Protection of individual shorttiose sturgeon from reproductive effects,
• Protection of populations of other fish from reproductive effects,
• Protection of benthic invertebrates from direct toxicity, and
• Protection of fish (including shortnose sturgeon) from direct toxicity.
Silvery minnows, channel catfish, and white perch were selected as representatives of the
fish community near the site. Channel catfish are benthic species and would be exposed to
contaminated sediments; they also were used as a surrogate species for shortnose sturgeon.
Silvery minnows are forage fish that utilize mudflats near the site. White perch are abundant
near the site and are recreationally important. Asiatic clams (Corbiculaflumined) were used to
evaluate bioavailability of PCBs to benthic organisms and to evaluate the link from the site to
fish through the food web.
5.4.3.3. Measurement Endpoints and Approach
For the assessment endpoints relating to the protection offish from reproductive effects,
tissue residue effect threshold concentrations were developed from the literature and compared
with fish tissue concentrations of silvery minnow and channel catfish taken from near the site.
For the endpoint relating to the protection of benthic invertebrates from direct toxicity, sediment
concentrations were compared with toxicity threshold concentrations taken from the literature.
Concentrations in NAPL, groundwater, and surface water also were evaluated for their potential
toxicity to benthic invertebrates that may be exposed in the discharge area. The results of a
qualitative benthic community assessment conducted in 1991 also were evaluated to determine
whether the benthic community appeared to be at risk. For the assessment endpoint relating to
direct toxicity to fish, concentrations in surface water were calculated for low-flow and
average-flow conditions based on concentrations in groundwater and compared with ambient
water quality criteria and maximum allowable toxicant concentrations from EPA guidance.
Concentrations of contaminants in dams were used as evidence to link contamination at the site
to concentrations in higher trophic levels and as an integrator of exposure throughout the mudflat
area.
A weight-of-evidence approach was used for each assessment endpoint The aquatic
areas near the site were divided into three areas for evaluation based on the gradient of
5-15
-------
contamination. An area within 15 meters of the site contained the highest sediment PCS
concentrations and was considered to be the discharge area for NAPL and groundwater. The
mudflat within 30 meters of the site contained elevated concentrations of PCBs and other
contaminants and was evaluated separately. Other areas of the mudflat and Delaware River near
the site contained lower concentrations of the other contaminants and were considered as a third
area for evaluation.
Tissue residue toxicity reference thresholds were developed from the literature by
compiling available studies and selecting the 10th and 50th percentiles of the effect
concentrations available. To evaluate sediment concentrations, both the arithmetic mean and
95% upper confidence interval were compared with effects range low, effects range medium, or
apparent effects threshold concentrations (Table 5-4). Maximum allowable toxicant
concentrations for the fathead minnow (for PCBs) were divided by 100 to account for lack of
chronic toxicity information and to extrapolate for the lack of species-specific information for
shortnose sturgeon and other fish species of concern.
5.4.3.4. Risk Characterization
Assessment endpoints relating to fish reproduction effects were evaluated using tissue
residues and literature effects thresholds. Sampling of Asiatic clams and fish from near the site
indicated that PCBs are accumulating in biota. PCB congener analysis of clams, sediment, and
groundwater demonstrated a similar pattern of PCB accumulation as found in mudflat sediment
and groundwater, indicating that the PCBs found in the clams come from the Metal Bank site.
Table 5-4. Mean and upper 95% confidence limit (CL) concentrations
Ong/kg) of total PCBs in sediments near the Metal Bank of America site
normalized to dry weight and total organic carbon (TOC)
Area (N)
Riprap (13)
Nearfield (<30)
Farfield (>30)
Dry weight
mean
5.9
3.8
0.87
Upper 95% CL
9.4
5.0
1.2
TOC normalized
mean
150
79
30
Upper 95%
240
110
44
CL
Effects range low: 0.023 mg/kg dry weight.
Effects range medium: 0.18 mg/kg dry weight.
5-16
-------
Asiatic clams from five stations in the mudflat contained 0.2 to 1.0 mg/kg total PCBs. Silvery
minnows accumulated 0.55 to 2.8 mg/kg (whole body), and filets and whole-body channel
catfish near the site contained 1.1 to 4.0 mg/kg (wet weight). Fish tissue (both silvery minnow
and channel catfish) exceeded the 10th percentile tissue toxicity threshold (0.2 mg/kg) but not the
median threshold (7.0 mg/kg), which indicates potential risk of reproductive problems in these
species. Shortnose sturgeon may be at risk if they are more sensitive than channel catfish
because of their longevity, habits, and higher lipid content. However, the relative sensitivity of
shortnose sturgeon compared with catfish or other species is not known.
The potential for direct toxicity to benthic invertebrates was evaluated using sediment
concentrations, literature toxicity thresholds, and benthic community analysis. Sediments in the
mudflat contained up to 16 mg/kg total PCBs.
The benthic community survey conducted in 1991 did not include concurrent sediment
analysis, so distance from the site was used as an indicator of PCB concentrations. Samples
taken closer to the site exhibited reduced diversity. Sediment concentrations near the site greatly
exceeded the highest toxicity thresholds for PCBs, PAHs, and phthalates, indicating that the
benthic community is at risk. The risk is greatest in the area closest to the site. PAHs and
phthalates do not exceed toxicity thresholds beyond 30 meters from the site, but risk from PCBs
extends out into the Delaware River. Exposure to NAPL would result in acute toxicity due to
high concentrations of PCBs, PAHs, and phthalates. It is most likely that benthic fauna would be
affected if they were exposed to NAPL.
The potential for direct toxicity to fish was evaluated using calculated surface water
concentrations and toxicity reference: thresholds available from EPA guidance. Only PCBs
exceeded toxicity thresholds in water. Predicted PCB concentrations for sturgeon (1.34-1.97
ng/L) exceeded toxicity reference concentrations (1 ng/L) only within 15 meters of the site. The
water threshold was exceeded in this area by only a factor of 2.
Discharges of NAPL would be expected to be confined to a small area near the riprap.
However, concentrations of PAHs, phthalates, and PCBs in NAPL exceeded toxicity reference
concentrations by five orders of magnitude, and acute effects to any fish directly exposed to
NAPL would be expected.
5.4.3.5. Conclusion
The risk assessment concluded that the Metal Bank of America site posed risk to fish
reproduction (including shortnose sturgeon) and to the benthic community hi the mudflat
adjacent to the site. A major strength of the risk assessment was the inclusion of a substantial
analysis of uncertainty around the data and conclusions. The risk assessment identified that the
5-17
-------
major concern for the site is the effects of bioaccumulation of PCBs hi fish and shellfish, which
demonstrates integrated exposure through surface water, sediment, and food web pathways. This
information is providing the basis for the selection of a remedy to reduce sediment contamination
to concentrations below those associated with reproductive effects in fish and to control the
discharge of contaminated ground water and LNAPL. The data also will provide the basis for
monitoring effectiveness of a remedy for the site. A final record of decision is expected in 1997.
Although no threshold concentrations
for sediment were provided hi the risk
assessment for sediment for reproductive
effects, relationships between sediment and
biota concentrations calculated in the risk
assessment were used to estimate protective
sediment concentrations based on site-
specific bioavailability and effects thresholds
from the literature. The risk assessment
provides an evaluation and compilation of
available tissue residue effects concentrations
for PCBs that has proven useful at other PCB
sites throughout the country.
5.4.4. Data Quality Objectives Process
This section is an example of a
process, not of a specific case study. Federal
agencies manage a wide variety of ecological
resources at various sites. Diverse sites, such
as many of those found throughout DOE and
the U.S. Department of Defense (DoD), pose
many technical challenges that are not
typically associated with smaller, simpler
sites (e.g., industrial and commercial sites
measured in acres, sites with single
contaminants, sites without radionuclide
contaminants). For example, DOE and DoD
sites may include relatively undisturbed and
sensitive habitats (e.g., wetlands, semiarid
IMe of the BQO Process in Ecological
Bisk Assessment
The DQO process involves th© following
seven steps:
1* State the problem
>. jj- j, •" •'•'•'
f S f «•/<• '
2. ^Identify the decision^
3* Identify inputs to the decision
4. Define the study boundaries
S* Develop a decision Me
^ •* •*
•6* Specify tolerable limits on decision errors
7. Optimize the design for obtai»ing date*
The. DQO process offers risk assessors a
standardized procedure $br designing an
effective, efficient risk assessment It is a
strategic planning-based approach whose
objective is to ensure that "data of sufficient
quality and quantity to support defeasible
decision making'* are collected^ without
"unnecessary,, duplicative^ or overly precise
data" being collected (U>S.EPA» 1993> The
DQO process meets this objective through
the application of seven planning steps
(based on the scientific method) that are
"designed to ensure that the type* qaiantityt
and quality of environmental data used, ia
decision making are appropriate for the
intended application" (ILS, EPA* 1993).
5-18
-------
deserts), threatened and endangered species, woodland habitats, former agricultural lands, and
highly disturbed industrialized lands. On any given site, specific types of ecological resources
may occur entirely within the site boundaries or may be distributed across and beyond site
boundaries.
These sites have a great range of contaminant profiles, sizes, climates, elevations, biomes,
ecosystems, and habitat types. The assessments are complex and vary greatly in purpose, scope,
approach, and implementation. Therefore, the ability to increase the efficiency and effectiveness
of ecological risk assessments by a more standardized design and conduct is important. The data
quality objectives (DQO) process developed by EPA (1994) offers an effective means of
achieving this objective, and it is being used to assist in the design and conduct of some
ecological risk assessments by DOE and other Federal agencies. The DQO process also offers
risk assessors and other participants a means for identifying and substantiating necessary changes
in scope, approach, cost, and schedule change for technical reasons during the conduct of me
assessment.
5.5. NATURAL RESOURCE DAMAGE ASSESSMENT AND ECOLOGICAL RISK
ASSESSMENT
5.5.1. What Is Damage Assessment?
CERCLA Section 107(a)(4) (c) establishes liability for damages for injury to, destruction
of, or loss of natural resources, including the reasonable costs of assessing such injury,
destruction, or loss. Natural resource!? are defined to include land, fish, wildlife, biota, air, water,
groundwater, and drinking water supplies and other resources belonging to, managed by, held in
trust by, appertaining to, or otherwise controlled by the United States, any State or local
government, any foreign government, or any Indian tribe. Natural resource damage assessment
is the process used to assess damages to natural resources from releases of oil or hazardous
substances and to obtain compensation to restore injured natural resources and their services.
The damage assessment process used by natural resource trustee agencies is guided by a series of
regulations. Under the National Contingency Plan, natural resource trustees are defined to
include States, tribes, and five Federal agencies (the Departments of Commerce, Interior,
Agriculture, Energy, and Defense).
Regulations describing procedures for assessing damages to natural resources from
discharges of oil or releases of hazardous substances under CERCLA were promulgated by the
U.S. Department of the Interior (DOI) and can be found in 43 CFR Part 11. Recently, NOAA
promulgated regulations describing procedures applicable to oil spills under the Oil Pollution Act
of 1990 (OPA), which can be found in 15 CFR Part 990. The paradigm for conducting damage
5-19
-------
assessments embodied in the OPA regulations also is being adopted for CERCLA damage
assessments.
Under the OPA regulations, the assessment process involves three phases:
preassessment, restoration planning (including injury assessment and selection of appropriate
restoration measures), and restoration implementation. Preassessment activities include
determining if natural resources are in the affected area and if they have been exposed to the
contaminants, as well as whether the resources could have been injured by the release.
Preliminary evidence for injury is compiled in this stage, and at the conclusion of the
preassessment, natural resource trustees should be able to decide whether to proceed with
restoration planning activities. Restoration planning is directed toward evaluating potential
injuries to determine the need for and scale of restoration activities. Injury assessment activities
determine the nature and extent of injuries to natural resources and the services they provide.
Following this assessment, restoration options are evaluated to determine their potential for
returning natural resources to their condition had the injury from the release not occurred.
Restoration implementation entails carrying out projects that compensate the public for the
injured natural resources and services. Responsible parties are liable for the cost of restoration
and for reasonable assessment costs.
I
5.5.2. Contrasts Between Ecological Risk Assessment and Damage Assessment
The CERCLA lead response agency is responsible for conducting an ecological risk
assessment. Natural resource trustees are responsible for conducting damage assessments. In
accordance with Section 107(f)(l), natural resource damages must be used to restore, replace, or
acquire the equivalent of injured natural resources. An ecological risk assessment can provide
information on injuries to natural resources, but by law Superfund money may not be used to
conduct damage assessments. Injury is defined by regulation as death, disease, behavioral
abnormalities, cancer, genetic mutations, physiological abnormalities, and physical deformities.
The objective of ecological risk assessment is to determine whether an ecological risk is
present at a site, link the risk to site-specific contamination, and provide sufficient information to
determine site action. The link between exposure to contaminants and adverse effects is critical
in an ecological risk assessment. To justify a remedial action based on ecological concerns, a
risk assessment must establish that an actual or potential ecological threat exists at the site. The
natural resource damage assessment process requires the trustees to demonstrate that injury has
occurred to natural resources and services, not just that there is potential ecological risk. This
typically requires that extra quality assurance information is collected or more rigorous studies
5-20
-------
are conducted, especially in the event that the damage assessment must withstand court
challenge.
An ecological risk assessment can provide necessary information for a damage
assessment because it establishes a causal link between site contaminants and adverse effects and
provides information concerning injury, but it may not provide a complete assessment of all
injuries to natural resources. Ecological risk assessments are designed to evaluate baseline
ecological risk and often evaluate the most sensitive receptors present, to ensure that all biota at
the site are protected. In contrast, natural resource damage assessments may focus on particular
representative species of interest to the trustees (for example, recreationally important species)
that may or may not be the most sensitive receptors present.
5.5.3. Requirement for Coordination of Assessments
CERCLA and the NCP require the lead response agency to coordinate assessments with
the natural resource trustees and to notify the trustees when potential injury is identified. This
requirement benefits all participants in the remedial process because early involvement of the
trustees can improve ecological risk assessment and can facilitate the settlement of liability at the
end of the process.
5.6. RISK ASSESSMENT METHODOLOGY DEVELOPMENT
Methodology development areas of particular interest under CERCLA include the
following:
• Address design needs specific to ecological risk assessments in the work plan phase,
including interested parties and management needs.
• Improve exposure/effects models, extrapolation techniques for various exposure
pathways, and validation techniques.
• Develop the use of chemical bioavailability, additional tissue-based toxicity thresholds,
and scientifically sound thresholds for screening values for soil contaminants.
5.7. SITE REMEDIATION AND THE ROLE OF ECOLOGICAL RISK ASSESSMENT
Superfund risk managers typically address three key questions at every site: (1) Do site
releases present an unacceptable risk to important ecological resources? (2) If the answer is yes,
should the site be actively cleaned up or will the remedy do more damage (and thus not provide
short-term protectiveness)? and (3) If cleanup is warranted, how do you select a cost-effective
response and cleanup levels that provide adequate protection?
5-21
-------
As was seen in the case studies, EPA considers the results from a battery of toxicity tests,
field studies, and food-chain models to determine whether or not observed or predicted adverse
effects are unacceptable. It can then use the same studies to select chemical-specific cleanup
levels that are believed to be protective at that site.
Whether or not to clean up a site is often the most difficult risk-based decision to make.
I : :, • 'i
Even though an ecological risk assessment may demonstrate that unacceptable ecological effects
have occurred or are expected to occur in the near future, removal or in situ treatment of the
contamination may do more ecological damage (often due to widespread physical destruction of
habitat) than leaving it in place. When evaluating remedial alternatives, the NCP highlights the
importance of considering the long-term and short-term impacts of the various alternatives in
determining which alternatives "adequately protect human health and the environment." A
remedy that does significant short-term ecological damage often would not be considered to meet
the NCP threshold criteria of "protective."
Assuming remediation is technically practicable and not cost prohibitive, risk managers
consider the long- and short-term ecological impacts of active remediation versus natural
attenuation of the contaminants. The evaluation of ecological impacts from implementing
remedial alternatives is part of the ecological risk assessment process and should be discussed in
a feasibility study. In most cases, unless they are very large, sites with persistent contaminants
that are also mobile are remediated. At sites with contaminants that degrade or with sediment
contaminants that will become unavailable because of natural deposition of uncontaminated
sediment over them, preventing additional releases may be the most appropriate remedy.
Before making a response decision, the risk manager, in consultation with an ecological
risk assessor, often considers many of the following factors:
• The magnitude of the observed or expected impacts of site releases on the affected
ecosystem component (e.g., fish population, benthic community)
• The likelihood that these impacts will occur or continue
• The size and functional value of the impacted area in relation to the larger ecosystem
• Whether or not the impacted area is a highly sensitive or ecologically unique environment
• The recovery potential of the impacted ecosystem and expected persistence of the
chemicals of concern under the site conditions
• i 1 . ' '• ! .
• Short-term and long-term impacts of the remedial alternatives on the site habitat and
larger ecosystem
5-22
-------
• Effectiveness of the remedy; that is, whether there are other continuing, nearby, non-
Superfund releases or other types of stressors that will continue to adversely impact the
ecosystem after the cleanup is implemented
• Community opinion on the value of the affected portion of the ecosystem and of the
natural resources affected
• Whether or not there will be any remaining residual risks that may need to be addressed
by a natural resource trustee.
It is the responsibility of the risk manager, hi consultation with the risk assessor, to select
a remedy and ensure cleanup levels for the site that are reasonable. This decision can be made
only after a thorough consideration of all nine criteria described in the NCP. Because of the high
complexity of ecosystems and the large number of species potentially affected at every site, there
will usually be a relatively high degree of uncertainty concerning the levels deemed to be
protective—are they too high or too low? At these sites, monitoring of the affected ecological
receptors should be performed after the remedy has been implemented in order to determine if
recovery is occurring in a reasonable time frame and whether or not an additional response action
is warranted.
5.8. REFERENCES
Barnthouse, LW; Suter, GW, II. (1986) User's manual for ecological risk assessment. OKNL-6251. Oak Ridge
National Laboratory. Oak Ridge, TN.
Falandysz, J. (1986) Metals and organochlorimes in adult and immature males of white-tailed eagle. Environ
Conservation 13:69-70.
Falandysz, J; Jakuczun, B; Mizera, T. (1988) Metals and organochlorines in four female whitetailed eagles. Marine
Pollut Bull 19:521-526.
Finley, MX; Stickel, WH; Christensen, RE. (1979) Mercury residues in tissues of dead and surviving birds fed
methylmercury. Bull Environ Contam Toxicol 21:105-110.
Henriksson, K; Karppanen, E; Helminen, M. (1966) High residue of mercury in Finnish whitetailed eagles. Omis
Fenn 43:38-45.
Koeman, JH; Hadderingh, RH; Bijiveld, MF1 J. (1972) Persistent pollutants in the white-tailed eagle (Haliaeetus
albicilld) in the Federal Republic of Germany. Biol Conserv 4:373-377.
Lee, H; Lincoff, A; Boese, BL; et al. (1994) Ecological risk assessment of the marine sediments at the united
Heckathorn superfund site. Final report to Reigion DC. Prepared for the Pacific Ecosystems Branch, Environmental
Research Laboratory, U.S. Environmental Protection Agency, Newport, OR. EPA ERL-N-269.
5-23
-------
Oehme, G. (1981) Zur Quecksilbeffijckstandsbelastung tot aufgefundener Seeadier, Haliaeeius albicilla. In: den
Jahren 1967-1978. (In German with English summary.) Hercynia 18:353-364.
U.S. Environmental Protection Agency. (1992) Framework for ecological risk assessment. Risk Assessment Forum,
Office of Research and Development, Washington, DC. EPA/630/R-92/001.
U.S. Environmental Protection Agency. (1993) Guidance for planning in support of environmental decision making
using the data quality objectives process (interim final). Quality Assurance Management Staff, Washington, DC.
U.S. Environmental Protection Agency. (1994) The new generation of environmental protection. EPA's five-year
strategic plan. Washington, DC. EPA/200/13-94-002.
U.S. Environmental Protection Agency. (1997) Ecological risk assessment guidance for Superfund: process for
designing and conducting ecological risk assessments (interim final). Risk Assessment Forum, Washington DC
EPA/540/R-97/006. NTIS PB97-963211.
5-24
-------
POTENTIAL USES OF
ECOLOGICAL RISK ASSESSMENT
-------
-------
6. AGRICULTURAL ECOSYSTEMS
6.1. SUMMARY
The U.S. Department of Agriculture (USDA) has primary responsibility for agricultural
production issues. Mandatory risk assessment of production agriculture has been established
only recently. The Federal Crop Insurance Reform and Department of Agriculture
Reorganization Act of 1994 established new statutes for regulatory analysis requirements of
Department regulations. Included is the requirement of a risk assessment and cost-benefit
analysis for all proposed major regulations, defined as regulations having an annual economic
impact of $100 million and primarily Effecting human health, human safety, or the environment.
Two of the case studies in this chapter are USD A conservation programs that were conducted
under the risk assessment requirement, while the other case study involves shrimp aquaculture.
The aquaculture case study followed EPA's ecological risk assessment approach but had
proceeded only as far as problem formulation at the time this report was prepared. The
assessments done for the USDA conservation programs relied heavily on the ecological risk
assessment approach in the early stages of development, but the assessment teams modified some
aspects of the framework to address: specific agency requirements, scope and scale issues, and
management goals for the associated regulations. As a result, there was some deviation from the
ecological risk assessment process. For example, the time and information available for the
assessment necessarily limited the degree of detail and quantisation possible in the assessment.
Due to statutory mandates and other factors, identified assessment endpoints did not always
involve an ecological entity and attribute. And, as with the nonindigenous species chapter, there
was some overlap of risk assessor and risk manager roles; in risk characterization, there was
strong emphasis on providing results directed at risk management objectives.
The two USDA conservation program assessments included in this chapter were done for
the new Environmental Quality Incentives Program (EQIP) arid the revamped Conservation
Reserve Program (CRP). Both programs seek to reduce the adverse impacts of agricultural
practices on natural resources on and off the farm. EQIP provides assistance to producers to
encourage the application of conservation strategies to farming activities that can result in
environmental degradation. CRP enrolls the most environmentally sensitive cropland into
permanent resource-conserving covers for 10-15 years. The objectives of the EQIP and CRP
assessments included (1) identifying those agricultural activities and practices that place natural
resource values at risk; (2) characterizing the mechanisms that result in risk; (3) characterizing
the magnitude and extent of the environmental risk; and (4) where possible, making
6-1
-------
recommendations to risk managers and decision makers on how the results of the assessment can
be used in program development and implementation.
In problem formulation, both assessments developed assessment endpoints, conceptual
model diagrams, and analysis plans. Both assessments were largely qualitative in focus. EQIP
focused on agricultural practices posing risks to the environment, including crop production and
grazing and livestock production. Conceptual diagrams were developed to hypothesize the
cause-and-effect pathways of environmental risk. These pathways were associated with (1)
soil/land disturbances, (2) irrigation water application, (3) pesticide application, (4) nutrient
application, (5) brush and noxious weed invasion, (6) pasture grazing, (7) rangeland disturbance,
and (8) confined livestock production. A qualitative evaluation of ecological effects was based
on analysis of these pathways and interpretation of agricultural use and impetct maps for the
continental United States. The assessment presented the principal causes of use and quality
impairment of rivers, streams, lakes, estuaries, reservoirs, and ponds from agricultural activities
by region in tabular form. Data were presented as miles of use impairment and percentage of
water quality impairment from such factors as pesticides, sediment, pathogens, and salinity.
Livestock concentrations on aper-State basis also were presented to indicate where
environmental stressors from associated activities might be the greatest. The risk
characterizatioii section of the assessment identified the magnitude of environmental
consequences and delineated how those consequences can be addressed by proven on-farm
conservation strategies. It discussed the potential for risk reduction and EQIP baseline
comparisons and summary conclusions based on the four previous conservation programs
replaced by EQIP. The risk characterization focused mainly on recommendations to risk
managers. The assessment team attempted to analyze where the cumulative effects or impacts of
agricultural activities are occurring across the United States. Both recovery of ecological
systems and major uncertainties in the assessment were discussed.
For CRP, the task was complicated by the huge scope and complexity of the problem; as
many as 36.4 million acres of environmentally sensitive cropland have been enrolled in the
program. It was extremely difficult to establish detailed and consistent databases that empirically
describe the stressor-environmental component relationships and their impacts. Problem
formulation was similar to that for EQIP, but, in contrast, the CRP conceptual diagrams were
limited to crop production activities and did not include livestock and grazing components.
Analysis of ecological effects were similar for EQIP and CRP. The main difference is that the
CRP analysis was based, in places, on evaluating the potential impacts as if there had not been a
CRP in place for more than 10 years. For risk characterization, the assessment team addressed
the identity and location of the type of cropped acreage that should receive priority for
6-2
-------
enrollment in CRP. The risk assessment will help national-level policy makers to generally
target the situations and areas where participation in the program is most likely to address
environmental degradation. With this information, national-level policy makers can work with
the States and localities in these areas to refine the application of CRP activities toward solutions
to the environmental problems. As with EQIP, major uncertainties and time to natural resource
recovery were discussed. However, in contrast with EQIP, no direct recommendations to risk
managers were made in this assessment.
The third case study addresses the potential introduction and spread of nonindigenous
pathogenic shrimp viruses to shrimp aquaculture and to the wild shrimp fishery in the United
States. Outbreaks of these viruses on U.S. shrimp farms and the appearance of diseased shrimp
in U.S. commerce prompted the Federal interagency Joint Subpommittee on Aquaculture to
initiate an ecological risk assessment. Following EPA's ecological risk assessment guidelines
process, a problem formulation step was used to develop a conceptual model, list the potential
effects of the viruses on shrimp and other aquatic species, summarize the basic life history of
shrimp, identify potential stressors affecting the shrimp population, and identify potential
pathways for the exposure of wild shrimp to the viruses. After using the framework as a tool for
organizing information, the work group proposed several options for doing an assessment. It
also recommended that a formal ecological risk assessment be; done to provide information
needed to address international trade issues, national and State regulatory obligations, and the
needs of industry, environmental groups, and the public.
The challenge to government agencies that conduct risk assessment on agricultural
production is to develop an iterative process early hi regulatory development. This includes
identifying when risk assessment will be required, clearly identifying risk management
objectives, and establishing an iterative process between risk assessment and risk management
for the course of program development. Also, it is necessary that there be clearly identified
strategies for using these tools so that programs can minimize adverse ecological impacts while
achieving other agricultural goals. Risk management must play a role hi the development and
use of risk assessments if the intent is for risk assessment to aid in the decision-making process
of regulatory development.
6.2. INTRODUCTION
Federal agencies traditionally have used a variety of tools to influence the impact of
agriculture on ecological resources. These tools include regulations, research and development,
training and education, financial incentives, and information management. In some cases,
6-3
-------
ecological risk assessments are now being used to aid decision makers in using these traditional
tools more effectively.
6.2.1. Historical and Current Use of Risk Assessment in Agricultural Production
6.2.1.1. Environmental Impacts of Production Practices
Agriculture is a multifaceted industry that depends heavily on the availability of
sustainable land, water, plant, and animal resources. However, agricultural production generally
has had an adverse impact on these resources and on the ecological framework (both on and off
the farm) that ties the resources together in a sustainable way. Future demand for food and fiber
requires a balancing of agricultural production technologies and ecological principles to create
agroecosystems capable of high levels of production on a sustainable basis with minimum
adverse off-site impacts.
Although some persons question the validity of applying the term "ecosystem" to farmed
land, there is no question that there are ecological interactions and relationships on farmed land.
These encompass climatic changes, soil quality factors, water quality and quantity, desirable and
undesirable insects, diseases, small mammals, bird species, and other forms of wildlife that have
adapted to agricultural landscapes. Although these interactions are not part of a "natural"
ecosystem, they still exist and offer opportunities to apply ecological principles about energy
flow, nutrient cycling, and biodiversity that can help promote sustained productivity.
Ecological risk assessment has played only a minor role in agricultural development in
the United States. This limited use is largely the result of timing. Most of the modern
agricultural system was in place before risk assessment came into use. However, agriculture
continues to evolve, and risk assessment is now being done to support public decision making
about land-use policies, new technology, and alternative production systems.
By 1920, most of the lands in the United States that could be used for agriculture were
already in production. About 500 million acres were used for growing crops, including hay
production. Another 700 to 800 million acres of public and private lands were used for livestock
grazing.
By the 1950s, most of the technologies associated with modern agriculture were already
widely used and today are continually refined. These technologies include the introduction of
exotic plants and animals, plants produced in monoculture, chemical fertilizers and pesticides,
irrigation, mechanization based on the internal combustion engine, and barbed-wire fencing.
Applying these technologies to the vast land areas used for agricultural production has had a
significant impact on ecological systems, both on and off the farm.
6-4
-------
The U.S. Department of Agriculture (USD A) has primary responsibility for agricultural
production issues. The U.S. Department of the Interior, through the Bureau of Land
Management, manages the majority of public lands leased for grazing livestock, in concert with
the Forest Service. (Any risk assessments conducted with regard to these lands are discussed in
Chapter 8.) This leaves USD A to address the issues associated with production on private lands.
Agencies within USD A regularly conduct risk assessments for various activities of the
Department. Some of these activities are discussed elsewhere in this report. However,
mandatory risk assessment of production agriculture has been established only recently. The
Federal Crop Insurance Reform and Department of Agriculture Reorganization Act of 1994
established new statutes for regulatory analysis requirements of Department regulations.
Included is the requirement of a risk assessment and cost-benefit analysis for all proposed major
regulations (Public Law 103 [PL 103J). For purposes of this statute, a major regulation is
defined as one that has an annual economic impact of $100 million and primarily affects human
health, human safety, or the environment. The Reorganization Act requires USDA to conduct
thorough analyses that make clear the nature of the risk beingi managed, the reasoning that
justifies the proposed rule, and a comparison of the likely costs and benefits of reducing the risk.
The Reorganization Act also established the Office of Risk Assessment and Cost-Benefit
Analysis, whose function it is to ensure that these analyses are based on reasonably obtainable
and sound scientific, technical, economic, and other date.
6.2.1.2. Environmental Impacts on Production
Although ecological risk assessments on the impacts of biological, physical, and chemical
stressors on agricultural production aire not common, some work has been done and more is
planned for the future. One example is a risk assessment conducted on wildlife damage to field
corn (Wywialowski, 1996). In another example, EPA's report Framework for Ecological Risk
Assessment (U.S. EPA, 1992) was used by the National Crop Loss Assessment Network
(NCLAN) to develop an approach for assessing the impact of ozone on crop production (U.S.
EPA, 1993). A goal of the NCLAN risk assessment was to provide risk managers with a better
understanding of the potential impacts of ozone on crop production and develop more
appropriate ozone standards. The regulatory impetus behind:this study was the Clean Air Act
(1970), which required EPA to set National Ambient Air Quality Standards (NAAQS) for "any
air pollutant which, if present in the air, may reasonably be anticipated to endanger public health
or welfare." EPA is responsible for developing and promulgating both primary (human health)
and secondary (public welfare) NAAQS. The 1977 amendments to the Clean Air Act required
that the criteria for the NAAQS be periodically reviewed and revised to include new information.
6-5
-------
In 1978, EPA conducted a review of the literature to determine the impact of ozone on vegetation
and published an analysis. As a result of the analysis, EPA accepted the primary ozone standard
as a reasonable secondary standard. However, the credibility of this "secondary" standard
suffered, and NCLAN was established to conduct a study to assess the impact of ozone on
agricultural resources and to provide the most useful data and criteria for the review of the
standard. At the time of this writing, the study has played a significant role in the proposed new
secondary standard for ozone.
The NCLAN study was initiated by EPA as a risk assessment and, with some
modification, followed the four steps of EPA's framework report: problem formulation,
characterization of ecological effects, characterization of exposure, and risk characterization. In
problem formulation, the relevant ecological components were identified and described, and
relevant endpoints defined. Next, ozone exposure-plant response was studied experimentally in
the field, and ecological effects response models were developed. The exposure characteristics
then were described and documented. Finally, risks were characterized in both crop yield and
economic terms.
:: • ! | :
As a national risk assessment, NCLAN was limited by time and funding constraints. As a
result, there was limited spatial representativeness (there were only six sites for a national
assessment) and few experimental designs implemented to assess the effects of interacting
stressors (U.S. EPA, 1993). Nonetheless, the results of the ecological risk assessment did play a
significant role in developing the proposed new secondary standards for ozone.
6.2,1.3. Environmental Impacts of Aquaculture
Aquaculture systems have expanded rapidly in recent years. Although these production
systems have the potential to impact significantly on aquatic ecosystems, public policies for
these new systems were made without the benefit of an ecological risk assessment until recently.
Regulatory control over aquaculture production is shared by several agencies, including the
National Oceanographic and Atmospheric Administration, National Marine Fisheries Service,
U.S. Fish and Wildlife Service, Animal and Plant Health Inspection Service (USDA), and EPA.
State agencies are also heavily involved in establishing permit (effluent) and production
regulations.
Ecological risks associated with aquaculture are now coming into focus. Aquaculture
systems, such as catfish production in the southern States, are the aquatic equivalent of feedlots
for cattle, hogs, and poultry. However, with aquaculture, pollutants and pathogens, via effluents,
have a greater potential to impact natural aquatic ecosystems because of the similar or identical
trophic levels of potential receptor species. Periodic draining of catfish ponds contributes to (1)
6-6
-------
increased turbidity, (2) organic, pesticide, and nutrient loading.; and (3) in some situations,
adding chemicals used for disease treatment into receiving water bodies. There is also a potential
risk that wastewater could transport pathogens to wild species.
Another potential ecological impact of aquaculture is the threat to native species from
farm stock escapees. For example, the wild Atlantic salmon is facing genetic alteration from
escapees from Norwegian fish farms. These farms expect to produce 330,000 tons of Atlantic
salmon this year, compared with a total wild Atlantic salmon catch of 3,800 tons in 1995. The
fish in these farms have been selected over five generations for fast growth and a high fat
content. In 1995, between 200,000 and 650,000 domesticated salmon escaped and intermingled
with native fish. In addition to crossbreeding, the fish raised hi confinement are known to carry
parasites such as sea lice, and although vaccinated, they also may carry pathogenic bacteria and
viruses to native fish species.
Based on these and similar experiences, ecological risk assessment is now being done on
selected aquaculture production systems.
6.2.2. Applicability of EPA Ecological Risk Assessment Framework and Guidelines to
Agricultural Ecosystems
Only recently has ecological risk assessment been used in public policy decision making
for agricultural issues. With hindsight, we can see benefits that could have been derived by
applying our existing understanding of ecological risk assessment to agricultural development
over the past 200 years. Such assessments could have aided in mitigating adverse ecological
impacts in many ways. For example, public programs for conserving soil and water resources,
regulating agrochemicals, protecting ecologically sensitive areas, and controlling the introduction
of exotic species could have been put in place before millions of acres of land were put into
production and new technologies promoted.
Such assessments also could have played a role in using agriculture more effectively in
protecting ecological values. Many wildlife and fish species benefit from such agricultural
activities as planting of shelterbelts, building of ponds and other water retention facilities,
planting crops that provide winter feed, and controlling fire and disease. Earlier ecological
assessments might have resulted in a more systematic development of such benefits.
Human ecology also has been profoundly affected by agriculture. American society
would not be the same without the benefits derived from its solid agricultural foundation. Earlier
ecological risk assessments could have led to policies that would have made these benefits even
greater. For example, understanding the impacts of overgrazing and plowing fragile lands .could
have mitigated the worst excesses of the Dust Bowl. Early ecological risk assessments also
6-7
-------
might have identified the value of farmlands in regulating urban growth and might have resulted
in farmland protection policies that would have prevented some urban congestion.
Despite its long history, agriculture continues to be a dynamic process, with the continual
development of new technologies, conservation measures, and production systems. The three
case studies provided in Section 6.3, two on the USDA conservation programs and one on
1 ..• i •• ' .. i • i. ; .• i '!,|
aquaculture, all relied heavily on the EPA framework report and on the report Draft Proposed
Guidelines for Ecological Risk Assessment (U.S. EPA, 1996). In the early stages of
development of the risk assessments, the guidelines report was not yet completed. Working
outlines for the assessments were developed using the framework report, but modifications were
necessary because of the scope and scale of the assessments and the physical and chemical nature
of the multiple stressors being assessed. In addition, there was strong emphasis in the risk
characterization process on providing results directed at risk management objectives.
6.3. CASE STUDIES
The two case studies presented in Section 6.3.1 concern nationwide, multiple-stressor risk
assessments conducted by two agencies in USDA. These assessments were the first conducted
under the risk assessment requirement of the Federal Crop Insurance Reform and Department of
Agriculture Reorganization Act of 1994. The third case study (Section 6.3.2) is an
interdepartmental report developed to provide the Joint Subcommittee on Aquaculture with a
basis for discussing and selecting among a range of options for conducting a risk assessment on
shrimp viruses. This report was developed using EPA's framework and guidelines reports (U.S.
EPA, 1992,1996) and followed these documents very closely. The assessments done for the
USDA programs relied heavily on the framework report in the early stages of development.
However, the assessment teams found these documents, in places, to be either unduly restrictive
or inadequate for the scope, scale, and purpose of the assessments given the management goals
for the associated regulations. The result is that there is some deviation from the framework and
guidelines reports in these assessments, particularly with regard to the development of
assessment endpoints and the focus of the risk characterization sections of the documents. The
framework report will probably continue to be modified, as needed, for use in USDA program
and regulation development.
6.3.1. Risk Assessment of USDA Conservation Programs
AS far as can be determined, only two ecological risk assessments of agricultural
production impacts using the EPA framework and guidelines reports have been conducted by a
Federal agency—namely, USDA—as the result of new congressional mandate P.L. 103. These
6-8
-------
two assessments were done for the new Environmental Quality Incentives Program (EQIP) and
the revamped Conservation Reserve Program (CRP). Bom programs seek to reduce the adverse
impacts of agricultural practices on natural resources on and off the farm. These resources
include soil and water, wetlands, and wildlife habitats. EQIP provides for cost-sharing funds,
incentive payments, and technical and educational assistance to producers to encourage the
application of conservation strategies to farming activities that can result in environmental
degradation. CRP enrolls the most environmentally sensitive cropland into permanent resource-
conserving covers for 10-15 years.
Ecological risk assessments were done on both of these programs (FSA, 1997; NRCS,
1997). The objectives of the assessments included (1) identifying those agricultural activities
and practices that place natural resource values at risk; (2) characterizing the mechanisms that
result in risk; (3) characterizing the magnitude and extent of the environmental risk; and (4)
where possible, making recommendations to risk managers and decision makers on how the
results of the assessment can be used in program development and implementation.
In assessments for both programs, the resources to be considered at risk—soil and water,
air, wetlands, wildlife habitat, and grazing lands—were provided by draft regulation before
conducting the assessments. This led to the development of assessment endpoints that might not
be considered common under the EPA framework and guidelines reports. The risk assessors for
the two programs determined that endpoints such as air quality and cultural and historic
resources are intimately and ecologically associated with the designated resources. They are at
risk from the impacts of some agricultural practices and should be considered in the management
decisions made in the two programs.
6.3.1.1. Environmental Quality Incentives Program
6.3.1.1.1. Background. EQIP has four environmental mandates: (1) combine into a single
program the functions of the rescinded Agricultural Conservation Program, the Great Plains
Conservation Program, the Water Quality Incentives Program, and the Colorado River Basin
Salinity Control Program; (2) execute EQIP in a manner that maximizes environmental benefits
per dollar expended; (3) provide flexible technical and financial assistance to farmers and
ranchers who face the most serious threats to soil, water, and related natural resources, including
those threats to grazing lands, wetlands, and wildlife habitats; and (4) provide assistance to
farmers and ranchers to comply with the Conservation Title of the 1996 Farm Bill and other
Federal and State environmental laws.
In creating EQIP, Congress, in the Federal Agriculture Improvement and Reform Act of
1996, provided an initial identification of environmental resources considered at risk. These
6-9
-------
resources were identified as soil, water, and related natural resources, including wetlands,
grazing lands, and wildlife habitats. However, in conducting this assessment, several additional
resources were identified at risk: (1) air quality, (2) cultural and historic resources, and (3)
landscape resources.
The assessment consisted of technical evaluations and analyses that attempted to
characterize the relationships among agricultural production activities, ecosystem stressors, and
resulting adverse ecological effects on particular natural resources. The assessment had three
sections: (1) problem formulation, (2) analysis of ecological effects, and (3) risk
characterization.
6.3.1.1.2. Prpblent formulation. Problem formulation includes an analysis plan, a brief
discussion oi| identification of missing data, and recommendations for additional data collection,
analysis, and evaluation. During the problem formulation stage, data were gathered and used to
identify those agricultural practices or activities posing the greatest risks to the environment.
These were identified generally as crop production and grazing and livestock production.
Conceptual diagrams were developed to hypothesize the cause-and-effect pathways of
environmental risk. These pathways were associated with (1) soil/land disturbances, (2)
irrigation water application, (3) pesticide application, (4) nutrient application, (5) brush and
noxious weed invasion, (6) pasture grazing, (7) rangeland disturbance, and (8) confined livestock
production. Examples of two of the conceptual diagrams are presented in Figures 6-1 and 6-2.
Identified in the conceptual diagrams were the specific assessment endpoints associated
with the resources at risk: structure of off-site resources and habitats, livestock or plant yields,
wetland functions, viability of aquatic communities, good air quality, survival of threatened or
endangered species, extent of natural habitats, quality of cultural resources, potable water
supplies, diversity of terrestrial and avian wildlife species, survival and diversity of terrestrial
and avian communities, function of riparian areas, diversity of natural habitats, and quality of
landscape resources. Aquatic communities, threatened and endangered species, wetlands,
livestock and plant yields, potable water supplies, air quality, and terrestrial and avian wildlife
.'...."'. . ' ! ' H * •
communities were assessment endpoints common to most hypothesized pathways. This reflects
the interconnectivity of agriculturally related natural resources. A detailed discussion of the risk
initiators, system stressors, ecological effects, and assessment endpoints identified in the
diagrams was included in the assessment.
6.3.1.1.3. Analysis of ecological effects. Owing to the lack of comprehensive data and the
uncertainties associated with extrapolation of site-specific data to a landscape scale, the analysis
6-10
-------
SJOSS3J1S
sjujodpug
luaiussassy
6-11
-------
JOJB!JJU|
SJOSS9JJS
See Soil/Land Disturbance Diagram
Reduced Sol! Depth
a
1
t
St
I
a
•I
•s
s
a
e
U
2
I
6-12
-------
of the hypotheses developed through the conceptual diagrams is in qualitative, narrative form.
The discussion centers on the previously identified resources at risk and provides an overall
evaluation of the types and kinds of activities found to be placing the natural resources at risk.
With available data, and in cooperation with the Natural Resources Inventory (NRI) staff,
maps of the continental United States were generated indicating the status (based on 1992 data)
of agriculturally related land uses and potential or actual impacts of agricultural activities. These
included maps of acres in cropland, wind and water erosion on cropland, sediment delivered to
rivers and streams from sheet and rill erosion on farm fields, cropland with conservation needs,
potential nitrogen and phosphate fertilizer loss from farm fields, pesticide runoff and leaching
potential by watershed for field crop production, irrigated cropland, rangeland status, palustrine
wetlands on croplands, and others. Examples of the maps are given in Figures 6-3 and 6-4.
The assessment presented the principal causes of use and quality impairment of rivers,
streams, lakes, estuaries, reservoirs, and ponds from agricultural activities by region in tabular
form. Data were presented as miles of use impairment and percentage of water quality
impairment from such factors as pesticides, sediment, pathogens, and salinity. Livestock
concentrations on a per-State basis also were presented to indicate where environmental stressors
from associated activities might be the greatest.
6.3.1.1.4. Risk characterization. The risk characterization section of the assessment identified
the magnitude of environmental consequences and delineated how those consequences can be
addressed by proven on-farm conservation strategies. It discussed the potential for risk reduction
and EQIP baseline comparisons and stiimmary conclusions based on the four previous
conservation programs replaced by EQIP. The risk characterization focused mainly on
recommendations to risk managers. The assessment team attempted to analyze where the
cumulative effects or impacts of agricultural activities are occurring across the United States. By
using thematic maps of the different resource or landscape features, resources and areas can be
rated as to their risk of degradation. In this manner, a more accurate picture of the types,
locations, and extent of agricultural activities and their relationship to environmental resources
potentially at risk can be ascertained.
Cumulative effects also were assessed, to the extent possible, in an effort to provide risk
managers with a more complete, in-depth analysis for use on an ecoregion basis. The 10
agricultural production regions of the country were used to represent ecoregions. This choice
was made because the States included within each agricultural1 production region were found to
have environmental or ecological similarities. Also, many of the available environmental data
used in this assessment were already presented in this format.
6-13
-------
•8
w
H
d
•s
I
§
I
u
C5
a,
2
o
ft
OS
6-14
-------
6-15
-------
Using this ecoregion approach, it was possible to identify specific farm production
regions facing significant environmental risks. These risks are due to a combination of factors,
including high-intensity agriculture, geologic/geographic conditions, and climate, all acting
simultaneously to exacerbate the on-farm and off-site environmental impacts identified in the
conceptual diagrams.
The major conclusion of the risk assessment was that agricultural production activities, if
done in the absence of conservation technologies and practices, can have serious environmental
impacts. However, the introduction, acceptance, and implementation of resource conservation
technologies can significantly reduce these threats.
The risk assessment team found that the best solutions for environmentally stressed
resources would be conservation measures applied in concerted, concentrated efforts in priority
areas, with smaller scale efforts going to sectors outside priority areas. EQIP should employ a
multiplicity of conservation measures, simultaneously and on a large scale. EQIP should address
not only on-site problems, but also off-site unintended adverse consequences and cumulative
effects. With this "fusillade" conservation approach, remedial actions will have greater effects
than could occur otherwise. Over time, significant ecological improvement should be observed
and downward environmental trends will move away from present "at risk" conditions.
The risk assessment also identified the need for additional data so that risk managers can
be provided with a more complete analysis of all the environmental hazards related to
agricultural production. Better environmental monitoring and evaluation tools need to be
designed to assemble the actual effects of the application of conservation practices on the
environment and on production agriculture.
Several sources of uncertainty also were identified during the analysis. One is associated
with the interrelationships among all the resources of the ecosystem, not just the agricultural
community. Time also adds a dimension of uncertainty. Long-term on- and off-farm effects
may not be noticed until the resource has been so damaged that the productive capacity is beyond
mitigation or restoration. A complete and quantitative environmental risk assessment may be
difficult to perform for several reasons. The effects of applied resource conservation practices
may not be seen immediately. What is done on one farm, tract, or ranch may register little to no
effect, from a cumulative standpoint, on a watershed, hydrologic unit, or ecosystem. In addition,
there is vast uncertainty associated with the role of agricultural production in landscape-scale
ecological degradation.
6-16
:,!;',, i;,. I,,.'Hit:. ,,'i11:..
-------
6.3.1.2. Conservation Reserve Program
6.3.1.2.1. Background. CRP is authorized under subtitle D ofTitle XII of the Food Security
Act of 1985, as amended. The statutory purpose of CRP is to assist owners and operators in
conserving and improving soil, water, air, and wildlife resources on their farms and ranches by
converting highly credible and other environmentally sensitive cropland to permanent resource-
conserving covers for 10 to 15 years. CRP is USDA's largest single conservation program. As
many as 36.4 million acres of environmentally sensitive cropland have been enrolled in the
program. Annual costs have reached nearly $2 billion, and the program has produced substantial
soil erosion reduction, water quality improvement, and wildlife habitat enhancement benefits.
The assessment team acknowledged the difficulties associated with conducting a risk
assessment on a nationwide scale. It sitated that an assessment of the risks to the environment
associated with agricultural production activities is highly complex. Activities undertaken for
crop production form a very interdependent and complex system of cause-and-effect linkages,
including feedback mechanisms and buffers, with the natural resource base. Often, long and
varying time lags are associated with the occurrence of an event or activity and its impact on one
or more elements of the resource base. Also, because of the large and diffuse set of cropping
activities and farming operations, it is difficult to trace the impacts back to the original source.
Further, similar environmental impacts can be caused by nonagricultural activities, and isolating
the specific cause-and-impact relationships is often very difficult. Finally, other factors clearly
outside the control of farm producers, such as weather and market forces, further complicate the
diverse, complex, and dynamic set of environmental cause-and-effect relationships associated
with agricultural cropland use. Because of this incredible complexity and diversity, it was
extremely difficult to establish detailed and consistent databases that empirically describe the
stressor-environmental component relationships and their impacts. These information shortfalls
illustrate the uncertainties associated with supporting the hypotheses established in the
assessment.
6.3.1.2.2. Problem formulation. As noted earlier hi this chapter, the problem formulation
section of the CRP risk assessment is functionally identical tq that of the EQIP risk assessment
(see Section 6.3.1.1.2). The only difference is that the CRP conceptual diagrams and
accompanying discussion were limited to crop production activities and did not include livestock
and grazing components. Pertinent conceptual diagrams developed by the EQIP risk assessment
team were recreated for the CRP risk assessment.
6-17
-------
6.3.1.2.3. Analysis of ecological effects. This section of the CRP risk assessment is also very
similar to the analysis section of the EQIP risk assessment. Water impairment tables and data
and pertinent NRI maps were incorporated. However, also included were maps of the
continental United States (developed by the Economic Research Service) for discussion of air
quality issues. Air quality was stipulated in the legislation of this program as a resource to be
considered at risk. Populations affected by cropland wind erosion and EPA particulate matter
(PM-10) nonattainment areas (July 1996) were represented on the maps.
The main difference between the CRP analysis and the EQIP analysis is the reference
point for presentation of some of the data. The CRP analysis was based, in places, on evaluating
the potential impacts as if there had not been a CRP in place for more than 10 years. For
example, in a discussion of cropland erosion rates, the analysis stated, "Without enrollment of
acreage in CRP, about 145 million acres would have eroded in excess of T and of the 2.14 billion
tons of soil that would have eroded about 1.1 billion would have exceeded the sustainable T rate.
With CRP enrollment, over one-third of the United States cropland, 131 million acres, is eroding
at an average annual rate greater than T." The "T" rate is defined as the maximum erosion rate
that can occur while allowing a soil to indefinitely sustain a high level of crop production.
6.3,1.2.4. Risk characterization. The assessment team intended to present information that
would be useful in making decisions about the identity and location of the type of cropped
acreage that should receive priority for enrollment in CRP. The principal contribution of the risk
assessment was to present and combine information that will allow national-level policy makers
to generally target the situations and areas where participation in the program is most likely to
address environmental degradation. With this information, national-level policy makers can
work with the States and localities in these areas to refine the application of CRP activities
toward solutions to the environmental problems.
Time scales for natural resource recovery as a result of program actions were addressed in
similar fashion to the EQIP risk assessment. However, recovery was estimated to occur much
faster because of the almost complete cessation of the production activities creating the
environmental stressprs. A chronicle of uncertainties associated with a risk assessment of this
type also was presented in similar fashion to that of the EQIP risk assessment.
Discussion centered on the topics of erosion-related impacts, wildlife habitat, fertilizer
and pesticide application, and wetlands. Under erosion-related impacts, reference was made to
the extent to which CRP has already contributed to erosion reduction and the impacts if CRP
lands are returned to production. The amount (tons) of sediment delivered to water bodies as a
result of erosion and of airborne soil from wind erosion was presented and discussed. The
6-18
-------
discussion of wildlife habitat was based on 10 geographic regions identified as priority areas hi a
1995 U.S. Congress Office of Technology Assessment report. Additional agriculturally related
geographic regions also were identified and discussed.
Patterns of fertilizer and pesticide use, areas for their potential impacts, and estimates of
reductions in their use as a result of the previous CRP signups were presented. The same NRI
maps associated with fertilizer and pesticide applications that were Included in the analysis
section of the EQIP risk assessment were presented hi this risk characterization. Finally, a
discussion of the location and acreage of cropped wetlands was presented.
In stark difference with the EQIP risk assessment, no direct recommendations to risk
managers were made in this assessment.
6.3.2. Report on the Ecological Impacts of Nonindigemous Shrimp Viruses
This section presents an evaluation of potential shrimp virus impacts on wild shrimp
populations in the Gulf of Mexico and southeastern U.S. Atlantic coastal waters. In a
preliminary report to the Joint Subcommittee on Aquaculture, ithe Shrimp Virus Work Group
(with members from the Department of Commerce, USDA, EPA, and Department of the Interior)
used EPA's framework report (U.S. EPA, 1992) to conduct a preliminary analysis of the
potential problem and to identify optional plans for performing a complete risk assessment.
Although the risk assessment has not been performed, this case study does illustrate another way
to use the framework report to aid risk managers (Shrimp Virus Work Group, 1997.)
6.3.2.1. Background '.
Nonindigenous shrimp viruses may be a threat to the sustainability of U.S. marine
resources. New highly virulent diseases have been found in foreign shrimp aquaculture facilities,
and the United States has greatly increased importation of shrhnp produced hi these facilities.
The viruses pose no threat to human health, but there have been catastrophic disease outbreaks
with 50% to 95% loss rates on U.S. shrimp farms as well as diseased shrimp found in
commerce. Also, new information on the susceptibility of wild shrimp and other crustaceans to
the virus has come to light. Shrimp harvesting and processing in the United States is a $3-
billion-a-year industry. As a result, there have been calls for a risk assessment of the potential
threat of the viruses to marine resources. ;
The Shrimp Virus Work Group was formed by the Joint Subcommittee on Aquaculture to
assess potential risks. The work group used the risk assessment framework to organize existing
information, determine the need for a formal assessment, and formulate options for performing
an assessment. The problem formulation step was used to develop a conceptual model, develop
6-19
-------
an overview of economic impacts, list the potential effects of the viruses on shrimp and other
aquatic species, summarize the basic life history of shrimp, identify potential stressors affecting
the shrimp population, and identify potential pathways for the exposure of wild shrimp to the
viruses.
After using the framework as a tool for organizing information, the work group proposed
^^ options for doing an assessment. It also recommended that a formal ecological risk
assessment be done to provide information needed to address international tirade issues, national
and State regulatory obligations, and the needs of industry, environmental groups, and the public.
6.3.2.2. Management Goals
The process was initiated with the risk assessors and managers agreeing on the scope of
the potential assessment and setting management goals. The goal of the analysis was to provide
information to help prevent the establishment of new disease-causing viruses in wild populations
of shrimp in the Gulf of Mexico and southeastern U.S. Atlantic coastal waters, while minimizing
possible impacts on shrimp importation, processing, and aquaculture operations.
• . • . i • . • .• • i (
6.3.2.3. Problem Formulation
The work group performed the three steps of problem formulation: (1) define assessment
endpoints, (2) develop the conceptual model, and (3) develop an analysis plan.
6.3.2.3.1. Assessment endpoints. In identifying potential assessment endpoints, the work group
focused on linking the management goal with the environmental values to be protected. The
primary assessment endpoint selected is the survival, growth, and reproduction of wild penaeid
shrimp populations in the Gulf of Mexico and southeastern U.S. Atlantic coastal waters. The
focus was on the wild penaeid shrimp in and around the Gulf because of the societal and
ecological importance of these shrimp populations and their known susceptibility to the stressors
in question (the viruses).
A secondary endpoint is the ecological structure and function of the coast and near-shore
marine communities as they affect wild penaeid shrimp populations. This endpoint was selected
because the shrimp population cannot be protected without considering the ecological system it
inhabits. For example, other crustaceans such as copepods, amphipods, and crabs share habitat
with the shrimp during key stages of the shrimp life cycle. The other crustaceans may be
alternative hosts for the viruses and serve as a potential reservoir and vector for transmission.
6-20
-------
6.3.2.3.2. Conceptual models. Developing the conceptual model aids the risk assessor in
formulating the risk hypotheses that v/ill be evaluated during the assessment. The modeling
process was used by the work group to identity the most significant linkages among human
activities, stressors, and the assessment endpoints.
Diagrams were used to communicate important pathways in a clear and concise way and
to identify major sources of uncertainty. The two major pathways for the imported viruses to
enter the domestic ecosystems were identified as aquaculture and shrimp processing. Sources of
aquaculture infection include contaminated feed, broad stock, transport vehicles and containers,
and bird and animal transport. The aquaculture shrimp can then infect native populations
through escapement, pond flooding, sediment and solid waste disposal, etc. Processing plants
are generally located on waterways, and processing wastes are generally discarded directly into
the adjacent water. Other sources of infection include infected bait shrimp, ship ballast water,
nonshrimp translocated animals, and natural spread.
The modeling process considered a variety of stressors but focused on four particularly
virulent species of nonindigenous viruses found in imported shrimp but not yet detected in native
U.S. shrimp. Other anthropogenic stressors such as harvesting, contaminants (e.g., organic
matter that lowers dissolved oxygen), and habitat destruction also were considered hi the model.
Environmental stressors, including temperature, salinity, and jpredation, were considered as well
because they affect shrimp population dynamics. Shrimp exposure to these stressors was
evaluated at the various stages of the shrimp's life cycle.
Direct and indirect viral effects included in the model were individual shrimp mortality
and population effects, as well as effects on other species and indirect ecological effects. The
indirect ecological effects could include changes in ecological structure (species composition)
and ecological function (predator/prey relationships, competition for niches and habitat, and
nutrient cycling).
The modeling process confirmed the complexity of the socioeconomic and natural system
being assessed and revealed numerous data gaps. Twenty-seven significant data gaps were
identified, and most of these will not be easily filled. Examples include (1) shrimp population
models that adequately explain variability of wild populations; (2) distribution and effects of
viruses on nonshrimp organisms; (3) distribution and genetic! diversity of offshore populations;
(4) concentrations, frequency, duration, location, and environmental media of the viruses; (5)
evaluation of the effects of interactions among multiple stressors; and (6) number and size of
U.S. aquaculture operations in relation to receiving waters harboring native shrimp.
6-21
-------
,' 1 ',;;;;:', ,;;;,:,:,„,:,,
6.3.2.3.3. Analysis plan. The analysis plan evaluates risk hypotheses and summarizes the
assessment design, data needs, measures, and methods for conducting the analysis phase of the
risk assessment In a complex assessment such as the potential shrimp assessment, the plan
should identify (1) the pathways most important to the exposure and specify the relationships
most critical to evaluating risks; (2) the measures of effects, exposure, and ecosystem
characteristics to evaluate; and (3) how to address data gaps. In this case study, the work group
did not perform a formal assessment. Instead, it laid out a brief plan with options for doing an
assessment in the future.
6.3.2.4. Analysis and Risk Characterization
This phase consists of two activities: characterization of exposure and characterization of
risk, The work group identified 15 considerations that should be included in any future exposure
characterization and 9 considerations for a risk characterization. Neither set of characterizations
has been done.
The risk characterization is the final phase of the risk assessment. Although the work
group did not perform this phase, it indicated that confidence in the results of a future assessment
Could be strengthened if there were agreement between several different lines of evidence. It
recommended pursuing several Ikies of evidence on exposure pathways: (1) laboratory bioassay;
(2) viral outbreaks in aquaculture; (3) effects, or lack of effects, of viral exposure in wild
populations; and (4) predicted effects based on exposure scenarios.
:|
6.3.2.5. Summary
The work group's summary focused on information gathered on exposure and ecological
risks that could be assessed from available information, and on its list of data gaps and research
needs. Two l?;ey pieces of information were that some countries knowingly export infected
shrimp, and that despite extensive efforts by the U.S. Marine Shrimp Farm Program, State
agencies, and producers to prevent viral outbreaks, there have been numerous disease outbreaks
on U.S. shrimp farms in recent years. Therefore, there is reason to take seriously the possibility
of the wild population becoming infected.
The work group concluded that proceeding with a full risk assessment at this time would
I, i! . ill , ' ' " :! '!. •;! ' , il. i Hi"": . ' !if ij '
result in a high level of uncertainty because of the many data gaps. However, there may not be
time to do the research needed to reduce the data gaps because of the nature of the potential risk.
The work group identified three assessment options: a quick qualitative risk assessment, a longer
t|rm quantitative risk assessment, and a tiered assessment. The tiered assessment would start
with the qualitative assessment and then refine it as data become available.
6-22
„ I,,,::::::1 ,*"'',:
-------
6.4. RISK ASSESSMENT METHODOLOGY DEVELOPMENT
The shrimp virus case study demonstrates that the framework report (U.S. EPA, 1992) is
useful even for preliminary presentation and identification of a potential problem. The available
data can be arranged in such a way that formal risk assessment follows in a prescribed and
logical fashion. In the case study report, information and data needs were enumerated, and
scenarios for conducting the assessment under different sets of available data were discussed.
However, this case study is an example of the single-stressor (virus types), single-receptor (wild
shrimp species) risk assessment that was fully developed in EPA's framework and guidelines
reports (U.S. EPA, 1992,1996).
EPA's framework and guidelines reports provided a basic methodology for developing
the risk assessments for the conservation programs. However;, the reports did not fully develop a
protocol for conducting risk assessment on multiple physical and chemical stressors at the scope
and scale necessary for these programs. The methodology needs to be modified so that risk
assessment guidelines can be developed for future conservation and other agroecosystem
programs. The EQIP and CRP risk aissessments focused, heavily on the problem formulation
phase of the process. Quantitative analysis of ecological effects and exposure was beyond the
scope of the assessments, given the time and financial resource constraints. It is unlikely that
there will be sufficient data, time, or finances in the near future to conduct such an empirical
analysis for these broad-based, nationwide environmental programs. The immediate goals
should be: ;
• Develop a basic methodology and set of guidelines for developing agroecosystem risk
assessment. :
• Identify sources of available data useful for agroecosystem risk assessments.
• Develop a process for directing research activities at governmental or nongovernmental
institutions for the production or collection of relevant environmental data.
« Develop a process for directing research activities at governmental or nongovernmental
institutions for the development of analytical models to support use of data in ecological
risk assessment. •
• Encourage interagency and rnultidisciplrnary collaborative efforts in data collection,
baseline assessment, and environmental monitoring to support risk assessment.
• Encourage interagency and multidisciplinary collaborative efforts in conducting
ecological risk assessments on a landscape scale.
6-23
-------
6.5. RISK MANAGEMENT
The challenge to government agencies that conduct risk assessment on agricultural
production is to develop an iterative process early in regulatory development. This includes
identifying when risk assessment will be required, clearly identifying risk management
objectives, and establishing an iterative process between risk assessment and risk management
fpr the course of program development. Agroecosystem risk assessment is a new requirement for
USD A, and it is not possible at this early date to determine how it will affect future regulatory
development It will also take time to determine whether risk assessment will affect future risk
j | • ',;, Si ' ' 1 ; I .. ; ,. , JSi!'1' ' , !} ] ' , ', ' i
management activities based on annual program evaluation. The predicted potential benefits
from the required analyses of the National Environmental Policy Act have suffered from a
process that has resulted hi the preparation of environmental assessments and environmental
impact statements very late in regulatory development. All these tools were designed to aid in
the decision making that occurs during regulatory development, and they are not used to their
fullest when approached hi this way.
It is necessary that there be clearly identified strategies for using these tools so that
programs can minimize adverse ecological impacts while achieving other agricultural goals.
Arguably, using an ecological risk assessment framework hi developing agricultural strategies
would help improve ecological protection. However, ecological risk assessments have had only
limited and recent use in assessing new policies, technologies, and production systems.
Consequently, it is not clear how significant the impact will be on public agricultural decision
making.
Risk management must play a role in the development and use of risk assessments if the
intent is for risk assessment to aid hi the decision-making process of regulatory development:
• Private as well as public risk managers will need to participate in the assessment planning
process because so many groups of stakeholders are associated with agriculture.
• Risk managers will need to make decisions balancing the need for a timely, scientifically
sound and credible risk assessment to support decision making with the time commitment
and monetary investment required to remove data gaps and reduce uncertainty in the
assessment.
• Risk managers will have to balance'often-conflicting goals because of the broad
geographical scope, interdisciplinary complexity, and different statutory guidelines of
agencies participating in the assessments. Implementing and regulatory agencies,
Federal, State, and local governments, and public and private sector activities frequently
have different missions and goals that must be reconciled if assessments are to be useful
and effective.
6-24
-------
6.6. NEXT STEPS
The U.S. Department of Agriculture may wish to address the following recommendations
to further incorporate ecological risk assessment in the management of agricultural ecosystems.
• Ecological risk assessments are a relatively new tool in 'agriculture. Promoting the use of
this tool will require a sustained promotional and training program for risk assessors and
managers in both the public and private sectors.
• Reducing data gaps will require some adjustments in research priorities to put more
emphasis on understanding the structural and functional relationships of agroecosystems.
• Statutes, regulations, and policies of the agencies involved in ecological risk assessments
should be reviewed to identify and remove any barriers that limit the effectiveness of
interagency assessments, particularly when they involve operating and enforcement
agencies. Such barriers could result from conflicting requirements for assessment scope,
data, and methodologies; standards of review; and. the role of the assessment in the
decision-making process.
6.7. REFERENCES
Farm Services Agency (FSA). (1997) Conservation Reserve Program environmental risk assessment. Farm Services
Agency, Washington, DC.
Natural Resources Conservation Service. (1997) Environmental Quality Incentives Program environmental risk
assessment. Natural Resources Conservation Service, Washington, DC.
Shrimp Virus Work Group. (1997) An evaluation of potential shrimp virus impacts on wild shrimp populations in
the Gulf of Mexico and southeastern U.S. Atlantic coastal waters. A preliminary report to the Joint Committee on
Aquaculture, March 24, 1997. ;
U.S. Environmental Protection Agency. (1992) Framework for ecological risk assessment. Risk Assessment Forum,
Office of Research and Development, Washington, DC. EPA/630/R-92/001.
U.S. Environmental Protection Agency. (1993) Ecological risk assessment case study: the National Crop Loss
Assessment Network. In: A review of ecological assessment case studies from a risk assessment perspective. Risk
Assessment Forum, Office of Research and Development, Washington, DC. EPA/630/R-92/005.
U.S. Environmental Protection Agency. (1998, May 14) Guidelines for ecological risk assessment. Federal Register
63(93):26846-26924.
Wywialowski, A. (1996) Wildlife damage to field corn in 1993. Wildl Soc Bull 24(2):264-271.
6-25
-------
-------
7. ENDANGERED/THREATENED SPECIES
7.1. SUMMARY
The determination of the impact of physical, biological,: and chemical stressors on the
survival of a given species fits well within the framework for ecological risk assessment. The
National Research Council (NRC) report on "Science and the Endangered Species Act" (NRC,
1995a) states that "the concept of risk is central to the implementation of the Endangered Species
Act." It calls for enhanced use of biologically explicit quantitative models and evaluation of
multiple stressors for risk assessment when evaluating endangered and threatened populations.
The evaluation of the likelihood of extinction of a species fits within the definition of risk
assessment. However, this chapter focuses on the analysis phajse of an ecological risk
assessment; the full adaptation of the ecological risk assessment framework to this topic has not
yet been done. ;
The other phases of ecological risk assessment are not as clearly addressed in this chapter.
Problem formulation could be considered to be largely statute driven, with the question "What
are we trying to protect?" clearly stated to be endangered and threatened species and the means to
conserve the ecosystem upon which they depend. In the risk characterization, addressed hi part
in Section 7.3.1.3, the likelihood of extinction within a given time frame and the uncertainties of
the risk of extinction are discussed. The impacts of future events are also considered.
This chapter focuses on the usie of specific modeling tools for estimating the risks of
extinction of endangered or threatened populations. Population biology parameters that
influence the probability of extinction include random demographic or environmental changes,
loss of adaptive variation, environmental catastrophes, accumulation of deleterious genetic
factors, and habitat fragmentation. The effects are determined; by collection of scientific and
commercial data on population numbers and rate of decline. Population models are used to
produce information on population stability, time to extinction, and even time to recovery—if the
stressors are removed. Life history models can identify stages of an animals life where it is most
susceptible to impacts from stressors.
A sound scientific process and. peer review are critical to ensure that a consensus
scientific product is put forward to decision makers. It is clear that various components of
ecological risk assessment are used in listing of endangered and threatened species. Whether it
will be further applied will depend on the scientists or managers wanting additional information
such as likelihood and consequences from population models, evaluating uncertainty, or
conducting sensitivity analysis.
7-1
-------
7.2. THE ENDANGERED SPECIES ACT OF 1973
When the Endangered Species Act (ESA) was passed in 1973, it represented a bipartisan
response to the decline of many wildlife species around the world. The ESA is regarded as one
of the most comprehensive wildlife conservation laws in the world. Its purposes are to conserve
the ecosystems upon which endangered and threatened species depend and to conserve and
, i I i,:-.;,: '! I
recover listed species. Under the law, species may be listed as either "endangered" or
"threatened." Endangered means that a species is in danger of extinction throughout all or a
significant portion of its range. Threatened means that a species is likely to become endangered
within the foreseeable future. All species of plants and animals, except pest insects, are eligible
for listing as endangered or threatened.
As of April 30,1997,1,081 U.S. species were listed, of which 447 were animals. The list
includes both U.S. and foreign species and covers mammals, birds, reptiles, fishes, snails,
clams/mussels, crustaceans, insects, arachnids, and plants. Groups with the most listed species
are (hi order) plants, birds, fishes, mammals, and clams/mussels.
The law is administered by the U.S. Fish and Wildlife Service of the U.S. Department of
the Interior and the National Marine Fisheries Service of the U.S. Department of Commerce.
The U.S. Fish and Wildlife Service has primary responsibility for terrestrial and freshwater
organisms, while the National Marine Fisheries Service's responsibilities concern mainly marine
species such as salmon and whales.
The 1973 ESA replaced earlier laws enacted in 1966 and 1969 that provided for a list of
endangered species but gave them little meaningful protection. The 1973 law has been
reauthorized seven times and amended on several occasions, most recently in 1988. The ESA
i • • i •" • -I i
was due for reauthorization again in 1993, but legislation to reauthorize it has not yet been
enacted. The ESA has continued to receive appropriations while Congress considers
reauthorization, allowing conservation actions for endangered species to continue. The ESA is a
complex law with a great deal of built-in flexibility. Some basics of the law, including key
terms, are given hi Sections 7.2.1 through 7.2.1.11.
7.2.1. Purpose
When Congress passed the ESA hi 1973, it recognized that many of our Nation's native
plants and animals were in danger of becoming extinct. They further expressed that our rich
natural heritage was of "esthetic, ecological, educational, recreational, and scientific value to our
Nation and its people." The purposes of the ESA are to protect these endangered and threatened
species and to provide a means to conserve the ecosystems upon which they depend.
7-2
-------
7.2.2. Listing
Species are listed on the basis of the best scientific and commercial data available.
Listings are made solely on the basis of the species' biological status and threats to its existence.
The U.S. Fish and Wildlife Service bases all listings on sound science and uses peer review to
ensure the accuracy of the best available data.
7.2.3. Species
The definition of "species" includes any subspecies offish or wildlife or plants, and
distinct population segments of vertebrate fish or wildlife species. This allows for populations of
vertebrate animals to be protected in regions of the country where they are in trouble without
requiring protection in areas where they are doing well. For example, bald eagles are listed as
threatened in the lower 48 States, but are not listed at all in Alaska where they are more
numerous. The Clinton Administratiom has issued new guidelines to clarify the definition of
"distinct population segments" under the ESA.
7.2.4. Candidate Species
The U.S. Fish and Wildlife Service maintains a list of "candidate" species. These are
species for which the Service has enouigh information to warrant proposing them for listing as
endangered or threatened but that have not yet been proposed for listing. The Service works with
States and private partners to carry out conservation actions for candidate species to prevent their
further decline and possibly eliminate Ihe need to list them as endangered or threatened. As of
April 30,1997, there were 182 candidate species.
7.2.5. Recovery
The law's ultimate goal is to "recover" species so they no longer need protection under
the ESA. The ESA provides for recovery plans to be developed, describing the steps needed to
restore a species to health. As of April 30,1997, 653 of the listed U.S. species under the
Service's jurisdiction had approved recovery plans. Appropriate public and private agencies and
institutions and other qualified persons assist in the development and implementation of recovery
plans. Recovery teams may be appointed to develop and implement recovery plans. The Clinton
Administration has issued new guidelines requiring the involvement of interested "stakeholders"
in recovery plans.
7-3
-------
7.2.6. Consultation
The ESA requires Federal agencies to consult with the Service to ensure that the actions
they authorize, fund, or carry out will not jeopardize listed species. In the relatively few cases
where the Service has determined that the proposed action would jeopardize a species, it must
issue a "biological opinion" offering "reasonable and prudent alternatives" about how the
proposed action could be modified to avoid jeopardy to listed species. It is rare that projects are
withdrawn or terminated because of jeopardy to listed species.
7.2.7. Critical Habitat
The ESA provides for designation of "critical habitat" for listed species. Critical habitat
includes geographical areas "on which are found those physical or biological features essential to
the conservation of the species and which may require special management considerations or
protection." Critical habitat may include areas not occupied by the species at the time of listing
but that are essential to the conservation of the species. Critical habitat designations affect only
Federal agency actions or federally funded activities.
7.2.8. International Species
The ESA is the law that implements U.S. participation in the Convention on International
Trade in Endangered Species of Wild Fauna and Flora (CITES), a 130-nation agreement
designed to prevent species from becoming endangered or extinct because of international trade.
The law prohibits trade in listed species except under CITES permits.
7.2.9. Exemptions
The law provides a process for exempting development projects from the restrictions of
the ESA. This process permits completion of projects that have been determined to jeopardize
the survival of a listed species, if a Cabinet-level Endangered Species Committee decides the
benefits of the project clearly outweigh the benefits of conserving a species. Since its creation in
1978, the Committee has been called on only three times to make this decision.
7.2.10. Habitat Conservation Plans
'•' _ Ji ' ; i
This provision of the ESA is designed to relieve restrictions on private landowners who
want to develop land inhabited by endangered species. Private landowners who develop and
implement an approved "habitat conservation plan" that provides for conservation of the species
can receive an "incidental take permit" that allows their development project to go forward.
7-4
-------
7.2.11. Definition of "Take"
Section 9 of the ESA makes it unlawful for a person to "take" a listed species. The ESA
states, "The term take means to harass, harm, pursue, hunt, shoot, wound, kill, trap, capture, or
collect or attempt to engage in any such conduct." The Secretary of the Interior, through
regulations, has defined the term "harm" in this passage as "an act which actually kills or injures
wildlife. Such act may include significant habitat modification pr degradation where it actually
kills or injures wildlife by significantly impairing essential behavior patterns, including breeding,
feeding, or sheltering." This regulation, has been in place since |l975 and was amended in 1991
to emphasize that only actual death or injury of a protected animal would constitute a legal
violation.
7.3. ESTIMATING RISK
Sections 7.3 and 7.4 represent the state of the practice on the use of risk assessment in the
ESA. The sections were taken, with permission from the National Research Council (NRC),
from Chapter 7 of the NRC report" Science and the Endangered Species Act" (NRC, 1995a).
The sections have been modified in some places as necessary for clarity within this document.
The concept of risk is central to the implementation of the Endangered Species Act. The
National Academy of Sciences (NAS) committee was asked to review the role of risk in deci-
sions made under the act, review whether different levels of risk apply to different types of
decisions made under the act, and identify practical methods for assessing risk.
Risk is the probability that something (usually a bad outcome) will occur. Risk
assessment aims to estimate the likelihood of a particular (usually bad) outcome occurring. Risk
management is an integrating framework that assesses the likelihood of bad outcomes and
analyzes ways to minimize the risk of bad outcomes, or at least to respond appropriately if they
occur. Many risk assessments follow the framework developed by the National Research
Council to apply to human health (NRC, 1983); an example of la specific risk assessment
framework is the one developed by EPA's Risk Assessment Forum (U.S. EPA, 1992), which
tracks patterns of exposure to harmful substances and responses of ecological systems to these
exposures. The sometimes confusing terminology of risk assessment and some of the issues in
applying risk assessment to ecological systems were describedjby Policansky (1993); further
examples were discussed by the National Research Council (1993).
The main challenges involved in the implementation of the ESA are the risk of extinction
and the risk management issues associated with unnecessary expenditures or curtailment of land
use in the face of substantial uncertainties about the accuracy of estimated risks of extinction and
about future events. Here we consider the problem of estimating the risk of extinction and the
7-5
-------
limitations of pur current ability to estimate this risk. Models are an important tool for analyzing
the consequences of complex processes, because intuition is often not reliable. In some cases,
. ",. „ I'i.ll I I I ' ,: ,il", '! ' I 'I 1 '
the predictions of the models are not precise because information is lacking or because the
underlying processes are not fully understood. They are valuable as guides to research and as
tools for analyzing the comparative effects of various environmental and management scenarios.
7.3.1, Estimating the Risk of Extinction
Since the inception of the ESA, there have been enough developments in conservation
biology, population genetics, and ecological theory that substantial scientific input can be used in
the listing and recovery-planning processes. The following text synthesizes and evaluates the
various approaches and conclusions that have emerged from recent attempts to understand the
vulnerability of small populations to extinction. The material focuses on random changes in
population sizes and hi their structure, changes in genetic variability, environmental fluctuations,
and habitat fragmentation. Additional theoretical and field research is needed to resolve or
reduce uncertainties, but existing analyses give insight into the relative magnitude and possible
scaling of various influential factors hi the extinction process. More thorough and technical
reviews were provided by Dennis et al. (1991), Thompson (1991), and Burgman et al. (1992).
7.3.1.1. Sources of Risk
Habitat loss, effects of introduced species and, hi some cases, overharvesting are almost
always the ultimate causes of species extinction. Decline of populations to a low density makes
them vulnerable to chance events and sets into play the extinction risks outlined below. When
conditions have deteriorated to the point that a wild population cannot maintain a positive growth
rate, no sophisticated risk analysis is required to tell us that extinction is inevitable without
human intervention. Our attention here is focused on cases hi which a population with a positive
capacity for growth in an average year is still vulnerable to chance events that cause short-term
excursions to low densities. Limitations of these approaches are discussed in Section 7.3.1.2.
7.3.1.1.1. Rarifloni demographic changes. Demographic features, such as family size, sex, and
age at mortality vary naturally among individuals. In populations containing more than about
100 individuals, individual variation averages out and has little effect on the dynamics of
population growth. However, hi small populations, random variation hi demographic factors can
occasionally reach such an extreme state that extinction is certain. This can arise, for example, if
i
all members of one sex die before reaching maturity or if all progeny are of the same sex, as was
7-6
-------
the case with the dusky seaside sparrow (Ammodramus maritimus nigrescens) after loss of
habitat led to its population decline.
Substantial effort has been expended to develop general models for predicting the risk to
small populations of extinction due to demographic stochastitity. Several assumptions must be
made about the ways in which populations grow, in particular about the way population growth
rates respond to density. From the standpoint of an endangered species, the simplest conceivable
model assumes that the population has been pushed to its limits—resources (habitat and food
availability) have become so scarce that, on average, the expected number of births in an interval
is the same as the expected number of deaths. In this case, with individual births and deaths
being random, the mean time to extinction for a population stajrting with N individuals is simply
W generations (Leigh, 1981), that is, the time to extinction increases linearly with the population
size.
A more common situation is one in which resources are sufficient to support an average
positive population growth when the population density is belpw a threshold. Due to chance, the
actual growth rate in any generation will deviate somewhat from its expected value, and in the
rare event that the cumulative growth rate realized over several consecutive generations is
sufficiently negative, the population size will be reduced to zero (i.e., extinction will occur).3
All the demographic models discussed in this section assume that all members of the
population are functionally identical. There is no variation based on age or sex; individuals are
assumed to be identical with respect to reproductive and mortality rates. Thus, strictly speaking,
the results apply best to short-lived asexual organisms or to hermaphrodites that synchronously
reproduce toward the end of their life, as do many annual plants and some invertebrates. Models
incorporating age structure, which are appropriate for vertebrates, require information on the
mean and variance of age-specific mortality and fecundity schedules (Lande and Orzack, 1988;
Tuljapurkar, 1989), information that is limited for even the best-studied species in nature.
For species with separate sexes (most vertebrates and many other organisms), another
source of demographic stochasticity can lead to extinction. When the population is small, there
is some probability that all of the offspring produced in a generation will be of the same sex. For
a population at size N, the probability of this event is 2(0.5N),;and the reciprocal of this quantity,
2N-', gives the mean extinction time if sex-ratio fluctuations are the only source of extinction.4
3 With this type of model, the mean time to extinction increases exponentially with the product of the
expected population growth rate at low density, r, and the population carrying capacity, K, where K can be viewed
as the number of individuals that a reserve can sustain at stable density.
"The derivation of this relationship is as follows: the probability that an individual is male is 0.5, and the
probability of all individuals being male in a sample of TV individuals is 0,5N. Thus, the probability that the sample
7-7
-------
Unless the population is very small, sex-ratio fluctuations alone are unlikely to cause
extinction. However., if the population birth rate is a function of the number of females, as is
usually the case, sex-ratio fluctuations will generate fluctuations in the population birth rate.
This type of synergism can reduce the mean survival time of a population by orders of magnitude
relative to expectations from models that ignore sex (Gabriel and Burger, 1992). For example, if
the number of adults the environment can support (K) is less than 25 individuals or so, the mean
time to extinction can be as low as 100 generations, even when the maximum rate of population
growth is quite high.
r|^ie Pre?eding results apply to populations for which the initial density is at the carrying
capacity. When a species is recognized as endangered, however, it usually has declined
dramatically, at which point the recovery goal is to increase the population density to some
4 J.ii''.i ' '„"' , ', ',j ' i • - . .', ,n .| , •
higher sustainable level. Richter-Dyn and Goel (1972) developed a general solution for the mean
extinction time starting from an arbitrary density, again assuming that random fluctuations in
birth and death rates are the only source of extinction risk. Their model is quite flexible in that it
allows for any pattern of density-dependence hi the birth and death rates.
7.3.1.1.2. Random environmental changes. Demographic stochasticity becomes less important
as the density of a population increases and individual differences average out; however, this is
not the case when temporal variation In an exogenous factor, such as the weather, influences the
reproductive or survival rates of all individuals in a population simultaneously. Environmental
fluctuations influence different individuals to different degrees, but to this point, the theory has
only been developed for the situation hi which all individuals respond in an identical manner to
4 . '!, • i , ', :• ; 'I I '
environmental change. The discussion below expands on the preceding section by incorporating
environmental as well as demographic stochasticity.
Most models consider the population to be growing with an average growth rate of r per
tf ;; „'."'', ' " • i : , ''«. "I" ' • ';!! I . !
capita per year, and variance in this rate among generations, Ve, is due to environmental
fluctuations. Typically, it is assumed that the variance is independent of population size and that
there is no correlation between the state of the environment in one generation and the next. Such
assumptions are probably rarely fulfilled in natural populations, and violations of them would
most likely enhance the risk of extinction, as when generations of poor growth conditions tend to
, JH'I i,, I? ' ' ' '' | "' '"i '' "1 !
be clustered. These caveats aside, a general prediction of models that incorporate environmental
stochasticity is that the mean extinction time is determined by the ratio r/V—the higher the
consists of either all males or all females is 2(Q.5N), in which case the population goes extinct through its inability to
reproduce.
7-8
-------
average growth rate and the lower the variance, the longer the population is likely to survive.
Moreover, the rate of increase of population longevity with increasing K is much slower when
environmental stochasticity is present than when demographic stochasticity operates alone.
Depending on the magnitude of Ve relative to r, even populations with several hundreds or
thousands of individuals can be vulnerable to environmental stochasticity.
The theory just discussed treats environmental variation as a factor that drives variation in
the intrinsic rate of population growth., r. Although this is certainly likely to be true in many
cases, environmental factors can also define the carrying capacity of a population. Thus, an
alternative approach to the treatment of environmental stochasticity is to let K, as well as r, vary.
Variation in K alone cannot cause extinction, unless the carrying capacity actually declines below
zero. However, K puts a ceiling on the attainable population size, and bottlenecks in K can
magnify the effects of demographic stochasticity by enhancing the variation in the population
growth rate due to the smaller sample of reproductive adults. Only limited work has been done
on these issues (see Roughgarden, 1975; Slatkin, 1978).
7.3.1.1.3. Catastrophes. Catastrophes are extreme forms of environmental variation that
suddenly and unpredictably reduce the population size. To the extent that these events are
determined by the weather, lightning fires, epidemics, etc., human intervention can do little to
influence their frequency. However, because catastrophes affect most members of a population
to more or less the same extent, it is clear that, on the basis of chance alone, larger populations
will have an increased likelihood of some individuals surviving this kind of event.
Hanson and Tuckwell (1981) and Lande (1993) have considered the time to extinction for
populations exposed to randomly occurring events, each reducing the population size to a
constant fraction of its current size, the former using a logistic: and the latter an exponential
growth model. In these models, there is no demographic or environmental stochasticity of the
kinds noted above. Rather, extinction occurs only when, by chance, a cluster of catastrophes
occurs. Provided the long-run growth rate is positive, the mean extinction time increases
exponentially with the carrying capacity under this model, with the rate of scaling increasing
with the frequency of occurrence and magnitude of catastrophes. Assuming catastrophes act
locally, spatial subdivision of a species provides a simple means of protection against extinction
caused by devastating events.
7.3.1.1.4. Accumulation of deleterious genetic factors. The reduction of a population to a low
density has several negative genetic consequences that can magnify vulnerability to extinction.
Most species harbor far more than enough deleterious recessive genes to kill individuals if they
7-9
-------
were to become completely homozygous (Simmons and Crow, 1977; Charlesworth and
Charlesworth, 1987; Rails et al., 1988; Hedrick and Miller, 1992). This large genetic load is
essentially unavoidable because it is maintained by a deleterious mutation rate of approximately
one per individual per generation (Mukai, 1979; Houle et al., 1992). In large populations,
deleterious genes, particularly lethal genes, have only minor consequences—the frequencies of
most deleterious genes are kept low by natural selection, and their expression is minimal because
they are usually masked in the heterozygous state. This situation can change dramatically in
small populations. During bottlenecks in population size, mildly deleterious genes, previously
kept at low frequency by natural selection, can rise to high frequency by chance. When these
genes become completely fixed (reach a frequency of 100%), a permanent reduction in
population fitness results.5
Although some deleterious genes may be purged from a population early in a population
bottleneck (Templeton and Read, 1984), the continued maintenance of a population at small size
can only magnify the long-term accumulation of mildly deleterious genes. As noted above,
deleterious mutations arise at a rate of about one per individual per generation. Provided the
individual selective effects of these genes are small (on the order of 1/4 Ne or less), they will
accumulate at the genomic mutation rate, causing a decline in mean fitness of approximately s
per generation (Lynch, 1994). Thus, s = 0.025 (as described in footnote 3), a small population
would be expected to experience a roughly 2.5% decline in fitness per generation due to
deleterious mutations alone, and the rate of mutation accumulation declines with increasing
population size. If the effective population size '(Ne) is greater than 1,000, mutation
accumulation is essentially halted for tune scales relevant to endangered species management.
However, if the accumulation of deleterious genes reaches the point at which the net reproductive
rate of individuals is less than 1, the population is incapable of replacing itself. At this point, the
population size begins to decline, and random drift progressively overwhelms natural selection;
consequently, decline in fitness accelerates through the accumulation of deleterious mutations.
•' ,,''!'' • ' , ',',:! !! il
This synergism, whereby the rate of decline in fitness increases with the accumulation of
'Roughly speaking, ifNc is the effective number of breeding adults and s is the selection intensity opposing
a deleterious gene in the homozygous state, then selection is ineffective if4Nes < 1. Typically, because of high
variance in family size, the effective population size is one-third to one-tenth the actual number of breeding adults
(Heywood, 1986; Briscoe et al., 1992). Thus, as a first approximation, if the number of breeding adults is less than
2ts, natural selection will be essentially incapable of eliminating a deleterious gene—its future frequency will be
governed by chance, with the probability of fixation being equal to the initial frequency. The current wisdom is that
s f°rM average mutation is approximately 0.025 (Simmons and Crow, 1977; Houle et al., 1992). Noting that
2/0.025 - 80, this implies that a substantial number of the rare deleterious genes in a population can drift to high
frequency if the number of breeding adults is reduced to 100 or fewer individuals for a prolonged period.
iii
7-10
-------
deleterious genes, has been referred to as a "mutational meltdown" (Lynch and Gabriel, 1990;
Lynch et al., 1993) and, once initiated, can lead to rapid extinction.
7.3.1.1.5. Loss of adaptive variation within populations. Most populations, even those
undisturbed by human activity, are exposed regularly to temporal and spatial variation in
physical and biotic features of the environment. In principle,, some species can cope with such
selective challenges by simply migrating to suitable habitat (Pease et al., 1989). However,
endangered species often live in higMy fragmented habitats with inhospitable barriers; migration
might not be an option. This leaves adaptive evolutionary change, which requires heritable
genetic variation, as the primary means of responding to selective challenges (habitat
degradation, global climatic change, species introductions, etc.) that threaten species with
extinction.
Consider a population that is faced with a gradual change in a critical environmental
factor, such as temperature, humidity, or prey size. If the rate of change is sufficiently slow and
the amount of genetic variance for the relevant characters in the population sufficiently high, then
the population will be able to evolve slowly in response to the environmental change, without a
major reduction in population size. If the rate of environmental change is too high, the selective
load (reduced viability and fecundity) on the population will exceed'the population's capacity to
maintain a positive rate of growth, and although the population might respond evolutionarily, it
will become extinct in the process. Thus, for any population, there must be a critical rate of
environmental change that allows the population to evolve just fast enough to maintain a stable
size. Lynch and Lande (1993) showed that this critical rate is directly proportional to the genetic
variance for the character upon which selection is acting.
Several factors influence standing levels of genetic variation for characters associated
with morphology, physiology, and behavior. Most forms of natural selection cause a reduction
in the genetic variance by eliminating extreme genotypes, the exact amount depending on the
intensity of selection. Small populations also lose an expected l/2Ne of their genetic variance
each generation because of the chance loss of some genes by random genetic drift. Mutation
adds genetic variation to each generation of a population. When populations are kept at a
constant size and under constant selective pressures, they ultimately evolve an equilibrium level
of genetic variance, at which point the loss due to selection and drift is balanced by mutational
input. i
For large populations, the magnitude of this equilibrium variation is debatable, because it
depends on the gametic mutation rate and the distribution of mutational effects, neither of which
are very well understood (Barton and Turelli, 1989). Howevpr, for populations with effective
7-11
-------
sizes of a few hundred or fewer individuals, the expected amount of variation for a typical
quantitative character is nearly independent of the strength of selection and proportional to the
product of the effective population size and the rate of mutational input of variation (Burger et
,:,•»; •:,',, , ' , • ,! • '!!"!!•, i ,ii i ,
al., 1989; Foley, 1992). This implies that for populations containing hundreds or fewer
individuals, the rate of environmental change that can be sustained for a prolonged period is
directly proportional to the effective population size. In other words, a doubling in population
size effectively doubles the evolutionary potential of the population.
Some attempts to identify a critical minimum population size for captive populations
from a genetic perspective have focused on goals such as the maintenance of 90% of the genetic
variation present in the ancestral (predisturbance) population for 200 years (Franklin, 1980;
Soule et al., 1986). Goals of this nature take into consideration the fact that populations that are
dwindling hi size cannot be in equilibrium. However, these goals are rather arbitrary with
respect to choice of acceptable loss and time span. For long-term planning, an alternative
approach is to consider that above a certain effective population size, the dynamics of genetic
variation are influenced predominantly by selection and mutation, so that any further increase in
the effective population size would not significantly influence the amount of genetic variation
;'T " ,' ! ' i' ', ' ., ill I
maintained in the population. Based on the above arguments and because the effective
: ! , l , ' ' ' 'I..',"' '! i '
population size is generally several-fold less than the actual number of breeding adults
(Heywood, 1986; Briscoe et al., 1992), populations must have about 1,000 individuals to
maintain their genetic variation.6
7.3.1.1.6. Habitat fragmentation. A major area of uncertainty in conservation biology concerns
the degree to which population subdivision influences the vulnerability of species to extinction.
Even for fairly simple, single-factor investigations hi which demographic or environmental
sources of randomness are assumed to dominate (Quinn and Hastings, 1987,1988; Gilpin, 1988),
the debate about the effectiveness of a single large reserve as opposed to several small ones is far
from being resolved. An advantage of a single large reserve is that it is buffered from
demographic stochasticity, but multiple small reserves can buffer an entire species from
extinction due to local catastrophes and environmental stochasticity. On the other hand, small
isolated populations are precisely the ones that are expected to suffer from inbreeding depression,
mutation load, and loss of adaptive potential. Much of the recent theoretical and empirical work
The actual number depends in part on the biology of the organisms involved, such as sex ratio, breeding
behavior, and so on. It can be greater than 1,000 if the effective population size is much smaller than the actual
population size.
7-12
.•ttli, ii's'll ;
U , liiiiiiii.:, ,1!:;: I
-------
on the dynamics of populations with EL metapopulation structure can be found in recent volumes
by Gilpin and Hanski (1991) and Burgman et al. (1992). •
Population subdivision adds another dimension to species viability analysis, because
questions are focused not just on the risk of extinction for an individual deme, but for an entire
complex of demes. Levins (1970) called a collection of partially or totally isolated populations
of the same species a metapopulation, and his early models for site occupancy form the
conceptual basis of most current efforts hi this area. Levins showed that in an ideal world
consisting of an effectively infinite number of subpopulations; each with a constant probability
of extinction E and a recolonization rate C, the entire metapopulation will eventually reach an
equilibrium with a fraction 1 - E/C of the total sites occupied. jBecause of the randomness of
extinction and colonization, the specific sites that are occupied will vary in time.
The intuitive notion behind Levins's work is that unless the extinction rate is zero, the
total amount of suitable habitat for a species is unlikely ever to be completely occupied.
Elimination of suitable but unoccupied patches of habitat reduces the recolonization rate by
making it more difficult for migrants to find suitable sites. Thus, habitat removal could
theoretically have the paradoxical effect of increasing the fraction of apparently suitable habitat
that is unoccupied, but this is due only to an overall decline hi metapopulation size.
Lande (1987) introduced a series of habitat-occupancy models showing that if suitable
patches are dispersed to a large enough degree that migrants are unlikely to find them, the local
extinction rate will exceed the colonization rate. Thus, there exists a minimum fraction of the
total landscape throughout a region mat must be suitable for a species to persist. These
extinction thresholds, defined by the demographic and dispersal properties of the species,
demonstrate that locally abundant species can sometimes be very close to extinction if the
proportion of suitable habitat is near the extinction threshold. This again emphasizes that
population size alone is not always a good indicator of vulnerability to extinction.
Lande's (1987) models are idealized in that they envision a world consisting of two kinds
of habitat patches—hospitable and inhospitable, all of equal size. The real world, of course, is
more complex. Patches differ in size; and shape, patch quality is usually a continuous variable,
and some patches are connected by corridors, others not at all (see NRC, 1995a, chapter 5).
More generalized approaches are discussed by Ak?akaya and:Ginzburg (1991). A significant
feature of their approach is the inclusion of a correlation between the extinction probabilities of
adjacent patches. This correlation, if positive, causes a reduction in the expected time to
extinction. In other words, if all patches hi an area became inhospitable at the same time, there
would be no refuges available.
7-13
-------
For many species, the adverse consequences of habitat fragmentation are not caused so
much by a loss of total area as by changes in the quality of habitat due to the development of
edge effects on the margins of reserves (Lovejoy et al., 1986). Edge effects range from
microclimatic changes resulting from structural changes in the environment to major alterations
hi the vegetational community to invasions by exotic species from agricultural and urban
settings. The complete impact of edge effects may require several years to develop and may
ultimately extend for several kilometers beyond the edge of the reserve. Some attempts have
been made to capture the key features of edge effects in mathematical models (Cantrell and
Cosner, 1991,1993). The issues are very complex because they involve interspecific
interactions, such as competition between reserve and invading species. Ultimately, the practical
application of any of these models requires a deep understanding of the ecology of the species
under consideration.
7.3.1.1.7. Supplementation. An increasingly common strategy for maintaining wild populations
of endangered species is augmentation with stock from breeding facilities, as in the case of
hatcheries for Pacific salmonids. An implicit assumption of such procedures is that recipient
populations, when they still exist, actually derive some benefit from an artificial boost in
population size. There are, however, several reasons why long-term deleterious consequences of
supplementation may outweigh the short-term advantage of increased population size.
First, over evolutionary time, successful populations are expected to become
morphologically, physiologically, and behaviorally adapted to their local environments. Thus,
the introduction of normative stock has the potential to disrupt adaptations that are specific to the
local habitat. This type of problem takes on added significance when the population employed in
stocking has been maintained in captivity. Captive environments are often radically different
from those in the wild, and over a period of several generations, "domestication selection" can
potentially lead to the evolution of rather different behavioral or morphological phenotypes
(Doyle and Hunte? 1981; Frankham and Loebel, 1992; NRC, 1995b)—genotypes that perform
well in the captive environment are expected to gradually displace those that do not.
Furthermore, an overly protective captive breeding program may simply result in a relaxation of
natural selection and the gradual accumulation of deleterious genes. For hatchery salmonids,
egg-to-smolt survivorship is typically 50% or greater, as compared with 10% or less in natural
populations (Wapies, 1991; NRC, 1995b).
Second, local gene pools can be co-adapted intrinsically (Templeton, 1986). Just as the
external environment molds the evolution of local adaptations by natural selection, the internal
genetic environment of individuals is expected to lead to the evolution of local complexes of
7-14
-------
genes that interact in a mutually favorable manner. The particular gene combinations that evolve
in any local population will be largely fortuitous, depending in the long run on the chance
variants that mutation provides for natural selection. The breakup of co-adapted gene complexes
by hybridization can lead to the production of individuals that have lower fitness than either
parental type (outbreeding depression) and takes its extreme fo;rm in crosses between true
biological species that cannot produce viable progeny. However, outbreeding depression can
even occur between populations that appear to be adapted to identical extrinsic environments.
The most dramatic evidence comes from reduced fitness in crosses of inbred lines of flies
(Templeton et al., 1976) and plants (Parker, 1992), but crosses between outbreeding plants
separated by several tens of meters can exhibit reduced fitness (Waser and Price, 1989), as can
crosses between fish derived from different sites in the same drainage basin (Leberg, 1993).
Outbreeding depression in response to stock transfer is a major concern in the management of
Pacific salmon, which are subdivided into demes that are home to specific breeding grounds
(Waples, 1991; Hard et al., 1992; NRC, 1995b).
Third, augmentation of wild populations with stock from captive breeding programs can
have negative ecological or behavioral consequences. Unlike genetic effects, which can take
several generations to emerge fully, ecological and behavioral'effects can be immediate. For
example, high-density hatchery populations offish are prone to epidemics involving diseases that
are uncommon in the natural environment. Such events provide strong selection for
disease-resistant varieties of hatchery-reared fish, which subsequently can act as vectors to the
wild population. The Norwegian Atlantic salmon is now threatened with extinction resulting
from a parasite brought to Atlantic drainages by resistant stock from the Baltic (Johnsen and
Jensen, 1986).
Fourth, if a wild population is small because of habitat loss or alteration, the increased
population density that results from augmentation can increase competition for food, space, or
whatever else the habitat provides. That competition can further reduce the size of the wild
population. Harvest of augmented wild populations (particularly if harvest levels are based on
total population) can reduce the wild segment of the population unless the harvest effort is
directed away from the wild population. A captive breeding and reintroduction program is
appropriate only when there is no alternative means of ensuring short-term population viability
or when there is strong evidence of historical gene flow. Habitat loss and degradation are the
main reasons species become threatened or endangered; therefore, the protection of habitat plays
a greater role in preserving these species than captive breeding and reintroduction. For example,
as of 1991, the species specialist groups of the International Union for the Conservation of
Nature, which are international groups of scientists with expertise on specific kinds of animals,
7-15
-------
had completed conservation plans for 1,370 mammals. Of the recommendations in these plans,
51,7 concern protecting or managing habitat, while only 19 concern captive breeding and
reintroduction (Stuart, 1991).
Captive breeding and reintroduction are appropriate when suitable unoccupied habitat
exists and the factors leading to extirpation of the species from this habitat have been identified
and reduced or eliminated. Under these circumstances, captive breeding and reintroduction of
threatened and endangered species can be part of a comprehensive strategy that also addresses
the problems affecting species in the wild (Foose, 1989; Povilitis, 1990; Ballou, 1992; NRC,
1992a). For example, captive breeding and reintroduction enabled the peregrine falcon (Falco
peregrinus) to repopulate much of North America after the use of DDT was eliminated (Cade,
1990). Similarly, Arabian oryx (Oryx leucoryx) were successfully reintroduced in several areas
of their original range where hunting was prohibited (Stanley-Price, 1989).
Captive breeding and reintroduction programs should be avoided when possible;
however, once the need for such a program has been identified, it is advisable to initiate it as
soon as possible. Starting the program before the wild population has been reduced to a mere
handful of individuals increases a program's chances of success. Starting sooner provides time
to solve husbandry problems, increases the likelihood that enough wild individuals can be
j
captured to give the new captive population a secure genetic and demographic foundation, and
minimizes adverse effects of removing individuals from the wild population.
Captive breeding and reintroduction programs are the most expensive forms of wildlife
management (Conway, 1986; Kleiman, 1989) and involve research and management actions.
Although genetic and demographic management techniques for captive populations are fairly
well developed and can be applied to most species (Ballou, 1992; Rails and Ballou, 1992),
husbandry and reintroduction techniques tend to be species specific. Zoos do not know how to
breed many species, such as cheetahs (Actinomyx jubatus), reliably in captivity. In such cases,
expensive and tune-consuming research on genetics, behavior, nutrition, disease, or reproduction
might be necessary to find the reasons for lack of breeding success. The reintroduction of
captive-bred individuals also poses substantial technical challenges. Considerable research, in
captivity and in the field, often is necessary during the early stages of the reintroduction process
to develop successful techniques (Kleiman, 1989; Stanley-Price, 1991).
7.3.1.2. Focusing Conservation Efforts
Life-history models can also help to identify the stages of an organism's life history most
likely to be sensitive to conservation efforts. For example, NRC (1992b) concluded from
life-history data and models that protecting juvenile and subadult sea turtles would have a greater
7-16
-------
effect on increasing population growth than reducing human-caused deaths of eggs and
hatchlings. Similarly, by performing ain analysis of the sensitivity of the population growth rate
of the northern spotted owl to various demographic parameters!, Lande (1988), based on the data
available then, concluded that the most important contributors to the owl's survival were the
adults' annual survival rate, followed by the survival rate of juveniles during their dispersal
phase, and annual fecundity.
7.3.1.3. Distribution of Extinction Times \
The preceding discussion summarizes the state of our knowledge of how various factors
contribute to the risk of population extinction. For practical reiasons, the existing theory focuses
almost entirely on the expected time to extinction. However, in the listing and management of
endangered species, the primary focus is usually on the likelihpod of extinction within a given
time frame (Shaffer, 1981, 1987; Mace and Lande, 1991). Risk analysis requires information on
the dispersion of the probability distribution of extinction times about the mean. For the models
previously cited and many others (Burgman et al., 1992), the distribution of extinction times
typically is strongly skewed to the right, with the most likely extinction time (the mode) being
substantially less than the mean. In general, it is probably more useful to estimate extinction
probabilities as a function of time for different population sizes than to identify some specific
MVP. '
One conceptually simple way of relating risk to the mean extinction time is to assume
that if the current ecological conditions remain stable, the probability of extinction per generation
also remains stable.7 That cannot be strictly true, even ki a constant environment, because
demographic and genetic sources of stochasticity will ensure that the probability of extinction is
not constant in time. For example, if by chance the population size dwindles, the risk of
extinction will be elevated above the average risk until the population has recovered to its
average size. i
7In this case, the conditional probability of extinction in any generation (given that the population has
survived to that point) is simply the reciprocal of the mean extinction time, i.e., pe = II„ where e is the mean time to
extinction measured in generations. Because the probability that extinction does not occur in (x - 1) consecutive
generations is (1 -£><,)*'', and the probability that those (x - 1) generations are immediately followed by extinction is
pe, the probability of extinction in generation x ispe(l -p,y~l. With this approach, the cumulative probability that
the population will be extinct by generation t can be computed by solving the preceding expression for x = 1 to x = t,
and summing these probabilities. Results in Gabriel and Burger (1992) and Tier and Hanson (1981) suggest that
this approach might provide a good first-order approximation to the distribution of extinction times due to
demographic and environmental stochasticity under a broad range of conditions.
7-17
-------
7.3.2. Limitations of Our Ability To Estimate Risk
We close this section by again emphasizing that the practical utility of any extinction
model depends on the validity of its underlying assumptions. Virtually all work on the
vulnerability to extinction has taken a single-factor approach, under the assumption that this will
at least yield an understanding of how the expected extinction tune scales with population size
when a single factor is operating. Other than analytical and computational simplicity, there
seems to be little justification for this approach to population viability analysis. Chapter 5 in the
1995 NRC report (NRG, 1995a) gives some examples of population viability analyses that have
been useful and points out the need to recognize the uncertainties discussed here. In nature,
populations are exposed to multiple sources of risk simultaneously. Synergism between different
risk factors is not reflected in many models, and therefore the risk of extinction can be
underestimated (see Gabriel and Burger, 1992). A field example of such synergism was
described by Woolfenden and Fitzpatrick (1991); epizootic infections of the Florida scrub jay,
which reduced local populations by 50%, also lowered reproductive success in the following
seasons even after the death rates had returned to normal.
Although analytical results are valuable as guides to research and as methods of
comparing the effects of various environmental and management scenarios, they are probabilistic
in nature, so they often ignore the underlying mechanisms. Perhaps their greatest potential is hi
combination with empirical evidence on extinction times, both in the laboratory and in the field
(see for example Pimm et al., 1993). It remains to be seen how relevant such results are to
natural populations. Most of the work on vulnerability of species has also focused on
nonfragmented populations and, except in the case of asexual populations (Lynch et al., 1993),
few formal attempts have been made to incorporate genetics into extinction models. There is a
clear need for models that predict distributions of extinction times as a function of population
density, demographic rates, mating system, environmental variation, etc. These models, which
can only be evaluated by computer simulation (Shaffer and Samson, 1985; Caswell, 1989;
Menges, 1992), can be expected to advance substantially in the next few years because
computational power is now widely available.
7.4. CONCLUSIONS AND RECOMMENDATIONS
Since the implementation of the ESA, numerous models have been developed for
estimating the risk of extinction for small populations. Although most of these models have
shortcomings, they do provide valuable insights into the potential impacts of various
' „ : ! • i, J: Ulilili ' , i, i '.,'». .Tin1 ,'!,, , , , !| 1
management (or other) activities and of recovery plans. With only a few exceptions, biologically
explicit quantitative models for risk assessment have played only a minor role in decisions
7-18
-------
associated with the ESA. They should play a more central role, especially as guides to research
and as tools for comparing the probable effects of various environmental and management
scenarios.
I
Despite the major advances that have been made in models for predicting mean extinction
tunes, the existing treatments still have substantial limitations. Most of the models are
unifactorial in nature and fail to incorporate the negative synergistic effects that multiple risk
factors have on the time to extinction. Efforts to jointly integrate genetic, demographic, and
environmental stochasticity into spatially explicit frameworks are badly needed.
Most extinction models primarily address the mean extinction tune. Because decisions
associated with endangered species usually are couched in fairly short time frames—less than 100
years—models that predict the cumulative probability of extinction through various time horizons
would have greater practical utility.
Results from population-genetic theory provide the basis for one fairly rigorous
conclusion. Small population sizes usually lead to the loss of genetic variation, especially if the
populations remain small for long periods. If the members of the population do not mate with
each other at random (the case for most natural populations), tiien the effect of small size on loss
of genetic variation is made more severe; the population is said to have a smaller effective size
than its true size. Populations with long-term mean sizes greater than approximately 1,000
breeding adults can be viewed as genetically secure; any further increase hi size would be
unlikely to increase the amount of adaptive variation in a population. If the effective population
size is substantially smaller than actual population size, this conclusion can translate into a goal
for many species for survival of maintaining populations with more than 1,000 mature
individuals per generation, perhaps several thousand in some cases. An appropriate specific
estimate of the number of individuals needed for long-term survival of any particular population
must be based on knowledge of the biology of the organisms involved, such as sex ratios and
breeding behavior. If information on the breeding structure of that species is lacking,
information about a related species might be useful.
7.5. REFERENCES
Ak9akaya, HR; Ginzburg, LR. (1991) Ecological risk analysis for single and multiple populations. In: Species
conservation: a population biological approach. Seitz, A; Loeschcke, V, eds. Basel, Switzerland: Birkhauser, pp.
78-87. '
Ballou, JD. (1992) Genetic and demographic considerations in endangered species captive breeding and
reintroduction programs. In: Wildlife 2001: populations. McCullough, D; JBarrett, R, eds. Barking, UK: Elsevier, pp.
262-275. :
7-19
-------
Barton, NH; Turelli, M. (1989) Evolutionary quantitative genetics: how little do we know? Ann Rev Genet
23:337-370.
Briscoe, DA; Malpica, JM; Robertson, A; et al. (1992) Rapid loss of genetic variation in large captive populations of
Drosophila flies: implications for the genetic management of captive populations. Conserv Biol 6:416-425.
I
Burger, R; Wagner, GP; Stettinger, F. (1989) How much heritable variation can be maintained in finite populations
by a mutation-selection balance? Evolution 43:1748-1766.
11 , i ... , , 'i ; sit, : . >i \
•;, , ' i ',' ' ' , ' ' . '• il, '. 1 il
Burgman, MA; Person, S; Akgakaya, HR. (1992) Risk assessment in conservation biology. New York: Chapman
and Hall.
, .„, , ,r .
Cade, T. (1990) Peregrine falcon recovery. Endang Species Update 8:40-45.
Cantrell, RS; Cosner, C. (1991) The effects of spatial heterogeneity in population dynamics. J Math Biol
29:315-338,
Cantrell, RS; Cosner, C. (1993) Should a park be an island? SIAM J Appl Math 53:219-252.
Caswell, H. (1989) Matrix population models. Sunderland, MA: Sinauer Associates.
•'" ' ":'-';l :•'» • ' . . ' 1 , ,1'!,' : >'\ \
Charlesworth, D; Charlesworth, B. (1987) Inbreeding depression and its evolutionary consequences. Annu Rev Ecol
Syst 18:237-268.
Conway, W. (1986) The practical difficulties and financial implications of endangered species breeding programs.
Int Zoo Yearbook 24/25:210-219.
Dennis, B; Munhplland, PL; Scott, JM. (1991) Estimation of growth and extinction parameters for endangered
species. Ecol Monogr 61:115-143.
Doyle, RW; Hunte, W. (1981) Demography of an estuarine amphipod (Gammarus lawencianus) experimentally
selected for high "r": a model of the genetic effects of environmental change. Can J Fish Aquat Sci 38:1120-1127.
Foley, P. (1992) Small population genetic variability at loci under stabilizing selection. Evolution 46:763-774.
Foose, TJ. (1989) Species survival plans: the role of captive propagation in conservation strategies. In: Conservation
biology and the black-footed ferret. Seal, US; Thome, ET; Bogan, MA; et al., SH, eds. New Haven, CT: Yale
University Press, pp. 210-222.
,:,,', J i ' :
Frankham, R; Loebel, DA. (1992) Modeling problems in conservation genetics using captive Drosophila
populations: rapid genetic adaptation to captivity. Zoo Biol 11:333-342.
Franklin, IR. (1980) Evolutionary changes in small populations. In: Conservation biology: an
evolutionary-ecological perspective. Soute, ME; Wilcox, BA, eds. Sunderland, MA: Sinauer Associates, pp.
135-149.
Gabriel, W; Burger, R. (1992) Survival of small populations under demographic stochasticity. Theor Pop Biol
41:44-71.
Gilpin, ME. (1988) A comment on Quinn and Hastings: extinction in subdivided habitats. Conserv Biol 2:290-292.
Gilpin, ME; Hanski, I, eds. (1991) Metapopulation dynamics: empirical and theoretical investigations. New York:
Academic Press.
7-20
„„, ,,,,i,,,;.,',. „;,,,:„
-------
Hanson, FB; Tuckwell, HC. (1981) Logistic growth with random density independent disasters. Theor Pop Biol
19:1-18. :
Hard, JJ; Jones, RP, Jr; Delman, MR; et al. (1992) Pacific salmon and artificial propagation under the endangered
species act. NOAA tech. memo. NMFS-NWFSC-2. U.S. Department of Commerce, Washington, DC.
Hedrick, PW; Miller, PS. (1992) Conservation genetics: techniques and fundamentals. Ecol Appl 2:30-46.
Heywood, J. (1986) The effect of plant size variation on genetic drift in populations of annuals. Am Naturalist
127:851-861.
Houle, D; Hoffmaster, DK; Assimacopoulos, S; Charlesworth, B. (1992) The genomic mutation rate for fitness in
Drosophila. Nature 359:58-60.
Johnsen, BO; Jensen, AJ. (1986) Infestations of Atlantic salmon, Salmosalar, by Gyrodactylus safaris in Norwegian
rivers. J Fish Biol 29:233-241. '
Kleiman, DG. (1989) Reintroduction of captive animals for conservation. BioScience 39:152-161.
Lande, R. (1987) Extinction thresholds in demographic models of territorial populations. Am Naturalist
130:624-635. ;
Lande, R. (1988) Demographic models of the northern spotted owl (Strix occidentalis caurind). Oecologia
75:601-607. '.
Lande, R. (1993) Risks of population extinction from demographic and environmental stochasticity, and random
catastrophes. Am Naturalist 142:911-927.
Lande, R; Orzack, SH. (1988) Extinction dynamics of age-structured populations in a fluctuating environment. Proc
Natl Acad Sci USA 85:7418-7421.
Leberg, PL. (1993) Strategies for population reintroduction: effects of genetic variability on population growth and
size. Conserv Biol 7:194-199.
Leigh, EG, Jr. (1981) The average lifetime of apopulation in a varying environment. J Theor Biol 90:213-239.
Levins, R. (1970) Extinction. In: Some mathematical questions in biology:. Gerstenhaber, M, ed. Providence, RI:
American Mathematical Society, pp. 77-107.
Lovejoy, TE; Bierregaard, RO, Jr; Rylands, AB; et al. (1986) Edge and other effects of isolation on Amazon forest
fragments. In: Conservation biology: the science of scarcity and diversity. Soule, ME, ed. Sunderland, MA: Srnauer
Associates, pp. 257-285.
Lynch, M. (1994) Neutral models of phenotypic evolution. In: Ecological genetics. Real, L, ed. Princeton, NJ:
Princeton University Press, pp. 86-108. ;
Lynch, M; Gabriel, W. (1990) Mutation load and the survival of small populations. Evolution 44:1725-1737.
Lynch M- Lande, R. (1993) Evolution and extinction in response to environmental change. In: Biotic interactions
and global change. Kareiva, PM; Kingsolver, JG; Huey, RB, eds. Sunderland, MA: Sinauer Associates, pp. 234-250.
Lynch, M; Burger, R; Butcher, D; et al. (1993) The mutational meltdown1 hi asexual populations. J Hered
84:339-344.
7-21
-------
Mace, GM; Lande, R. (1991) Assessing extinction threats: toward a reevaluation of IUCN threatened species
categories. ConservBiol 5:148-157.
Menges, ES. (1992) Stochastic modeling of extinction in plant populations. In: Conservation biology. Fiedler PL-
Jain, SK, eds. New York: Chapman and Hall, pp. 253-276.
Mukai, T. (1979) Polygenic mutations. In: Quantitative genetic variation. Thompson, JN, Jr; Thoday, JN eds New
York: Academic Press, pp. 177-196.
National Research Council. (1983) Risk assessment in the federal government: managing the process. Washington,
DC: National Academy Press.
National Research Council. (1992a) The scientific bases for the preservation of the Hawaiian crow. Washington,
DC: National Academy Press.
i,. " • i .I. .1 " • > • : i i
National Research Council. (1992b) Decline of the sea turtles: causes and prevention. Washington, DC: National
Academy Press.
....... ....... )••
National Research Council. (1993) Issues in risk assessment. Washington, DC: National Academy Press.
National Research Council. (1995a) Science and the Endangered Species Act. Washington, DC: National Academy
JtiCSS*
.., , , . ,|
National Research Council. (1995b) Upstream: salmon and society in the Pacific northwest Washington DC-
National Academy Press. '
Parker, MA. (1992) Outbreeding depression in a selfing annual. Evolution 46:837-841.
Pease, CM; Lande, R; Bull, J. (1989) A model of population growth, dispersal and evolution in a changing
environment. Ecology 70:1657-1664.
|
PinimSL; Diamond, J; Reed, TM; et al. (1993) Times to extinction for small populations of large birds. Proc Natl
Acad Sci USA 90:10871-10875.
, , ..... j
Policansky, D. (1993) Application of ecological knowledge to environmental problems: ecological risk assessment
In: Comparative risk assessment. Cothern, CR, ed. Boca Raton, FL: Lewis Publishers, pp. 37-51.
.,,,,...;,' , . , . •". , '• j '
Poyilitis T (1990) Is captive breeding an appropriate strategy for endangered species conservation? Endang
Species Update 8:20-23. &
Quinn, JF; Hastings, A. (1987) Extinction in subdivided habitats. Conserv Biol 1;198-208.
Quinn, JF; Hastings, A. (1988) Extinction in subdivided habitats: reply to Gilpin. Conserv Biol 2:293-296.
S ^ B™,\JD> (1"2) ManaSmS §enetic diversity in captive breeding and reintroduction programs. Trans
North Am Wildl Nat ResourConf 57:263-282.
Rails, K; Ballou JD; Templeton, A. (1988) Estimates of lethal equivalents and the cost of inbreeding in mammals.
wOnscrv oioi A:lo5-193,
-: ;:; • .: " „ .; .
Richter-Dyn, N; Goel, NS. (1972) On the extinction of a colonizing species. Theor Pop Biol 3:406-433.
' J- (19?5) A Simple m°del f°r P°Pulation dynamics in stochastic environments. Am Naturalist
-736.
7-22
' : :,,i :::
-------
Shaffer, ML. (1981) Minimum population sizes for species conservation. BioScience 31:131-134.
Shaffer, ML. (1987) Minimum viable populations: coping with uncertainty. In: Viable populations for conservation.
Soule", MA, ed. New York: Cambridge University Press, pp. 69-86.
I
Shaffer, ML; Samson, FB. (1985) Population size and extinction: a note on determining critical population sizes.
Am Naturalist 125:144-152. ;
Simmons, MJ; Crow, JF. (1977) Mutations affecting fitness in Drosophila populations. Ann Rev Genet 11:49-78.
Slatkin, M. (1978) The dynamics of a population in a Markovian environment. Ecology 59:249-256.
Soute, ME; Gilpin, M; Conway, N; et al. (1986) The millennium ark: how long a voyage, how many staterooms,
how many passengers? Zoo Biol 5:101-114.
Stanley-Price, MR. (1989) Animal reintroductions: the Arabian oryx in Oman. Cambridge, UK: Cambridge
University Press, 291 pp. ;
Stanley-Price, MR. (1991) A review of mammal reintroductions, and the role of the re-introduction specialist group
of IUCN/SSC. In: Beyond captive breeding: reintroducing endangered mammals to the wild. Gipps, JHW, ed.
Zoological Society of London Symposia 62. Oxford, UK: Clarendon Press, pp. 9-25.
Stuart, SN. (1991) Re-introductions: to what extent are they needed? In: Beyond captive breeding: reintroducing
endangered mammals to the wild. Gipps, JHW, ed. Zoological Society of London Symposia 62. Oxford, UK:
Clarendon Press, pp. 27-37. I
Templeton, AR. (1986) Coadaptation and outbreeding depression. In: Conservation biology: the science of scarcity
and diversity. Soule", ME, ed. Sunderland, MA: Sinauer Associates, pp. 105-116.
Templeton, AR; Read, B. (1984) Factors eliminating inbreeding depression in a captive herd of Speke's gazelle.
Zoo Biol 3:177-199.
Templeton, AR; Sing, CF; Brokaw, B. (1976) The unit of selection in Drosophila mercatorum. I. The interaction of
selection and meiosis in parthenogenetic strains. Genetics 82:349-376. :
Thompson, GG. (1991) Determining minimum viable populations under the Endangered Species Act. NOAA tech.
memo. NMFS F/NWC-198. U.S. Department of Commerce, Washington, DC.
I
Tier, C; Hanson, FB. (1981) Persistence in density dependent stochastic populations. Math Biosci 53:89-117.
Tuljapurkar, S. (1989) An uncertain life: Demography in random environments. Theor Pop Biol 35:227-294.
U.S. Environmental Protection Agency. (1992) Framework for ecological risk assessment. Risk Assessment Forum,
Office of Research and Development, Washington, DC. EPA/630/R-92/Q01.
Waples, RS. (1991) Genetic interactions between hatchery and wild salrrionids; lessons from the Pacific northwest.
Can J Fish Aquat Sci 48:124-133.
Waser, NM; Price, MV. (1989) Optimal outcrossing in Ipomopsis aggregata: seed set and offspring fitness.
Evolution 43:1097-1109.
Woolfenden, GE; Fitzpatrick, J. (1991) Florida scrub jay ecology and conservation. In: Bird population studies:
relevance to conservation and management. Perrins, CM; Leberton, JD; Hirons, JGM, eds. New York: Oxford
University Press, pp. 542-565.
7-23
8-1
-------
I'
. (1996) defines ecosystem
management as "„.management driven by explicit
goals,, executed by policies, protocols, aad practices,
and made adaptable by monitoring snd research, based
on our best understanding o£ ecological interactions
aad processes necessary to stMato, ecosystem
composition, structure and function."
Valley in southwest Virginia, Middle Platte River in south central Nebraska, Middle Snake River
in south central Idaho, and Waquoit Bay on the southern shore of Cape Cod. Each of these
assessments involved an integrated ecosystem approach to making land management decisions.
The role of science in these assessment activities, as in risk management, is to provide
objective information for decision making. That information is framed in a manner that meets
the needs of decision makers, while strictly maintaining scientific objectivity, integrity, and
quality control. Decision maker needs are agreed upon hi an interactive process between
: !!
decision makers and assessment managers. Although new research is seldom performed within
the assessment activity, synthesis of existing information often provides new knowledge or
I ' !|
perspectives about the ecosystems being assessed. In all of the case studies reported hi this
chapter, science played a critical role in
facilitating definition of the pertinent
management questions to be addressed,
establishing information on which to build
the assessments, maintaining quality control
and assurances protocols, analyzing and
synthesizing information, and working to
communicate results to responsible managers and interested parties.
8.2. INTRODUCTION
Ecological risk assessment is a process of organizing and analyzing data, information,
:l
assumptions, and uncertainties to evaluate the likelihood of adverse ecological effects (U.S. EPA,
1996). Ecosystem management is a process for maintaining the integrity of ecosystems over
time and space (Quigley et al., 1996a). Ecosystem sustainability increasingly is being stated as
the goal of ecosystem management. A variety of ecosystem assessments have been led by
Federal agencies in recent years. These assessments were intended to help decision makers and
other interested parties better understand and evaluate consequences of potential regulatory or
natural resource allocation actions within a larger social, economic, and ecological framework.
This chapter provides information on the ongoing development of the ecological risk assessment
process and the ecosystem management assessment process. The linking of these two processes
can bring improved organizational and analytical consistency to the assessment of information in
support of multiple scales of resource planning and decision making needed for ecosystem
management. The intent of this chapter is to promote dialogue between the two communities
and enhance cross-community appreciation of needs and approaches.
8-2
-------
Section 8.2 of the chapter presents an overview of several ecosystem assessments done in
recent years. Section 8.3 provides several Agency case study illustrations of assessment
approaches. Section 8.4 discusses risk assessment methodology development. Section 8.5
examines ecological risk assessment in the ecosystem decision-making context. Section 8.6
discusses possible next steps, beginning with a description of cost-benefit considerations
followed by suggestions for expanded use of the proposed EPA ecological risk assessment
guidelines in ecosystem assessments; the section concludes with an analysis of technical and
research challenges. \
A fundamental challenge to ecosystem management is the need to understand and
manage complex ecosystems simultaneously across large and small temporal and spatial scales
(Quigley et al., 1996a). In light of this challenge, decision makers are faced with making
complex social, economic, and environmental decisions. These decisions bring with them an
inherent level of uncertainty for decision makers and stakeholders alike. Decision makers and
stakeholders need to recognize this inherent uncertainty and be flexible enough to adjust their
decisions in the face of surprise. A general planning model for ecosystem management was put
forward by Quigley et al (1996a) (Figure 8-1). The process has four basic parts: monitoring,
assessments, decision, and implementation.
Ecological risk assessments are tools decision makers can use to help identify and, it is
hoped, reduce uncertainty throughout the decision-making process. Ecosystem assessments are
also tools to help decision-makers. Ecosystem assessments follow general concepts as shown in
Figure 8-1, but they do not have an existing set of definitional rules. The general concepts
include acknowledgment of stakeholders and their questions,'development of situational
analyses, identification of trade-offs and limits, development of an understanding of future
conditions, and assessment of risk for issues of concern. The primary reason for conducting
ecosystem assessments then is to provide a framework for decision makers and stakeholders to
help them understand and evaluate the consequences of actions with respect to regulation and/or
allocation of natural resources within the larger social, economic, and ecological context. The
ecosystem management process presented in Figure 8-1 outlines several sections where
ecological risk assessment can link with ecosystem management. In the assessments section of
the process, assessing risks for issues of concern is presented. In ecological risk assessment, risk
characterization involves addressing the likelihood and consequences, weight of evidence,
uncertainty, and other factors. For determining the impact of a stressor, multiple stressors, or a
management scenario, a qualitative or quantitative analysis of the likelihood and consequences of
the scenario would be valuable for the decision makers. Supporting the "how sure are we of this"
8-3
-------
Monitoring
Monitor biophysical outcomes
Monitor social and economic outcomes
Monitor societal values and goals
Recommend new assessments, new
decisions, and/or new implementation
Monitoring shows
a need for new
assessments or
added information
Monitoring
shows a need for
new decisions
Assessments
Acknowledge stakeholders and their questions
Develop situation analysis of biophysical,
social, and economic systems
Identify trade-offs and limitations
Develop an understanding of future conditions
Assess risk for issues of concern
Decisions
Select management goals
Develop management alternatives
Predict impacts of alternatives
Recommend preferred alternative
Select an alternative
Emerging
issues
Changing
societal values
and goals
New scientific
understanding
Monitoring
shows a need
for changes in
Implementation
Implementation
• Implement decisions on the ground
• Establish partnerships
• Publicize decision, facilitate participation
• Inaugurate adaptive management
I I i, _ 'IU'Wil i 1 '',',! 'II! I '•' it'I1. 'I '' ' ' ill • ' '• ' ' " • V MM 'ilii'.iii i1, i '• I1'1 |i III. 'i 'I I » , ',, : ' . i ''1111 ' ' . .1.
Figure 8-1. Ecosystem management model. Each step has several parts. Because the
model is iterative, external or internal influences can initiate any step in the process, and
the process never ends.
'» „', ,', ,,| ij ; ' '.
Source: Quigley et al., 1996a.
8-4
i,, JBIIiUL,,i,Ji ', i. ..m:,,;;,!.,, J;i,;Jliilili!l i IBlillliillllillJn, Lljiii.
i, i nil'in 1 mi,, i •,;.;, 'i;|, .I
-------
question, uncertainty and lines-of-evidence analyses provide additional information to the
managers. In the decisions section of the ecosystem management process, the prediction of
impacts from alternatives is presented. Prediction of the likelihood and consequences of a
management action on an ecological system can utilize ecological risk assessment methods.
Chapter 4 presents an effective process for nonindigenous species that could be adapted for
multiple stressors or management alternatives. !
It is within this context of ecosystem management, uncertainty, and adaptation that a
series of "lessons learned" workshops, designed as an adaptive learning approach to ecoregional
assessments, are being conducted to discuss and document the knowledge gained by various
assessment teams throughout the country.
The first iteration of ecosystem assessments, which include the Report of the Forest
Ecosystem Management Assessment Team, the Columbia River Basin Assessment, and the
Sierra Nevada Ecosystem Project, were mandated by the President. These were generally high-
cost projects ($6 million to $36 million) directed at a number of high-profile issues in the Pacific
Northwest and Northern California.
Second-generation ecosystem assessments were chartered by decision makers (Forest
Service Regional Foresters) for the purpose of providing state-of-the-art information needed to
revise forest land management plans. These are best represented by the recently completed
Southern Appalachian Assessment, the ongoing Great Lakes Assessment, the Northern Great
Plains Assessment, and the Ozark/Oua:chita Highlands Assessment. These are low-cost
alternatives ($0.5 million to $2 million) to the earlier generation noted above.
Key findings from the "lessons learned" workshops are summarized below (USDA,
1996). ;
• The assessment process
—Assessments are not decision-making documents. However, they do provide a
synthesis of information in support of multiple scales of resource planning and
decision making.
—Assessments should be issue driven.
—Data synthesis and acquisition need to be strongly focused on the assessment issues.
—Preassessment planning is critical to conducting an assessment.
—Process, structure, and function are the ecosystem components evaluated during the
assessment process. These components need to be analyzed at multiple spatial and
temporal scales.
8-5
-------
—Broad-scale assessments are a rich source of new information. Recognizing
emergent properties of ecosystems at broader scales is an important part of this new
information.
Linkages to other assessments and programs
—Inhere is a need to develop implementation, effectiveness, and validation monitoring
programs at multiple scales. These programs should update assessment information
over time.
—Ecoregional assessments can be linked using common information themes and
protocols.
—Cooperation with the Federal Geographic Data Committee will help ensure data
linkages among other national, regional, and landscape assessments.
I
Public involvement and partnerships
ill ; !!!«! " ,"•• . ' . . . .•..'••,• : ii ,| •
—Public participation for ecoregional assessments should be based on adaptive
1 i- /"SI " . •!•'•. ••'•. , ': i- •' •• i • ,' ' '. , .I*. , 1 !:• :' .', ! H 1| i !- .. '.i . ,i| , ,
management principles focused on achieving awareness and active involvement of a
diverse array of stakeholders.
—Public involvement is crucial to the success of assessments and provides benefits in
later decision-making forums.
>£ :':a!1 • ' ' . '' ' ' . • ' '' • ' 'i'"1 •' '• '' >'t' ' •' '!"; " '
—Because of their sheer size and the complexity of ownership patterns, ecoregional
assessments have a greater need for partnerships than any other planning process.
Assessment products
—Assessments produce various tangible and intangible products, including findings,
data, maps, references, changed relationships with participating agencies and the
public, and institutional and organizational change. Products that address immediate
needs and issues are most likely to get immediate use.
,.. .. ,j
I! • I ' '. i • I
Information management
—An interagency commitment needs to be made to ensure maintenance of data, maps,
, 'I •' l,l,,»:1 ' • '! ' ' • '! :,':'. H ' 'I '
iri§ta data, etc., for future assessment and monitoring efforts.
—Establishing an information management infrastructure before the assessment
should be a high priority.
8-6
-------
8.3. CASE STUDIES AND EXAMPLES
8.3.1. Interior Columbia River Basin Scientific Assessment8
The Interior Columbia River Basin Ecosystem Management Project was initiated by the
Forest Service (FS) of the U.S. Department of Agriculture and; the Bureau of Land Management
(BLM) of the U.S. Department of the Interior in response to decisions to adopt an ecosystem-
based management strategy; the need to replace interim direction; concerns about declining
forest, rangeland, and aquatic health; and concerns about single-species approaches to
conservation and management. The project area includes those portions of the Columbia River
Basin within the United States and east of the Cascade crest and portions of the Klamath and
Great Basins in Oregon (the Basin). The primary products called for in the charter include (1) a
framework for ecosystem management (Haynes et al., 1996), (2) an integrated scientific
assessment (Quigley et al., 1996a; Quigley and Arbelbide, 1996), (3) two environmental impact
statements (EISs) addressing management of FS- and BLM-administered lands within the Basin,
and (4) an evaluation of the EIS alternatives (Quigley et al., 1996b). The framework,
assessment, and evaluation of alternatives are products of the science team, and the EISs are
products of the EIS teams. In addition to these primary products, more than 40 scientific
publications are expected from this work over the next several years. The following material is
drawn mostly from the executive summaries of the science documents cited above. The Basin
includes 145 million acres, with the FS and BLM administering more than one-half (76 million
acres) of the area. This sparsely populated area covers portions of 7 States and 100 counties. It
encompasses a variety of climatic, topographic, socioeconomic, forest, and rangeland conditions.
It extends from the Continental Divide on the east to the Cascjade crest on the west. It includes
resources of international significance such as Yellowstone National Park and Hells Canyon. It
is home to some 22 Native American Indian tribes and more than 3 million people.
8.3.1.1. Framework
With the announcement by the FS and BLM of the intent to adopt an ecosystem-based
strategy came the need to frame the interactions among decisions at multiple levels and their
relationship with assessments. The framework assumes that the purpose of ecosystem
management is to maintain the integrity of ecosystems over time and space. It is based on four
ecosystem principles: ecosystems are dynamic, can be viewed as hierarchies with temporal and
spatial dimensions, have limits, and are relatively unpredictable. This approach recognizes that
8This section provides examples of a range of assessments prepared by a number of agencies. The views
expressed represent those of the authors of each assessment summary.
8-7
-------
•..'••.. . • !: il .,|1.'
people are part of ecosystems and that stewardship must be able to resolve tough challenges,
# n i , I •!. ' ..ii !;,i , • - , ; ' ' •','!, .''in li'i;.1] „ i, !'. I ';] ''.I ' i:, , • •!, „
including how to meet multiple demands with finite resources. The framework describes a
general planning model for ecosystem management (Figure 8-1) that has four iterative steps:
monitoring, assessment, decision making, and implementation. Since ecosystems cross
jurisdictional lines, the implementation of the framework depends on partnerships among land
managers, the scientific community, and stakeholders. It proposes that decision making be based
on information provided by the best available science and the most appropriate technologies for
land management.
I ! . • I | ' .
8.3.1.2. Integrated Scientific Assessment
This integrative assessment links landscape, aquatic, terrestrial, social, and economic
characterizations to describe biophysical and social systems. Integration was achieved through
the use of a framework built around six goals for ecosystem management and three different
views of the future. The assessment represents the largest and most comprehensive assessment
of ecosystems undertaken. The overall purpose of the assessment is to develop a better
understanding of the current, historical, and potential future biophysical, economic, and social
conditions and trends in the Basin, The assessment is not a decision document nor does it
", i, ,. • i'.ijii" .'"''i .. •'• : ' i r:•"•' •' ;il. IT . '!
resolve specific resource issues. Rather, the assessment provides the foundation for proposed
additions or changes to existing FS and BLM resource management plans to consistently manage
risks and opportunities at multiple scales. Some highlights of the findings include the following:
• • • •
• There has been a 27% decline in multilayer and a 60% decline in single-layer old-forest
structures from historical levels, predominantly in ponderosa pine and Douglas-fir forest
types.
• Aquatic biodiversity has declined through local extirpations, extinctions, and introduction
of exotic fish species, and the threat to riparian plants and animals has increased.
• Some watershed disturbances, both natural and human induced, have caused and continue
to cause risks to ecological integrity, especially owing to isolation and fragmentation of
fish habitat.
,. ' '-, !'! l . , • U I :
• The threat of severe lethal fires has increased by nearly 20%, predominantly in the dry
and moist forest types.
• Rangeland health and diversity have declined because of exotic species introductions,
historical grazing, changing fire regimes, agricultural conversions of native shrublands
and herblands, and woodland expansion in areas that were once native shrublands and
herblands.
8-8
-------
• Human communities and economies of the Basin have changed and continue to change
rapidly, although rates of change are not uniform.
There are tremendous opportunities to restore ecosystem processes and functions as well
as provide for the flow of goods and services demanded by society. In addition to tremendous
opportunities, risks are also associated with attaining these opportunities. Some risks are related
to natural events such as wildfire, insect, and disease outbreaks, while other risks are associated
with management activities such as road building, timber harvest, and prescribed fire. These
risks and opportunities vary greatly across the Basin. The assessment has characterized the
broad-level risks and opportunities across the Basin. Realizing the opportunities and managing
the risks involves working within the adaptive management framework presented.
8.3.1.3. Ecosystem Integrity
Drawing from the detailed assessment of historical and current conditions within the
Basin, two concepts were used to integrate the major functional areas to determine status of the
ecosystems. Maintaining the integrity of ecosystems is assumed to be the overriding goal of
ecosystem management. The integrity of ecosystems encompasses both social and biophysical
components; the health of the Basin's people and economy is not a separate issue from the health
and integrity of other ecosystem components. Ecological integrity refers to the presence and
functioning of ecological components and processes. The basic components of ecological
integrity include the forest, range, and aquatic systems, with a hydrologic system that overlays
the landscape as a whole. The counterpart to ecological integrity is socioeconomic resiliency
(measured at the county level), which in the context of ecosystem management reflects the
interests of people to maintain well-being through personal arid community transitions.
8.3.1.4. Composite Ecological Integrity
Integrity ratings were developed for five ecological components: forestland, rangeland,
forest and rangeland hydrologic, and aquatic systems. This information became the primary
basis for estimating composite ecological integrity for each subbasin (approximately 850,000
acres in size) within the Basin. Currently, 16% of the Basin is rated as having high relative
composite ecological integrity, 24% as moderate, and 60% as; low. Eighty-four percent of the
systems with high integrity are on FS- and BLM-administered lands, while 39% of the low-
integrity systems are on FS- and BLM-administered lands.
8-9
-------
8.3.1.5. Socioeconomic Resiliency
I
Socioeconomic resiliency, estimated at the county level for this analysis, dealt with the
adaptability of human systems. High ratings imply that these systems are highly adaptable;
changes in one aspect are quickly offset by self-correcting changes in other sectors or aspects.
High levels of Socioeconomic resiliency should reflect communities and economies that are
adaptable to change, where sense of place is recognized in management actions., and where the
mix of goods, functions, and services that society wants from ecosystems is maintained. A low
••• • I
rating applies to 54 Basin counties. Another 20 Basin counties were rated as having an
intermediate level of resiliency. A high Socioeconomic resiliency rating applies to the 26 Basin
counties that are more densely populated. While 68% of the area within the Basin is rated as
haying low Socioeconomic resiliency, 67% of the people of the Basin live in areas with high
Socioeconomic resiliency.
8.3.1.6. Findings From the Future Management Options
The current draft EIS has primarily considered three options: (1) continuation of current
approaches, (2) restoration emphasis, and (3) reserve area emphasis. Evaluations of the options
benefitted from the underlying science documents and assessments conducted in the basin. The
puSic comme^ Per*od on ^ ^^ EIS has closed. Land managers and stakeholders will now
engage in a dialog about the content and process of selection of the preferred strategy for
managing the FS- and BLM-administered lands.
, i
: ' j
8.3.2. The Southern Appalachian Assessment
The Southern Appalachian Assessment (SAA) is an ecological description of conditions
within a region encompassing parts of seven States. The area extends southward from the
Potomac River to northern Georgia and the northeastern corner of Alabama. The SAA assembles
the best available knowledge about the land, air, water, and people of the region. The SAA does
not specifically apply typical risk assessment tools. It does attempt to describe change in the
environment and the stresses that affect it. It is similar to risk assessment in that it avoids
recommending actions.
The recently completed assessment was not the first. Early in the 20th century, the
Appalachian landscape and its natural resources had been badly abused by destructive
agricultural practices and exploitive logging. In 1901, at the request of the U.S. Congress, the
Department of Agriculture conducted a similar assessment for the region. Its findings led to the
Weeks Act, which authorized the establishment of national forests and national parks in the
eastern United States.
8-10
-------
Although there was no specific statutory requirement for the latest assessment, national
forest management plans required by the 1976 National Forest Management Act had been in
place for more than 10 years and needed to be revised. The management of national forests and
other Federal lands is directly influenced by the biological, social, and economic conditions that
surround them. Also, Federal and State regulatory agencies were concerned that increasing
population pressures and economic development were adversely affecting environmental quality
in the region. Thus, there was a need for a comprehensive and credible source of information to
serve as a basis for planning.
Even before the SAA got under way, Federal and State Agencies in the Southern
Appalachian region had worked together on several projects of mutual interest. A coordinating
group had been established, initially to address land management problems, but later expanded to
include most environmental issues within the area. This was the Southern Appalachian Man and
Biosphere (SAMAB) program. SAMAB now includes 12 Federal and 3 State agencies.
Through the coordination of the SAMAB program, most of these agencies were involved in
some way in conducting the SAA. '
The SAA began in the spring of 1994. A dialog that involved SAMAB agencies and
forest planners outlined a number of issues that needed to be addressed. There was no single
issue producing conflict or confrontation, but there was widespread concern for the health and
welfare of the region's resources. Starting with an initial set of issues, a series of public
meetings was held at different locations within the area. People were told about the assessment
that was planned and asked about their concerns and suggestions. The issues and concerns
became the basis for a set of questions that the assessment would address.
The SAA was organized around four major environmental components: air, land, water,
and people. Interagency teams were established to address each of these themes. An initial
evaluation of the data indicated the need for a strong emphasis on map-based geographic
information system technology. An interagency policy group was formed to guide the
assessment. One of the group's first functions was to establish constraints or targets for the time
of completion, money and people available, size of reports, and sources of data. Early hi the
process, it was decided to invite the public to attend and participate in most aspects of the
assessment.
Each of the four major topics making up the assessment culminated in separate technical
reports (atmospheric, aquatic, terrestrial, and social/cultural/economic reports). Although the
analyses differ, the reports have several common features. Each starts with a set of questions that
were derived from the issue identification process. The questions served to guide the analysis
and to define the scope of the assessment. In addition, each interagency team was asked to
8-11
-------
. :
describe the current resource situation and, to the extent possible, look for past and future trends
t nii'i ,i : , Hi, ,h " :, "iii ,i i . • ....... "i, iiiii ..... . • '.'.; in • • .....
in resource condition. Part of the assessment also consisted of evaluating the quality of available
data sources and documenting future research and monitoring needs. The following paragraphs
give a brief summary of each assessment topic.
The atmospheric team concentrated its analysis on nitrogen oxide, sulfur dioxide,
particulate matter, and volatile organic compounds. These pollutants are important because the
secondary pollutants formed from them are suspected of reducing visibility, producing ozone,
and having consequent impacts on vegetation and human health; the pollutants also are important
because of the acid deposition impacts on terrestrial and aquatic environments. In addition, these
are the pollutants directly affected by the Clean Air Act legislation. The report describes the
location of emissions and concentrations where emissions are greatest, and it projects likely
future trends. Visibility is especially important in the S AA analysis because the Clean Air Act
es!*blished a8, a n^onal 8oal t*16 "prevention of any future, and the remedying of any existing,
impairment of visibility in mandatory Class 1 Federal areas where impairment results from man-
..... >!! ,• . •"'" "»! ' ,'•;.'. ',!.:' , • • " : ! ..' '»"-»'. , . ......... r ., VN
made pollution." The majority of the visibility data was obtained in the seven Class 1 areas
within the SAA region.
The terrestrial report is divided into two separate sections: (1) plant and animal resources
and (2) forest health. The report responds to the considerable interest in the status of threatened,
endangered, or sensitive species. Of more than 25,000 species known to inhabit the area, 472
were given special attention. The group includes 51 species that are federally listed as threatened
or endangered and 366 whose numbers are sufficientiy restricted that their populations are
considered at risk. Most of these species can be grouped into 19 associations based on similar
habitat requirements. Historically, the most significant event to affect the region's forests was
the initial logging that was largely accomplished in the early decades of this century. Perhaps
equally profound, although less dramatic, are the effects of a number of forest health factors.
The chestnut blight, gypsy moth, and dogwood anthracnose have altered species composition of
the region's forests. Other recently discovered diseases such as hemlock woolly adelgid and
butternut canker are also cause for concern. Although historic data are inconclusive, it seems
clear that the most serious threats to the health of the region's forests are coming from exotic
pests introduced from other parts of the world.
The headwaters of nine major rivers lie within the boundaries of the Southern
Appalachians, making it the source of drinking water for most of the Southeast. The aquatic
assessment compiled the best available data on water resource status and trends, riparian
condition, impacts of various land management or other human activities, water laws, aquatic
'!;/, . • ''ir'Sh! -' "'!»!: I : ' - •• - "Jt • ' V", :„ . ,|i " I .' ' •! " 'i1,
resource improvement programs, and water uses. The report discusses the distribution of aquatic
8-12
-------
species and identifies some problems., including degraded streams, eutrophication of lakes, and
the impacts of increasing human population and development. There is general agreement,
however, that water quality has improved significantly since the adoption of the Clean Water Act
in 1972.
Humans are a part of the ecosystem. Natural resource values are derived from the utility
and aesthetic or intrinsic benefits that come from human culture. The social/cultural/economic
assessment looked at four aspects of human influence: (1) communities and human influences,
(2) the timber economy, (3) outdoor recreation, and (4) roadless and designated wilderness areas.
The relationship between people and public lands in the Southern Appalachians has changed
greatly during the past two decades. The growing economy has become more diverse and less
dependent on manufacturing. Newcomers to the region, many of them retirees, resort owners, or
those employed in service industries, are more interested in scenery and recreation than in
resource extraction. Also, the increasing population throughout the area is fragmenting land use
and ownership, with adverse effects on wildlife habitat and timber availability. These changes
are reflected in diverse, and often incompatible, demands on public lands. The assessment was
aimed at better understanding the public and how their collective values have changed in recent
years. This should be useful to both land managers and community planners.
The S AA documents consist of four technical reports 'and a summary report. But equally
important are two other products of the assessment. The first is a set of five computer disks (CD-
ROMs) that contain all the maps and data used in the assessment in digital form. These were
distributed to the 400 selected Federal Depository Libraries used by the U.S. Government
Printing Office and to individuals who requested them. The second medium is the Internet. In-
depth versions of the text and data are available on the SAMAB, Forest Service, and Info South
home pages on the World Wide Web (WWW). i
The spirit of the SAA can best be summarized by a quotation from the documents: "The
Southern Appalachian Assessment was accomplished through the cooperation of federal and
state natural resource agency specialists. The strong emphasis placed on working together
toward a common goal is increasingly recognized as essential to effective government operation.
Teamwork has strengthened our understanding and conununication. With the assessment as a
framework for future action, government policy and management can become more consistent
and better coordinated." This basic principle is being applied as various groups work to further
apply the information contained in the SAA. ;
8-13
-------
8.3.3. EPA Watershed Assessments
^^' ot^er ^B^6^ ^ ^te a§encies> environmental groups, and communities are
placing increasing emphasis on community-based environmental protection and integrated
ecosystem management. This emphasis arises from a recognition that the impacts of multiple
human activities combine in the environment to cause significant adverse ecological effects that
are not amenable to regulation under current environmental law, Unless these stressors are
managed at the community level, local and national environmental goals may not be achievable.
As the Agency shifts emphasis from command and control toward voluntary compliance and
community-based environmental protection, it becomes critical that EPA provide the scientific
basis for community-level management decisions. States and local organizations need a process
and tools they are able and willing to use for determining what ecological resources are at risk
31x1 how best to protect those resources through management action. Case studies for evaluating
•'',(': j ''. : tV. , ..... :"'" ; • :" " .•'!!•' • , ,. :, I , , 'I,-1!. ! i, 111 il ' " P
risk to watershed ecosystems were initiated to develop examples and guidance on how to use
.'i'-i ! ' .":!";!,: '!(' ! ,i ; ' "' , . ; ..... „ i ' '. : ;i; ' i i| i • ' :• ••
science more effectively in ecosystem management.
j
8.3.3.1. Background
The watershed ecological risk assessment case studies were initiated in September 1993
to evaluate the feasibility of applying the ecological risk assessment process as provided in the
Frarriework for Ecological Risk Assessment (U.S. EPA, 1992) to the more complex context of
watershed ecosystem management. The Risk Assessment Forum and the Office of Water agreed
to jointly sponsor the development of prototype ecological risk assessment case studies in
j
watersheds under the guidance of a Risk Assessment Forum technical panel. The case study
watersheds served as natural laboratories where teams used the process of ecological risk
assessment to address ecosystem-level problems concerning diverse stressors, ecological values,
and political and socipeconomic concerns hi watersheds of different type, size, and complexity.
The case studies served as a mechanism for learning about key management and research
questions, limitations to the risk assessment process provided in the framework report, and issues
surrounding involvement by interested parties that must participate in resource management.
Watershed ecosystems were chosen as the landscape unit for ecological, pragmatic, and
• I'M , IS"! . i • i , . •' ::" •• ..... . ..... iiiii , ° ii ir °
programmatic reasons: (1) Watersheds are natural geomorphological units with definable
boundaries where water flows across the landscape and collects in surface water bodies and
groundwater. Because water flows across a landscape, the effect of human impacts occurring on
land and directly in the water become combined as water flows toward collection basins such as
rivers, lakes, wetlands, and estuaries, thus providing an effective landscape unit to assess the
combined and cumulative effect of multiple stressors. (2) Watershed ecosystems are highly
8-14
-------
flexible in size. The size defined is based on the type of issues and relevant management
decisions. A small community may be interested in the watershed in its valley and may focus
management efforts at its level of influence, even though its watershed is part of a larger system.
A State may choose to focus on a watershed that covers one-quarter of the State in order to
organize permitting activities. Multiple States may become involved in cooperative management
of large watersheds that cross political boundaries. Thus, watersheds can be local or regional in
scope, and can cover multiple ecologically diverse regions. (3) Clean abundant water will
increasingly become a highly valued limiting resource, both for direct human use and for
supporting ecosystems. (4) EPA is encouraging States to organize regulatory and nonregulatory
efforts according to watershed boundaries. This is intended to! focus efforts in such a way as to
promote the coordination of management efforts to improve environmental protection and reduce
management cost. Geographic areas defined by a watershed afe not appropriate for all
environmental problems requiring management. The type of assessment question being asked
determines the rationale for defining landscape boundaries. For example, a watershed would be
appropriate for addressing risk to aquatic resources within a stirface water body but would not be
effective for concerns about air pollution in a forest ecosystem that covers parts of several
watersheds. Although the case studies are focused on watershied ecosystems, the project's focus
is on the process of conducting risk assessments for ecosystem-level problems. This process is
readily adaptable to other ecosystem management problems. ;
8.3.3.2. Process
The case studies were initiated in 1993 through joint sponsorship by the Office of Water
and the Office of Research and Development and administered under the Risk Assessment
Forum through a technical panel.
8.3.3.3. Watershed Case Study Selection \
Early in 1993, a solicitation for candidate watersheds for the project was announced and
resulted in more than 50 applications. Watersheds were selected for the project on the basis of
specific selection criteria, including data availability, identification of local participants, diversity
of stressors, and significant and unique ecological values. The watersheds selected represent
different surface water types; an array of chemical, physical, and biological stressors; and a
diversity of valued ecological resources, scales, management problems, socioeconomic
circumstances, and regions. Case study teams were established and began work in September
1993.
8-15
-------
8.3.3.4. Case Study Teams
Each case study is being developed by an interdisciplinary, interagency team of scientists
and natural resource managers. Professionals recruited for the teams include EPA scientists and
managers from regions and program offices, State scientists and regulators, and scientists from
other Federal agencies, nongovernmental groups, industry, and academia. When forming teams,
every effort was made to recruit individuals with expertise in ecological risk assessment,
ecological processes, the ecological resources and stressors in the targeted watershed, and
ecosystem management Recruitment has been a continuing process throughout development.
^eam s*ze ranSes from 10 to 50 members and participants, and other professionals are consulted
as needed. The teams hold regular meetings (normally by conference call), and all teams have
met in the watershed as part of the work on the case study.
8.3.3.5. Characteristics of Selected Watersheds
,iii 'i, 1 ,'JliH . . . , 1.1 i , ;,
The Clinch River Valley contains highly valued fish and mussel communities and
includes the greatest diversity of mussels in North America, many of which are rare and
endangered. The valley is also a Last Great Place, but agricultural practices and mining are
8-16
-------
major stressors in the high-relief terrain environment. Protection of these valued resources must
be done in a socioeconomically depressed area.
The Middle Platte River wetlands support millions of birds migrating in the Central
Flyway, including the endangered whooping crane, as well as many resident species.
Competition for water in the Middle Platte River, part of our iNation's breadbasket, is a
politically charged issue. Hydrological modifications have changed the broad-braided river
wetlands of the Middle Platte to a 50-mile stretch of narrowed wetland systems.
The Middle Snake River, once charged by natural springs bursting from canyon walls, is
now primarily fed by irrigation return flows. Considered the most unpaired watershed among
the case studies, the Snake River has become an algae- and sediment-choked stream in many
parts of its reach. Better management of dams, irrigation return flows, sediments, and trout
hatcheries is central for protecting and restoring at least part of the river's function.
Waquoit Bay is the smallest watershed among the case studies, valued for its aesthetic
beauty, recreational opportunities, and commercial fisheries. Currently, residential development,
a Superfund site, and other activities in the watershed are placing these values at risk. The fairly
affluent community is seeking ways to reverse degradation and regain ecological values.
8.3.3.6. Resources to Support Case Study Development
The case studies were designed to demonstrate what can be accomplished using available
data and limited resources. The project was organized to approximate the kinds of expertise,
resources, and data likely to exist in communities that would be responsible for using guidance
for implementing ecosystem management at the community level. The following statements
characterize case study resources:
• Members and participants on watershed teams are professionals from diverse disciplines
whose time was volunteered by their organizations for the effort.
• A small minority of participants were familiar with risk assessment.
• Each watershed ecosystem is being evaluated within the context of many competing
socioeconomic and political concerns.
• The case studies are being conducted with minimal funding and a reliance on existing
data.
8.3.3.7. Lessons Learned
The watershed ecological risk assessment case studies were developed using available
guidance on ecological risk assessment as presented in the framework report (U.S. EPA, 1992).
8-17
-------
During case study development, several adjustments to this process were found to be valuable.
Each team's experiences added dimension to our interpretation about what adjustments were
needed. Sometimes teams experienced successes, sometimes readjustments and redirection. All
, , . , , i
were important learning opportunities.
We believe that the process that emerged from the case studies is sound and valuable and
will be the focus of detailed guidance in the future. However, all of the lessons learned are now
•••'•i i '! : ; '.its • *: ' n" ' • ' . ; : ' •' '- =' 'i '"i • i""11 ' ,. 'ii "i:: • . ..'
incorporated at a general level in the Agency's Draft Proposed Guidelines for Ecological Risk
Assessment (U.S. EPA, 1996). Specific issues and changes that emerged from conducting the
case Studies include how value-initiated risk assessments alter the process of problem
formulation, the importance and process of "planning" for establishing ecosystem management
goalSj how to develop and interpret management goals for an ecosystem-level risk assessment,
how to select and define assessment endpoints, how to develop conceptual models for watershed
ecosystems with multiple stressors, when and how to define measures and data that will be used
in the assessment, and the explicit need for analysis plans.
8.3.3.8. Reviews ajfid Current Status
In May 1994, the Risk Assessment Forum Ecorisk Oversight Committee held a peer
review of the draft problems formulations. Substantial discussion at that review centered on the
generation of management goals for the watershed and their interpretation into assessment
endpoints. The Risk Assessment Forum organized a second peer review in September 1994 that
focused on the analysis plans generated from conceptual model development. Significant
discussion centered on aspects of the risk assessment process that were changing as a result of
' "
case study development. Throughout development, case study drafts have undergone technical
peer review by independent professionals knowledgeable about the watershed. In July 1996, the
?' i' .'..', ?Ii .Till ' : ' ,'">! : : '*?.-• !'•: ;:;, Si •>, • "A il: ., •''••' •
"process" and "lessons learned" and draft "planning and problem formulation" sections of the
five case studies were presented to the EPA Science Advisory Board (SAB) for advice on work
in progress. The results of that peer review are published and available to the public on the SAB
Website (http://www.epa.gov/sab) as report number EPA-SAB-EPEC-ADV97-001. Based on
feedback, the case study teams continue to refine work on problem formulation and are moving
into analysis and risk characterization. Many of the teams are reconfiguring to ensure that
adequate expertise is on the teams for the next phase. Some teams have obtained substantial
grant and extramural funds to expand and improve the risk assessment based on the success of
the first phases of the case study work. It is anticipated that an additional 2 years will be
neqeSsary to compiete the full ecological risk assessment and finalize ecosystem-level guidance.
8-18
-------
8.3.4. Examples of U.S. Department of Defense Activities in Ecological Assessments
Ecosystem management was adopted by the U.S. military in recognition of the U.S.
Department of Defense's (DoD) responsibility as a manager of public trust resources that
encompass 25 million acres. It also was recognized that responsible management with a long-
term perspective will ensure the continuing availability of training resources, thereby enhancing
the sustainability of the military's readiness mission. The Army's Integrated Training Area
Management Program, implemented on more than 60 installations nationwide, is an excellent
example of the military's efforts to integrate land management objectives with combat
requirements through standard methods for monitoring land condition and trends, managing
training lands to their carrying capacities, and rehabilitating resources toward a natural state of
biodiversity. Within the Army Corps of Engineers Civil Works Program, which is responsible
for managing an additional 12 million acres of Federal lands and waters, there is a long history of
cumulative impact assessment of watersheds that is now helping to develop risk-based
approaches in many regions.
An excellent handbook for military resource managers, Conserving Biodiversity on
Military Lands, was recently published by DoD and the Nature Conservancy. In addition, the
Army published Tri-Service Procedural Guidelines for Ecological Risk Assessment in June 1996,
which provides cost-effective tiered procedures with which to coordinate the defense ecological
risk assessment efforts of contractors and follows the paradigm put forward in EPA's Framework
for Ecological Risk Assessment (U.S. EPA, 1992).
8.3.4.1. DoD's Ecosystem Management Policy ;
Initial DoD guidance established the goal of ecosystem management to balance
sustainable human activities, such as the support of DoD missions, with the maintenance and
improvement of native biological diversity. Ecosystem management is a balance of ecology,
economics, and social values. Partnering and public involvement are stipulated as means to
achieving shared goals and making decisions. Goodman (1994) outlined 10 ecosystem
management principles and guidelines, which can be summarized in four general themes:
ecological approach, stakeholder involvement and collaboration, scientific and field-tested
information, and adaptive management. This guidance was formalized in 1996 (DoD Instruction
4715.3, Environmental Conservation Program, May 3,1996). The following examples illustrate
how the military services and DoD are making strides toward full implementation of ecosystem
management. ;
8-19
-------
8.3.4.2. Site, Examples
At Camp Pendleton, California, a project entitled Alternative Futures for the Region is to
"examine the connections between urban, suburban., and rural development and the consequent
stresses on native habitats and biodiversity." The study poses an important question: How will
urban and suburban growth and change that is forecast and planned in the rapidly developing
area between San Diego and Los Angeles influence biodiversity? The question is particularly
" ' ' ' !| '
relevant for Camp Pendleton because it constitutes the largest unbuilt segment of land on the
southern California coastline and one of the most biologically diverse environments in the United
States. Given its position and cache of unbuilt land, Camp Pendleton is central to maintaining
the long-term.biodiversity of the region. Camp Pendleton plays a key role in the connectivity of
the region's ecosystems and over the long term faces the risk of becoming a "habitat island" for
,'"';';'' ' ' !''"?l|: " ''!5i'i '' ' ' -: •' ' , i ! • ' ' : i ',i i1 ' ' •" ''M1' , il ' f '' '' 'i'"'
species. Camp Pendleton is also key to the military's readiness mission, being the only facility
on the West Coast where amphibious assault maneuvers can be practiced. Camp Pendleton
resource managers believe that a regional perspective is necessary if a true ecological perspective
is to be achieved and that an ecological perspective enhances the long-term readiness mission.
The project asks, "Can appropriate management of biodiversity and landscape planning allow the
rrffiitary to more effectively manage its property and efficiently fulfill its mission?" From the
Camp Pendleton perspective it asks, "How might issues of biodiversity affect or influence land
management activities of the camp?" and "How might future development or conservation
'upstream' from Camp Pendleton influence hydrology, ecosystems, and biodiversity on the base
and thus potentially influence its primary mission of training?"
li'1** i.i',''',!l , . • il' l|
In the Chesapeake Bay Program (CBP), each Federal agency commits to managing the
,,, : / if ;f .:•!•! • • • t-, i j
8-20
-------
and analysis of information regarding inventories, resource assessments, scientific
documentation, and land management by all Federal, State, and local agencies and other
interested parties. Ultimately, a queryable database will provide land managers and resource
specialists with the tools for attempting to create a regional-scale database to affect dynamic,
sustainable ecosystem management. MDEI is an important example of DoD's ecosystem
management activities for several reasions: (1) It is an attempt to provide uniform data coverage
across an entire scientifically defined ecoregion, regardless of political or administrative
boundaries; (2) data collection, interpretation, documentation, :and sharing will be a significant
tool used for integrated planning and decision; (3) it provides an important model for sharing,
integration, and use of data for ecosystem management purposes by a broad and varied group of
participants; and (4) DoD's military trainers have been effectively integrated into the MDEI's
program.
Adaptive management means the ability to change management structures and protocols
to adjust to new or enhanced understandings advanced by the scientific community. Eglin Air
Force Base, a 463,000-acre facility near Pensacola, FL, is home to the largest remaining longleaf
pine system. Eglin and the surrounding landscape contain 153 rare species, including 13 that are
federally listed, and many exceptional occurrences of imperiled natural communities. In
partnership with the Nature Conservancy and 30 other organizations, Eglin has developed an
i
ambitious ecosystem management program featuring an adaptive approach. Among the natural
resource management program's most important goals is to restore and maintain the resiliency of
native species and ecosystems. Eglin's military and natural resource management staff believe
this approach best provides the broadest array of options for pjursuing the base's military mission
of testing conventional weapons and munitions. As it is being practiced at Eglin, adaptive
management is an integrated, science-driven, and policy-based set of methods and principles for
grappling with regional-scale environmental management problems. It seeks to answer two
fundamental questions: (1) How do ecosystems change, and (2) how do institutions learn and
adapt? Its goal is to integrate knowledge of ecosystem behavior with the policy processes of
human institutions and to create learning institutions that can adapt to ecological and social
change. With highly unique and high-value posts, camps, and stations, or "habitat islands" in the
context of this report, DoD and the U.S. military services will continue to partner with
neighboring ecosystem managers, experts, and the public to sustain our Nation's ecosystems for
present and future generations.
8-21
-------
8.4. RISK ASSESSMENT METHODOLOGY DEVELOPMENT
8.4.1. Expanded Use of the EPA Guidelines
Several agencies have gained experience with ecosystem assessments in recent years.
These assessments have varied in scope, cost, and specificity of problems addressed. The
,;;" ; f , ":~ ":'~, " ' .: . i,
agencies and scientists involved have learned lessons along the way, and there is general
consensus that the utility of the assessments has unproved as experience has been gained.
Likewise, practitioners believe that individual assessments should be tailored to address the
specific issues and circumstances generating the need for the particular assessment. However, as
the need for ecosystem assessments appears likely to continue or expand, continuation of a
completely "hand-crafted" approach is not efficient, will overuse available scientific resources,
and will not sustain improvement hi the assessment craft. We believe expanded use of the EPA
Guidelines for Ecological Risk Assessment (U.S. EPA, 1998) offers an opportunity for several
agencies to improve the efficiency and utility of ecosystem assessments. "While the guidelines
were developed for EPA use, judicious use by other agencies can provide government-wide
benefits.
".' • j- '".:. „:- • ' , ••: . : ' ' ••;', •", }.'•• ', '. , •'•
Ecosystem assessments do not focus on adverse impacts. In fact, their main focus is to
provide comprehensive, integrated information to assist with planning and decision making in an
ecosystem context. The EPA guidelines are designed to evaluate the likelihood of adverse
effects because they are based on a risk assessment paradigm. The conceptual impasse between
ecosystem assessments, which do not have an a priori focus on adverse impacts, and the EPA
ecological risk assessment guidelines, which do have an a priori focus on adverse effects, is more
apparent than real. Sustainability is the goal of ecosystem management, and the EPA guidelines
specifically address methodologies for translating sustainability goals to risk assessment
endpoints.
in ', ' ! : ,.,, ' ' ' . '.' "'" ' " ,:|' • j '
Dialog among parties interested in ecosystem assessment would benefit from an
agreement on general long-term goals for ecosystem management, such as sustainability, and the
translation of those goals to endpoints amenable to ecological risk assessment approaches..
Scientists and/or risk assessors should be involved as facilitators of this dialog while avoiding a
role as determiners of the goals and endpoints. Benefits for decision makers and other interested
parties will be greater accuracy, clarity, and precision of scientific information available for
decision making. Benefits for ecosystem assessment scientists and/or risk assessors are clarity of
i • . _ -i*l I.:!1 ' ' ' '•:• if K 'I .i
expectations and lessening of end product controversy.
Flexibility and rigor need to be balanced when use of the EPA guidelines is expanded for
application to ecosystem assessment. Traditional risk assessments require data and process rules
", ' ' : ;j.i ]-i1.:.!!1 ! i . ' , '! • " '. ' ' y,! :!v;;;" :'!'! J '; !
that are simply not available for most ecosystem assessments. The EPA guidelines clearly
8-22
-------
recognize the need to adjust risk assessment data rigor to the information available. The
guidelines are sufficiently flexible for application to most ecosystem assessments. Principals
responsible for ecosystem assessments need to embrace this flexibility while striving to retain as
much rigor as possible. Expanded use of the EPA guidelines will increase the value of the
Agency's investment in producing them while simultaneously increasing the value of ecosystem
assessments that utilize them. Agencies responsible for ecosystem assessments and other
ecosystem management activities should seek to understand the EPA guidelines and expand then-
use. EPA should actively seek to transfer guideline technologies to agencies with ecosystem
management responsibilities and expand their use.
8.4.2. Technical and Research Challenges
Ecosystem management is a complex topic that contains a variety of challenges.
Ecosystem management needs to be based on sound scientific studies and assessments.
However, it also needs to reflect societal values and issues, political and economic concerns, and
the decisions need to be legally defensible. Challenges include better linkages of research and
technical information to the way society makes decisions hi general and how ecosystem
management is implemented in particular. Developing effective methods to communicate
science and management options and consequences to the public, decision makers, and other
stakeholders is a critical need. Successful partnerships between elected officials, the public,
interested parties, and scientists have produced effective management programs for the
Chesapeake Bay, south Florida, and other regional programs. The experiences gained from these
partnerships need to be applied to other regional programs.
8.5. RISK ASSESSMENT IN ECOSYSTEM MANAGEMENT DECISION MAKING
Ecosystem management is the continuous process of holistically managing the physical,
biological, and human components of ecosystems. The concepts underlying risk management
are relatively straightforward (Marcot, 1986; Bartel et al., 1992; Burgman et al., 1993; Covello
and Merkhofer, 1993; Morgan et al., 1990; Lackey, 1994; Suter, 1993). However, the joint
application of the ecosystem and risk management concepts in a complex of ecological,
organizational, and sociological processes is quite difficult. :
8.5.1. The Risk Management Cycle and Ecosystem Management
Risk assessment is part of a cycle of processes that make up risk management.
Ecological risk assessment is described in earlier sections of this document. Potential application
of the risk management process to ecosystem management would involve eight phases:
8-23 :
-------
1. Hazard identification—identifying human actions or natural events, the conditions under
which they could potentially produce adverse effects, and the parts of the ecosystem that
might be affected.
:.'.'... , > ;. /::,: ,, ' , ' ', '.,:, ':': i,
2. Risk assessment—characterizing risks imposed by some proposed action by estimating
magnitudes1 of potential loss, exposure pathways, and likelihoods of occurrence.
3. Evaluation—judging the relative acceptability of assessed risks in light of policies,
standards, organizational or cultural norms, public opinion, and other expressions of
human values. Also, comparing different risks for their relative contribution to the
overall level of severity.
4. Adjustment—choosing strategies for modifying, avoiding, accepting, or otherwise
dealing with the risk profile of proposed actions or likely natural events. These choices
involve comparing risk adjustment benefits and costs of various strategies and policy
instruments and making difficult tradeoffs among risks and costs.
i
5. Implementation—interpreting the strategy mix in practical standards? guidelines, and
incentive systems.
6. Monitoring—tracking the effectiveness of the risk adjustment strategies by measuring
exposure pathways and risk endpoints sensing for "signal" events that could trigger
adaptive responses.
7. Adaptive Management—strategies can be implemented through modifications in the
proposed actions, mitigations for particular risks, or responses under a planned adaptive
management program (Holling, 1978; Walters, 1986).
8. Risk communication—translating the results of one phase to another among ecosystem
managers, scientists, policy makers, and the public. The traditional view of risk
communication was of a one-way flow of technical information from experts to the
public. Recent approaches emphasize multiway communication with an emphasis on
understanding the mental models and belief systems on which people judge the
acceptability of risks. For the risk management cycle to work successfully, risk
communication—clarity, completeness, accuracy, and compatibility with information
processing; styles—needs to be built into every phase of the cycle.
While these eight phases share some important similarities with the Ecological Risk
Assessment Framework, as described earlier in this document, differences are illustrative of the
gaps that currently exist between the framework and application of risk management in
ecosystem management.
8-24
I Ijlfj ;i| ..... V ..... ......... ,1 ..... ^Hif ..... fW ;.i|J§jidU|.Ji,ii St. J1- .......... iii'i ........ ! ..... ii i,,,'.;,!!.:.:, j :,| ...... ^.m &', .la.i ........ • ..... i. :'!.!: , ,; ........... ! , i ..... at ,!.,,:' ..... !£.- iiiJIiSL-'l ..... iiilJIJliii '• 3- i
. Hl'J i
•.ifclillill ,,;l!lllii!!iil-i:.
-------
8.5.2. Risk Management and Decision Quality
The effectiveness of the cycle depends in part on the quality of the human judgments and
decisions that support it. A high-quality decision is one that (1) solves the correct problem; (2)
clearly describes the problem, criteria, and alternatives to the decision maker; (3) generates and
evaluates many relevant alternatives; (4) makes choices consistent with criteria and information;
and (5) provides for learning that will improve future decisions.
Most successful attempts to improve decision making have involved better organizing
and structuring of basic cognitive tasks. Kleindorfer et al. (1993), Dawes (1988), Bazerman
(1994), and other decision scientists contend that unstructured;tasks are subject to many biases
and illusions. Generic decision tasks include (1) process mapping, (2) problem framing, (3)
intelligence gathering, (4) evaluation and choice, and (5) learning from feedback. Each of these
tasks are subject to unique biases and opportunities for improvement.
Risk management involves decisions about how to reduce probabilities, lower potential
losses, interrupt exposure pathways, or collect information to better predict events (MacCrimmon
and Wehrung, 1986; Head and Horn, 1991). Each phase of the risk management cycle
corresponds to one or more generic decision tasks. The cycle itself is a process map that lays out
a sequence of steps and prescribes methodologies and protocols. Hazard identification is a
problem-framing and intelligence-galhering task. Risk assessment takes these tasks to higher
levels of rigor by requiring probabilistic judgments and analysis of complex pathways. Risk
evaluation and adjustment are tradeoff evaluation and choice tasks. Risk monitoring is an
intelligence-gathering and learning task.
8.5.3. Risk Assessment as a Decision Aid
As a decision-aiding tool, risk assessment can be judged by six criteria (Covello and
Merkhofer, 1993): (1) logical soundness, (2) completeness, (3) accuracy, (4) acceptability, (5)
practicality, and (6) effectiveness. The first three criteria are measures of scientific discipline;
the last three relate to the users of the assessments' outputs.
Logical soundness is the degree to which the assessment conforms with fundamental
theoretical assumptions of basic science, the specific field, and the laws of probability.
Completeness refers to whether the assessment accounts for all considerations and scenarios that
are relevant to a reasonable choice of policies. Accuracy is whether the assessment correctly
describes sources of uncertainty, the probability of their events, and their effects. Sources of
inaccuracy include biases and errors in data collection, model specification, and expert judgment,
as well as inappropriate application of modeling methods. Inaccuracies can be minimized by
sensitivity testing, peer review, and comparison with other assessments or empirical patterns.
8-25
-------
Acceptability is whether decision makers understand the assessment and find it believable.
Barriers to acceptabiliry include lack of public credibility of risk experts, experts' limited
understanding of public risk perceptions, and failures to disclose limitations and uncertainties.
Cr^ikilitv Problem? stem from instability of results under different assumptions and data,
complexity of outputs, overuse of technical jargon, invasion of assessment results with advocacy
111 ' ' ' ' ''" ' ' ' ' ' '"' ' '' ' ' ' , Ml' ll| I I ll 'I ' J
for a risk adjustment option, and lack of trust for the agency doing the assessment. One ironic
barjier to acceptability is how uncertainty is displayed. Because humans prefer certainty in
answers and predictions, even if they are illusory, risk assessments that describe large degrees of
uncertainty tend to be rejected. When risk assessors are most candid about environmental
uncertainties, the risk manager or public is likely to be most disappointed because the assessors
cannot be definite. If risk assessors ignore or obscure uncertainties and gives unambiguous
predictions and advice, their credibility is damaged, especially if the risky event actually occurs
(Carpenter, 1993). Practicality relates to whether the assessment can be employed in a real-
•III!' , , ' ' , in II', •!' 'i'Vi . : , , , • , • ,• ,r , ' ,1,1,, ' ,. !!||j' 1,. , :r • • •' rtp , ii1/,, ,
wqrjd environment with deadlines and limited resources and information. Risk assessments
usually require interdisciplinary teams, iterations, and many interlocking steps. Risk
^essments, esjPeciaIlv large-scale assessments, need to be well managed; scientists who serve
on assessment teams do not have the inclination or abilities to manage product-driven efforts.
Too many risk assessments evolve into piecemeal research projects, with individual scientists
pursuing their own subjects and then trying to assemble what they have as the deadline nears.
The effectiveness of a risk assessment is ultimately how much it improves the risk
adjustment decision making in the ecosystem management organization. A risk assessment does
n°tStand alone f5 fSS estimate> a11 analysis, or a report, but is actually part of a way of thinking.
Using probabmstic-information in making decisions is not easy or natural for most human
beings. Most people focus disproportionately on magnitudes of dire outcomes, almost ignoring
probability information. People have many automatic or routine ways of dealing with risk that
can substitute for analyzing risks. Such behaviors include delaying the decision, delegating or
otherwise transferring the risk, ignoring uncertainties, treating uncertainties as if they were
certain, collecting information, modifying the management alternatives, buying insurance,
making contingency plans, and setting performance standards. If followed without any
knowledge of the nature or the degree of the risk being managed, these strategies can misallocate
intellectual, financial, and physical resources.
Many still approach risk assessment as a way of justifying and documenting decisions
that are already made. To these persons, a good risk assessment will be disappointing because it
will expose many incorrect and unfounded assumptions.
8-26
-------
8.5.4. Expert Judgment in Risk Management
Quality in professional, or so-called "expert," judgment is an important element of
decision quality, especially when there are no precedent events or data, and statistical evaluation
is not possible. In these situations, assessment quality may be gauged more on how the judgment
mobilizes knowledge, both theoretical and practical, to estimate effects and how the process
advances and contributes to the decision process.
Expert judgments of ecological structure and stressor responses enter into the
specification of measurement endpoints and the estimation of likelihoods and severities.
Judgments of acceptability of risks enter into the risk evaluatibn phase; judgments of managerial
feasibility are the basis for selecting risk adjustment actions. Quality judgments rationally use
scientific and other sources of information and are expressed in ways that can be understood and
used by decision makers and stakeholders. '
Quality expert judgments are important sources of infoirmation in risk assessment and risk
adjustment. Risk assessment is a form of judgmental hypothesis testing. The null hypothesis is
that there is no risk; alternative hypotheses are that the risk may be at various levels. If the risk
assessor judges that there is a risk, efforts may be made to adjust it. If the event that creates the
loss never actually occurs, the risk management process has made a false-positive error and
incurred unnecessary costs. If the expert judges that there is no risk and the event actually
occurs, there has been a false-negative error with losses that are perhaps unacceptably high. A
good risk management program will attempt to minimize the combination of false positives and
false negatives. The symmetry of the loss and cost distributions should guide the level of effort
put into assessment and adjustment strategies. Where false-negative losses overwhelm false-
positive costs, it is better to invest in more sophisticated assessment processes and more stringent
risk adjustments.
Three types of bias influence expert judgments. Task bias is caused by the improper
definitions of events or initiating conditions leading the assessor astray. Conceptual biases
include motivational biases (wishful thinking or advocacy) and cognitive biases (systematic
patterns of thinking that do not allow full expression of subject matter knowledge in probability
form). Expert judgments can be improved by structuring elicitation processes to keep judgments
relatively free of these biases and to folly reflect the knowledge of the experts. Process
guidelines include avoiding experts who have agendas or preconceived notions, obtaining
estimates from several sources, challenging experts to explain their rationale, encouraging
experts to move estimates from their initial estimate "anchors," requiring experts to provide
degrees of uncertainty, checking estimates against any record^ of similar losses, and removing
sources of distraction and motivational bias from the elicitation environment (Cleaves, 1994).
8-27
-------
8.5.5. Risk Evaluation, Adjustment, and Decision Quality
T*16 risk mana§er should fully evaluate a range of options for managing risks, including
incentive-based and other flexible policies for allowing managers to assess and adjust risk
according to site-specific information, experience, and knowledge. Risk policy is composed of
rules and standards and other instruments that signal which risks are most important and what
levels of those risks are acceptable. Protection standards are written to ensure particular
behaviors of human managers toward ecosystem components. Standards limit improper outside
factors from influencing resource management decisions. Standards convert probabilistic choice
"*? a detenninistic rule (Keeney, 1983). Whoever develops standards makes critical tradeoffs:
those who accept or implement standards may not have the same degree of discretion. Some
flexibility is beneficial for sites at smaller scales of analysis and choice. Standards that attempt
to minimize magnitude discourage managers from basing their decisions on the relationship of
magnitude and probability. Standards that consider probability as well as magnitude may
provide a useful tool for risk managers.
8.5.6. Risk Communication and Decision Quality
Implementing risk management in an ecosystem management context depends on clear
communicatio? amonS participants in the risk management cycle and on public support. The
public will not accept risk assessments or management policies just because they represent expert
judgment. Underlying psychological attributes of risk perception influence risk information.
These attributes incjude (1) voluntariness or controllability, (2) dread or vividness, (3) familiarity
with the outcome, (4) extent (degree of catastrophe) in the losses, and (5) future generational
impacts (Covelio et al., 1986; Sandman, 1985; Slovic, 1987; Cross, 1994). People focus on these
attributes to adjust their judgments of magnitude, frequency, and degree of exposure. Some
events have a high "signal" value in that they may symbolize the potential of more serious risks
in the future.
Involving the public in risk management is the best way to better understand and work
witii risk perceptions. Many techniques such as focus group interviewing, group facilitation, and
alternative dispute resolution can be applied to public involvement in risk assessment. Part of
this effort should be devoted to helping the public recognize and contribute to decision quality.
All parties should be able to ask questions about decision quality: What perspectives are
involved? What factors have been omitted? Are major uncertainties quantified and explained?
Have sensitivity analyses been conducted? How are the risk assessment and risk adjustment
tasks separated to minimize bias? An informed public will appreciate honest efforts to
8-28
-------
characterize risks, be more creative in suggesting risk policies, and be less inclined to reject
assessment results.
8.6. NEXT STEPS
This chapter has provided some examples of ecosystem assessments for ecosystem
management, views of how risk assessment methodologies could contribute to more efficient
ecosystem assessments, and a description of how improved ecosystem management decision
making would be the result of hybridizing ecological risk and1 ecosystem assessments. The
following "next steps" would help natural resource agencies move along that pathway.
Link ecological risk assessment and ecosystem management to improve organizational
and analytical consistency in support of multiple scales of resource management.
Expand the use of EPA's Guidelines for Ecological Risk Assessments (U.S. EPA, 1998)
across agencies to improve the efficiency and utility of ecosystem assessments.
i
• Improve valuation technology to provide better definition of societal values and
preferences and to better achieve awareness and active involvement of a diverse array of
stakeholders.
Link ecosystem assessments using common information themes and protocols that
provide analyses of ecosystem process, structure, and function at multiple temporal
spatial scales.
8.7. REFERENCES
Bartel, SM; Gardner, RH; O'Neill, RV. (1992) Ecological risk estimation. Chelsea, MI: Lewis Publishers.
Bazerman, MH. (1994) Judgment in managerial decision making, 3rd ed. New York: John Wiley and Sons.
Burgman, MA; Person, S; Akcakaya, HR. (1993) Risk assessment in conservation biology. London: Chapman and
Hall. ;
Carpenter, RA. (1995) Communicating environmental science uncertainties. Environ Prof 17:127-136.
Christensen, NL; Bartuska, AM; Brown, JH; Carpenter, S; D'Antonio, C; Francis, R; Franklin, JF; MacMahon, JA;
Noss, RF; Parsons, DJ; Peterson, CH; Turner, MG; Woodmansee, RG. (1996) The report of the Ecological Society
of America Committee on the Scientific Basis for Ecosystem Management. Ecol Applications (6)3:665-691.
Cleaves, DA. (1994) Assessing uncertainty in expert judgments about natural resoi>:ces. General Technical Report
SO-110. U.S. Department of Agriculture, Forest Service, Southern Forest Experiment Station, New Orleans, LA.
Covello, VT; Merkhofer, MW. (1993) Risk assessment methods: approaches for assessing health and environmental
risks. New York: Plenum Press.
8-29
-------
„•.,; iv ,1 : • . • ., ' • . •;<••:, \ 1 • n ,
Covello, VT; von Winterfeldt, D; Slovic, P. (1986) Risk communication: a review of the literature. Risk Abstr
Jt 1 / t™ i
Cross, FB. (1994) The public role in risk control. Environ Law 24:821-969.
Dawes, RM. (1988) Rational choice in an uncertain world. Orlando, FL: Harcourt Brace Jovanpvich.
Goodman, S. (1994) Memorandum from Deputy Undersecretary for Environmental Security Sherri Goodman on
ecosystem management, August 8, 1994.
Haynes, RW; Graham, RT; Quigley, TM, eds. (1996) A framework for ecosystem management in the Interior
Columbia Basin including portions of the Klamath and Great Basins. General Technical Report PNW-GTR-374.
U.S. Department of Agriculture. Forest Service, Pacific Northeast Research Station, Portland, OR.
Head, GL; Horn S, II. (1991) Essentials of risk management. Vol. I and II. 2nd ed. Malvern, PA: Insurance Institute
of America.
Holling, CS, ed. (1978) Adaptive environmental assessment and management. New York: John Wiley and Sons.
Keeney, RL. (1 983) Issues in evaluating standards. Interfaces 1 3 : 12-22.
, ••«'> -''i ':..., : r " ' ..... ; ,
er' PR5 Kunreuther, HC; Schoemaker, PJH. (1993) Decision sciences: an integrative perspective New
York; Cambridge University Press.
Lackey, RL. (1994) Ecological risk assessment. Fisheries 19(9): 14-18.
Little, IMD; Mirrlees, JA. (1994) The costs and benefits of analysis. In: Cost-benefit analysis. Layard, R; Glaister,
S, eds. Cambridge, UK: Cambridge University Press.
fvi ., , ';. ., 'j;; ;.;i , ' , ..... .,•.,,, ,': ' ' ;•. *),;, xi, j : : „•
MacCrimmon, KR; Wehrung, DA. (1986) Taking risks: the management of uncertainty. New York: The Free Press.
Marcot, BG. (1 986) Concepts of risk analysis as applied to viable population assessment and planning. In- The
management of viable populations: theory, applications, and case studies. Wilcox, BA; Broussard, PF; Marcot, BG,
eds. Stanford, CA: Center for Conservation Biology, Stanford University, pp. 1-13.
Morgan, M; Henrion, G; Henrion, M. (1990) Uncertainty: a guide to dealing with uncertainty in quantitative risk
and policy analysis. Cambridge, UK: Cambridge University Press.
Quigley, TM; Arbelbide, SJ, eds. (1996) An assessment of ecosystem components in the interior Columbia Basin
Md portions of the Klamath and Great Basins. General Technical Report PNW-GTR-XXX. U.S. Department of
Agriculture, Forest Service, Pacific Northwest Research Station, Portland, OR.
""
Quigley, TM; Haynes, RW; Graham, RT, eds. (1996a) Integrated scientific assessment for ecosystem management
in the, interior Columbia Basin and portions of the Klamath and Great Basins. General Technical Report PNW-GTR-
382. U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station, Portland, OR.
5 , :|i .:' >'(!'; , " '" •' "' •' ' ' ' ' ,| ' . . ' : :' ,. ' i,j.' f'i • '< '. |j I ;.i ; «•
Quigiey, TM; Lee, KM; Arbelbide, SJ, eds. (1996b) Evaluation of EIS alternatives by the science integration team
General Technical Report PNW-GTR-XXX. U.S. Department of Agriculture, Forest Service, Pacific Northwest
Research Station, Portland, OR.
Sandman, PM. (1985) Getting to maybe: some communication aspects of siting hazardous waste facilities Seton
Hall Legis J 9:442-465.
Slovic, P. (1987) Perception of risk. Science 236:280-285.
lift,,; , :,;! ;j| ••,„*' , i 'i'1" , i j .
8-30
-------
Suter, GW, II, ed. (1993) Ecological risk assessment. Chelsea, MI: Lewis Publishers.
U.S. Department of Agriculture (USDA). (1996) Report of the Lessons Learned Workshop: policy, process, and
purpose for conducting ecoregion assessments. USDA Forest Service. Albuquerque, New Mexico. July 30 to
August 1, 1996.
U.S. Environmental Protection Agency. (1992) Framework for ecological risk assessment. Risk Assessment Forum,
Office of Research and Development, Washington, DC. EPA/630/R-92/001.
U.S. Environmental Protection Agency. (1998, May 14) Guidelines for ecological risk assessment. Federal Register
63(93):26846-26924.
Walters, C. (1986) Adaptive management of renewable resources. New York: MacMillan Publishing Co.
8-31
-------
-------
9. THE USE OF ECOLOGICAL RISK ASSESSMENT FOLLOWING THE
ACCIDENTAL RELEASE OF CHEMICALS
9.1. SUMMARY
The application of an ecological risk assessment to the accidental release of chemicals is
an iterative process. For sites where chemical release is relatively frequent (harbors and
chemical transfer areas), planning for an accidental release of chemicals is common. These plans
address the chemical types and quantities to expect, personnel, and equipment needs. The
ecological risk paradigm problem formulation phase provides the structure for a planning process
that asks the question, "What are we trying to protect?" The analysis and risk characterization
phases, in a response plan, are qualitative in nature but relate to remediation of ecological
impacts. For releases at sites where no plans exist, the ecological risk assessment is qualitative
in nature, such as limiting the size and degree to which an area is impacted. As resource
managers and trustees respond to the release, the risk to natural resources is evaluated in detail.
The required remediation to reduce the risk and/or restoration of the resource is specified. This
I
chapter briefly reviews several environmental laws that deal with accidental releases and Federal
Government responsibilities and response to them. In addition, information is provided on
Federal and State agency trustees for natural resources, natural resource damage assessments, and
the role of trustee agencies hi working with the on-scene spill coordinator.
The case study describes the accidental release of contaminated dredged material off
Charleston Harbor, SC. About 2,500 tons of dredged material contaminated with dioxin, PCBs,
and other chemicals was released. The response team identified the resources to protect and the
data requirements, which were similar to the problem formulation phase in an ecological risk
assessment. An analysis plan was developed to determine the degree of contamination of the
dredged material and seawater in the barge. Various criteria were developed to determine areas
of sediment to clean up and whether release of seawater from the ship was permitted. These
activities characterized the risk to the affected resources and put forward remediation alternatives
to the risk manager.
Risk assessors and risk managers can follow ecological risk assessment steps as a guide
in developing spill contingency plans. This would lead to better responses to accidental chemical
releases and to greater protection for sensitive ecological resources.
9.2. INTRODUCTION AND LEGISLATION
When an accidental release of a chemical occurs, the public often feels that the only
acceptable response is immediate and total removal of the chemical and restoration of the
i
9-1 :
-------
environment. Since such action is rarely, if ever, achievable, these expectations are not met, and
the public is generally disappointed by the Government's action. The ecological risk assessment
process, by encouraging a structured approach to problem solving and greater public
involvement in the process, can help this situation. By definition, emergency or accidental
releases are unscheduled and unplanned. Therefore, the principles of ecological risk assessment
must be incorporated in advance into the processes that determine the manner in which an agency
will respond to a release. Strategies to protect and restore sensitive ecological resources must be
developed before an accidental release occurs.
Several key environmental laws deal with accidental releases and dictate how Federal
agencies are to respond. An important aspect of these laws deals with the concept of "trust
resources," Federal and State trustees, and natural resource damage assessments.
The Comprehensive Environmental Restoration, Compensation, and Liability Act
(CERCLA) and the National Contingency Plan (NCP) require mat the responding agency
1 v I 1 , ; '" t .Jft.!1 ii'W! ' '•'. ''•''':': " , . "•••', ", '• 'V* :"!->•: i ;!"" '.li!;::! : ;: fiN'! | ;• : - , .: '. ..
coordinate with the natural resources trustees pn natural resource issues to understand the
policies and legislative requirements the trustee groups may have.
The portions of the Fish and Wildlife Coordination A.ct (16 U.S.C. 661-667e, as
anxended) that directly relate to hazardous material and spill response are the amendments
fi ' • ' :'" *IM jiiil"! • ' i •' ': :."•'' "... i? ••• ' : "li, " .;.;;.if;; . if"1'i.-i •
enacted in 1946 that require consultation with the U.S. Fish and Wildlife Service and the fish and
wildlife agencies of the State where the "waters of any stream or other body of water are
proposed or authorized, permitted or licensed to be impounded, diverted...or otherwise controlled
„"'1 ' "„' !!|"'MiiiJ, .'"Si I ,, " I1' •• '' '... '' •' , ' ' .." i'1!':* , ' '. ,:ii|i', ' ",'!!! I.1!!!,'.:1! ' ' ,1"1!1 Xl1: I I1!1:,:1 ' ' !' '. ' ' • ,.!".'
orj modified" by any agency under a Federal permit or license. Consultation is to be initiated for
the purpose of "preventing loss of and damage to wildlife resources."
The Refuge Administration Act (16 U.S.C. 668dd-668e, as amended) governs the
administration and resource management issues on lands in the U.S. Fish and Wildlife Service
Refuge System. The portion of the act that may play an important role during a spill or release
relates to the function of the refuge and the purpose for which the refuge was created. Activities
related to refuge lands must be consistent and compatible with the major purposes of the refuge.
Therefore, assessment and characterization of the spill may be directly influenced by the function
of the refuge.
The Endangered Species Act (16 U.S.C. 1531 et seq.) regulates a wide range of
activities affecting plants, animals, and their habitats designated as endangered or threatened.
Section 7 outlines the procedures for Federal interagency cooperation to conserve federally listed
species and designated critical habitats.
The act has a provision for proactive conservation efforts by Federal agencies. Section
7(a)(l) directs all Federal agencies to utilize their authorities in furtherance of the purposes of the
9-2
-------
act by carrying out programs for the conservation of species listed pursuant to the act. This
section makes it clear that all Federal agencies should participate in the conservation and
recovery of listed threatened and endangered species. Under this provision, Federal agencies
often enter into partnerships and Memoranda of Understandings with the U.S. Fish and Wildlife
Service or the National Marine Fisheries Service for implementing and funding conservation
agreements, management plans, and recovery plans for listed species.
Section 7(a)(2) states that each Federal agency shall, in consultation with the Secretary
(of Interior and Commerce), ensure mat any action they authorize, fund, or carry out is not likely
to jeopardize the continued existence of a listed species or result in the destruction or adverse
modification of designated critical habitat. In fulfilling these requirements, each agency must
use the best scientific and commercial data available. This section of the act defines the
consultation process, which is further developed in regulations promulgated at 50 CFR 402.
The Migratory Bird Treaty Act (16 U.S.C. 703-712; 703-712, as amended) established
a prohibition, unless a permit is issued, to kill, attempt to take, possess, sell, or capture any
migratory bird. The act protects migratory birds and any part, nest, or egg. The U.S. Fish and
Wildlife Service administers this act and serves as the lead agency in the protection of these
i
animals. The act applies to migratory birds affected in spills and hazardous releases.
Many other statutes and executive orders apply in the event of a spill; however, the above
legislation outlines the major regulations that govern response activities and related operations.
Each agency has implemented response planning in a slightly different manner; therefore, the
exact response will be different depending on which agency has lead responsibility.
9.3. USE OF THE RISK ASSESSMENT PROCESS IN ACCIDENTAL RELEASES
CERCLA section 107(a)(4)(c) establishes liability for damages for injury to, destruction
of, or loss of natural resources, including the reasonable costs of assessing such injury,
destruction, or loss. Natural resources are defined to include land, fish, wildlife, biota, air, water,
ground water, and drinking water supplies and other resources belonging to, managed by, held in
trust by, appertaining to, or otherwise controlled by the United States, any State or local
government, any foreign government, or any Indian tribe. The statute also directs the President,
through designated representatives, to act on behalf of the public as a trustee of such natural
resources. Natural resource damage assessment is the process used to assess damages to natural
resources from releases of oil or hazardous substances and to obtain compensation to restore
injured natural resources and their services. The damage assessment process used by natural
resource trustee agencies is guided by a series of regulations. Under the NCP, natural resource
trustees are defined to include States, tribes, and five Federal agencies (the Departments of
9-3
-------
Commerce, Interior, Agriculture, Energy, and Defense). Note that EPA is not a trustee for
natural resources. (For further discussion, see Section 5.2.5.)
Natural resource trustees have technical expertise that can improve design of ecological
risk assessment studies, facilitate the interpretation of data and results, improve remedy selection
ft J:. * :,.'!!- " ;i,, ili'1 ill'' 'i J'li ,' ' i"1'11", ' :;"; ' ',. " • ,- ' • • • ,•• ",'!' ". • T'i "* •::• ^,u MI, ., . • , • , , J »
and design, and design and implement more effective remedial restoration. By incorporating
natural resource concerns into the remedial process, residual injury (and the associated liability
of the responsible parties) can be minimized, and the need for damage assessment activities can
be eliminated.
Numerous books, Government documents, and peer-reviewed and gray literature have
addressed the issues surrounding ecological risk assessment, and the process and procedural.
guidelines have been presented in many forums by multiple Government agencies (Wentsel et
al? 1996; U.SJBP4 1992, 1994,1996) and private individuals (Suter et al., 1983; Barnthouse et
al., 1990; Calabrese and Baldwin, 1993). Currently, little guidance exists that specifically
identifies the role of ecological risk assessment in an accidental release situation. Many of the
Hirl: •;•:', ''M: ?s!|l 'i " i;. , • & ,.• !,!••:.. ' , > ",! ;.f ., ,:lfi{; '•'(; Nl';|.i" . • •• r..
processes outlined in the above documents may have a direct bearing on spill and hazardous
waste site assessment and ecological evaluations. Many of the problems associated with the use
of ecological risk assessment at a hazardous release or spill involve the need for protecting
human health and safety in an expeditious manner, so there is not adequate time to implement an
ecological risk assessment in the traditional sense. Therefore, the ecological risk assessment
;,:,', , !, , mill' ,'i'!Fi , . , : ' • ,','.: ' , ' ' ' !,T, c7 '
process must be evaluated specifically for an emergency planning response.
In terms of the ecological risk assessment process, natural resource managers must (1)
know the location of important resources, however they are defined, and be able to identify
ii!,,; h ' i'ff ;/!$ ! , • , ; ; • '; f , . '! , ', ' '..;.. »'••• .. !('>•• ; '.; . Mi,'1 I ''I'.:
probable threats to the resources of concern (problem formulation); (2) identify the possible
impacts from the material that might be released (exposure characterization); and (3) know the
probable ecological impacts to the species of interest (risk characterization). Discussion between
the resource managers and the risk assessors will lead to a plan for protection of the biological
resources, including determining what the tradeoffs might be for different spill scenarios (risk
•jjf ! ''.i, "''!?:; 'WS! !• |; ' ?: " •, ' :' . ' '•' ,.''•'. 'I"' ' B ;•(. , a ii:1' " :• •',"•
mahagenieht). If the outcome of the management alternatives does not adequately protect the
biological resources, then alternatives can be considered and discussion continued until an
adequate response plan is established. The value of the ecological risk assessment process exists
in both preparedness planning and in the actual site-specific assessment of a given spill or
release.
9-4
-------
9.4. EXAMPLES OF ACCIDENTAL RELEASES
The dynamics of an emergency situation do not lend themselves to drafting written plans
leading to a site-specific ecological risk assessment. Typicallyj decisions must be made rapidly,
with every practical effort being made to coordinate inputs from many parties. Following
reporting of a spill or accidental release, a Federal on-scene coordinator (FOSC) takes charge. In
the case of a spill in inland areas, the FOSC is usually from the EPA Regional Office. For spills
in the marine or estuarine environment, the FOSC is from the Coast Guard. At the request of the
FOSC, usually depending on the nature and size of the event, a natural resource trustee (either
Federal or State or both) is responsible for providing liaison with the FOSC in order to represent
fish, wildlife, and sensitive environmental concerns relative to response activities.
The trustee will provide immediate input and identify the areas of greater ecological risk
to the FOSC. Recommendations relative to the protection of and/or response to these sensitive
areas will be submitted to the FOSC to mitigate any adverse ecological impact. The
recommendations also may initiate the collection of collateral data to determine if it is necessary
to develop a restoration plan. Biotic and abiotic collections may be required to establish,
quantify, and document adverse effect and pathways caused by the discharge or release. These
activities equate to problem formulation in the ecological risk assessment process and the
identification of complete exposure pathways.
One case study and several spills are presented that depict the range of planning and
management activities that have occurred in response to accidental releases.
9.4.1. Case Study: Patricia Sheridan Release of Contaminated Dredge Material
At 1:40 a.m. on October 12,1995, the 106-meter-long jbarge Patricia Sheridan was
intentionally grounded because of an extreme list developed during a storm approximately 2
miles seaward of Charleston Harbor and approximately 150 meters southwest from a Federal
navigation channel. The barge was carrying approximately 12,000 tons of dredge material
contaminated with dioxin and other hazardous substances from the Rowland Hook Marine
Terminal, Staten Island, NY, to an off-loading facility hi Corpus Christi, TX, for subsequent rail
car transport to a disposal facility in Utah. While the barge was listing hi approximately 10
meters of water, two hatch covers opened over the cargo. During the storm and, to some degree,
subsequently, an estimated 2,500 tons of dredged material spilled into the sea. Initially, the U.S.
Coast Guard responded to the incident as a salvage operation because of navigational safety
concerns. Most natural resource trustees were notified of the release on October 12.
9-5
-------
i, ' '" •* L i" MI '• , , , ":»• ,' •' ' , i' i,
The natural resource trustees considered a number of concerns in developing an
evaluation of potential risks to public welfare, including ecological risk, arising from this
incident. Screening-level values for the following environmental media were determined:
Surface water: This concern stemmed from the discharge of dioxin-contaminated
• • •• i" •
effluent water generated during the raising of the sunken barge. State and Federal water
quality criteria for the protection of aquatic organisms were the measurement endpoint.
Promulgated criteria/standards for dioxhi do not exist; however, a chronic toxicity value
for marine organisms for total polychlorinated biphenyls (PCBs) is 0.03
Sediment: Promulgated criteria/standards for dioxin, PCBs, or metals do not exist;
however, in lieu of sufficient site-specific data on toxicity, resource agencies rely on other
widely accepted screening guides, including effects range low (ERL), effects range
medium (ERM), and apparent effects thresholds (AETs).
• Biological tissues: Promulgated criteria/standards for fish or shellfish or other wildlife
for ecological effects do not exist In lieu of direct effects data, resource agencies may
rely on contaminant values in wildlife that if consumed by humans would pose
'• • • '•. - • i r *
unacceptable risk to humans, for example, Food and Drug Administration action and
tolerance levels. While concentrations below these levels are not considered by resource
agencies as necessarily protective of ecological receptors, higher concentrations are
suggestive of harm. In addition, if these concentrations are found in fish, shellfish, or
wildlife, the potential commercial, recreational, or subsistence value (i.e., natural resource
1 jii'l n , i'lilf ' i1 ' • i ' , , ' , , ,'! i ' • '" i/i, ' ' i'i 1
services) of these resources is significantly diminished.
The potential public welfare elements at risk were identified as:
• Recreational fisheries
'•'' ' 'I1 i ' •
• Nonrecreational fish, shellfish, and other aquatic life
• Ongoing and potential Federal/State actions, including Superfund sites, affected by
inadequate characterization of extent of contamination and risk
• Marine transportation services
• Designated ocean dredge materials disposal site
9-6
; ji I.-
• ill1' 'ill,' ,,l|ll|,i',il|lililll ,,;|l|||
-------
• Resources beyond the 12-mile territorial sea
• Resources beyond the 200-meter economic exclusion zone.
The inclusion of professional and experienced ecological risk assessors during emergency
and time-critical incidents generally should be sufficient to account for all the prescribed
elements considered in a remedial setting.
On October 13, State and Federal natural resource trustees received a copy of the
Department of the Army Corps of Engineers (ACE) dredging permit for the Rowland Hook
Marine Terminal and accompanying sediment sample and bidassessment analysis data that
served as the basis for the permit conditions. The data indicated that levels of dioxins, PCBs,
and several metals exceeded some ecological effects screening values, such as ERLs and ERMs.
However, the permit material provided suggested to the trustees that the primary contaminant of
concern was the dioxin congener 2,3.,7,8-TCDD. The permit'stated, "As a result, the
bioassay/bio-accumulation testing for dioxin [2,3,7,8-TCDD] indicates that the proposed dredged
material does not meet the criteria for unrestricted ocean disposal." Concentrations for sediment
characterization in only three samples analyzed for dioxin ranged from 74 parts per trillion (ppt)
to 140 ppt Test sediment concentrations ranged from 8.5 to 39 ppt. Concentrations above 10
ppt dry weight 2,3,7,8-TCDD toxic equivalents in sediments are considered by some as
potentially threatening. The dredge material failed to pass criteria allowing ocean disposal under
the Marine Protection, Research and Sanctuaries Act, and consequently permit language stated,
"It [dredged material] shall not re-enter ocean waters."
The natural resource trustee agencies, including the South Carolina Department of Marine
Resources, the U.S. Fish and Wildlife Service, and NOAA, requested the response agency to
treat the incident as a CERCLA release because of its potential threat to public welfare, which
included the bioaccumulation/biomagnification potential in natural resources such as fish and
shellfish and associated impacts to recreational fishing. In addition, uncontrolled releases of
dioxins could confound ongoing CERCLA actions in the Charleston Harbor area, specifically as
they might relate to culpability. It was determined that exceeding a reportable quantity was not
necessary to initiate a response action.
The Patricia Sheridan was 1 of 11 such barges transiting dredge materials from Howland
Hook, and no analytical data were known on the specific contents of any single barge load.
Because dioxins from the barge may comingle with dioxins already in the immediate
environment from other sources, including National Priority List (Superfund) sites, the risk
assessors successfully argued that congener-specific data might be the only way to assign
9-7
-------
fi'.li' ' il
ConsidermS ^ of these arguments, the response agency requested the potentially
responsible party (PRP) to provide such information to characterize the nature of continuation.
The response agency requested an incident-specific regional response team conference to
discuss the issue of discharge of the water contained in the cargo hoppers, which was necessary
to facilitate salvage. The hoppers contained an estimated 489,000 gallons of seawater. The State
, ,,,,,,1,
of South Carolina stated that its criterion for total dioxin of 1.2 ppq (parts per quadrillion) should
not be exceeded in discharge effluent waters. Because no Federal water quality criteria for
dipxins existed, the risk assessors considered the State criterion as a screening risk value for
*;"-"< "! . •"''.' 'I*1'"' . '• " ",',,,' ', ', ,' l!'1!!,;' !'!! i|,,,
marine surface water protection. It was requested that unfiltered water be analyzed, as dioxin
wquld likely be adsorbed to particulates and therefore not be free phase.
A Captain of the Port Order was issued requiring sampling and analysis of sediments for
constituents of concern in the area outside the barge. Results indicated that some metals and
dioxins released from the barge were now located on the seafloor. The U.S. Coast Guard, with
acknowledgment by the trustees, determined that the release represented an imminent and
substantial danger to public health and welfare.
Of greatest concern to one trustee agency, NOAA, was the potential contamination of the
federally maintained navigational channel. Under the NCP, NOAA's natural resource trusteeship
extends to natural resources managed or controlled by other Federal agencies and that are found
in, under, or using waters navigable by deep-draft vessels. Given that the ACE manages,
controls, and maintains Federal navigation into the Port of Charleston, NOAA expressed concern
"'• ' "". : • •' • • • •' ••• •••• : ;; I1" • •
over the potential threat to transportation services provided by surface waters. Threats included
the potential inability of the ACE to continue to dispose dredge materials from the Federal
navigational channel at the ocean materials disposal site, possible delays in maintenance
dredging associated with the need to find alternative disposal solutions, increased costs
associated with disposal at alternative disposal sites, and potential interruptions in commercial
and Department of Defense waterborne transportation.
The PRP contract laboratory reported 2,3,7,8-TCDD in barge water at 5.4 ppq and a total
dioxin concentration of 10.0 ppq. Again, the State of South Carolina criterion for total dioxin
was 1.2 ppq. Therefore, discharge of this water would violate State water quality criteria and
potentially present an unacceptable risk to marine aquatic resources.
The FOSC issued an Administrative Order on January 10, 1996, requiring the removal of
3.1 cm of sediments over a 4-acre area on the sea floor. To evaluate the effectiveness of the
'!,!i, . ' • „, ."..i1! !!!• l!!i|!|l'| ' , , , ' '",'',, ifiji'i1 ; I1 I J , » ' I1, : , i
remova* act*on> the U.S. Coast Guard and the natural resource trustees required pre- and
postbiological tissue analyses and postsediment sampling for 17 dioxin congeners. A second
I
round of postbiological tissue sampling, as in the first round of sampling, revealed a lingering
9-8
-------
potential problem at the site. As a noteworthy anecdote, the sampling identified a wider area of
dioxin contamination in the Charleston area offshore waters.
More than 4 acres and approximately 5,460 cubic yards around the grounding site were
dredged to remove the upper 3.1 cm of sediment. The preliminary results of the biota sampling
were presented by the PRP along with the quality assurance/quality control data. The data
demonstrated failure of the response to meet the conditions set by the FOSC in consultation with
the trustees. The FOSC ordered that a second set of samples be collected and analyzed. The
FOSC ordered the PRP to fulfill the requirements of the Administrative Order by continuing the
necessary biota monitoring. A second set of data also failed to meet the conditions set by the
trustees and the FOSC. However, the FOSC in consultation with the trustees determined that the
cost associated with additional removal would not be justified by reductions in threat, and
therefore, the response effort was determined to be ended. The natural resource trustees were
now in a position to determine whether or not residual conditions warranted an NRDA action for
restoring lost resources and services.
In August 1996, the U.S. Coast Guard consulted with the trustees regarding the second
set of biota monitoring data and wrote the PRP that it had met the conditions of the January 10,
1996, Administrative Order and that therefore the emergency response phase "...is hereby
rescinded."
This incident demonstrated that although marine accidents may begin as emergency
responses, they can develop into time-critical or nontime-critical responses that require greater
levels of planning and coordination. In addition, pre- and postsediment and biological tissue
sampling enabled the response agency and the natural resource; trustees to properly evaluate
response effectiveness, a step usually necessary to provide the trustees with sufficient
information on which to base a decision for pursuing NRDA action leading to restoration of lost
resources or services. '•
9.4.2. Types of Accidental Releases
9.4.2.1. John Day River Acid Spill '•
On February 8,1990, a tanker truck owned and operated by Thatcher Trucking Company
of Salt Lake City, UT, skidded off Highway 395 and rolled down an embankment into the North
Fork of the John Day River in north-central Oregon. The accident occurred near the town of
Dale, just south of the Camas Creek Bridge and immediately below the mouth of Camas Creek at
river mile (RM) 56.8. The contents of the tanker, approximately 5,000 gallons of 35.2%
hydrochloric acid, began leaking through a ruptured diaphragm in the pressure valve. An
estimated 3,500 gallons, or 33,500 pounds, of the acid discharged into the river and flowed
9-9
-------
'I11!1' V
I,*1!' ,
downstream at;an approximate rate of 1 mile per hour, causing substantial change in the acidity
i,i.:.< • ..<.,i, i 'i i , i. l| i i i .
of the river.
Natural resource trustees with the authority for managing and protecting natural resources
iq the impacted area include the Department of the Interior (DOI), represented by the U.S. Fish
and Wildlife Service and the Bureau of Indian Affairs; the State of Oregon, represented by the
Qregon Department of Fish and Wildlife (ODFW); and the Confederated Tribes of the Umatilla
ill;;: v "^ • ,.,;, ' •, „;i,:!,, ^(i 'i ; • , , ,„ , i ,i '"':',' „•,;,!i1 in',: ,., ,i .. i H. ' 4 ! ivi: , • , .„ ,
Indian Reservation (CTUIR). Assessment of natural resource injuries and development of a
restoration plan was coordinated between the trustees and the Confederated Tribes of the Warm
•• ••" '•' ••' • •• •• "< i
Springs Reservation of Oregon (CTWS).
Historically, the John Day River was one of the most significant anadromous fish-
producing rivers in the Columbia River watershed. The John Day River Basin continues to
support one of the largest remaining runs of wild spring chinook salmon (Oncorhynchus
tshawytscha) and summer steelhead trout (Oncorhynchus mykiss) with populations estimated to
range from 3,000 to 4,000 spring chinook salmon and 30,000 to 35,000 summer steelhead. The
basin also supports a population of Pacific lamprey (Lampetra tridentatd) as well as other
indigenous species. The management policy for the John Day River Basin is designed to
maintain native wild stocks of salmon and steelhead and to preserve the genetic diversity of the
native salmon and steelhead stocks for maximum habitat use and fish production.
The basin also supports a variety of resident fish species, such as rainbow trout
(Oncorhynchus mykiss), brook trout (Salvelinus fontinalis), bull trout (Salvelinus confluentus),
cutthroat trout (Salmo clarki), mountain whitefish (Prosopium williamsonf), channel catfish
(Ictalurus punctatus), and smallmouth bass (Micropterus dolomieuf). Bull trout are of particular
concern because they are a Federal candidate species petitioned for listing as threatened or
endangered under the Endangered Species Act. Other resident species common to the area
. &.'•: f . 'Sri •:',".': •. , »•" • i'-1?/1,1 '": ir: i ;. ill I :;, , i .;v
include chisehnouth (Acrocheilus alutaceus), suckers (Catostomus spp.), redside shiner
(Richardsonius balteatus), longnose dace (Rhinichthys cataractae), sculpins (Cottus spp.), and
northern squawfish (Ptychocheilus oregonensis).
Numerous natural resources within a minimum 12-mile stretch of the North Fork of the
jitf1"'1' ' • "::i " ' i"1" ' .'' ' • •'• • ! II il , ! ' - '
John Day River were injured as a result of the hydrochloric acid discharge. The pH of the river
was lowered from a normal background level of 8.0 to 2.4 (Oregon Department of
Environmental Quality, 1990). Hydrological modeling showed that the river did not have the
'*! ;• ,.?,•' . ' i '•••„;.:• "! •• :it1 i"
acid neutralizing capacity to recover to a pH of 6.5 even at RM 15.3, which is 41.5 miles
downstream 9fthe spill site. Thus, natural resources under the trusteeship of the DOI, the State
of Oregon, the CTUIR, and the CTWS were adversely affected by the acid spill.
9-10
; U, ILs i i !' :; •" W\ I. "£ '••'. ", ^ liMJ/. Ilililiil! • tii.i.\ ^.:,; .1 ii':..!,ll;:i:;:' i, M! • i.. I;',..': a, ibiJL. -. •. •. i; ..:ii: '. i .« i; !i' •::' • A'&HM .i. i« l'= -•*•.! i';liiS • h^v •> it i * "S •&• >••<•:< *• '•> i^KmLs
-------
The sensitivity offish to acidic; water varies, but pH levels below 6.0 can be detrimental
to many species (Haines and Baker, 1986; Gloss and Schofield, 1989; Wiener and Eilers, 1989).
Exposure to the acid was manifested in fish by burned, blistered, or discolored skin; singed fins;
bleeding gills; loss of scales; cloudy eyes; internal bleeding; and severe behavioral distress.
ODFW (1990) and Dougan (1990) estimated 98,000 to 145,000 fish were destroyed, including
4,000 anadromous fish, 300 bull trout, and 9,500 Pacific lampirey. The loss of 300 bull trout hi
the river is especially critical because the spill may have destroyed a large portion of the adult
bull trout population in this area (H. Li, personal communication, 1990). Although bull trout
primarily occur in the upper tributaries of the John Day River Basin, they seasonally utilize the
North Fork of the John Day River in the winter. The spill occurred at a time when a large
portion of the adult bull trout population was probably in the North Fork of the river.
In addition to adult fish, an estimated loss of 50% of the chinook salmon alevins was
reported (ODFW, 1990; Dougan, 1990). This estimate was based on a quantitative aquatic
invertebrate analysis that showed a 50% loss of invertebrates in the first mile below the spill site.
Aquatic invertebrates provide an essential food resource for many species of resident and
anadromous fish as well as other species. A reduction in aquatic invertebrate abundance had a
short-term impact on food availability. Long-term loss of natural production of salmonid species
and complete annihilation of at least one age class of locally spawning salmon and steelhead
occurred from the spill. Additionally, although direct mortality offish was not documented in
surveys beyond 12 miles downstream from the spill site, chronic effects most likely occurred in
these areas. \
Aquatic mammals, waterfowl, and endangered species that utilize the John Day River
Basin also may have been directly or indirectly impacted, by the spill. Loss offish from the
North Fork John Day River could have affected wintering bald eagles (Haliaeetus
leucocephalus), mink (Mustela visori), and river otter (Lutra canadensis) that are known to
forage in the river. Peregrine falcons (Falco peregrinus; an endangered species) nest in the
basin, and numerous waterfowl species use the river, including Canada geese (Branta
canadensis), common (Mergus merganser) and hooded mergansers (Lophodytes cucullatus),
mallard (Anas platyrhynchos), gadwall (Anas streperd), .American widgeon (Anas americand),
wood duck (Aix sponsd), and green-winged teal (Anas creccd). All of these species may have
been indirectly affected by the spill through destruction of their food base; changes hi foraging,
shelter, breeding, and rearing areas; or other factors essential for long-term survival.
In addition to fish and wildlife resources, the river supports significant tourism, hiking,
camping, subsistence fishing and trapping, and commercial arid sport fisheries. The impacted
area also has important cultural and archaeological values to the local Indian tribes. Tribal
9-11
-------
subsistence fishing in tributaries in the John Day River Basin and mainstem Columbia River
provides a culturally important food source for the tribes. Pacific lamprey, salmon, and other
indigenous species such as whitefish, suckers, and chiselmouth have been essential food fish for
the tribes of the John Day River Basin for centuries. The capacity of the river system to support
these consumptive and nonconsumptive activities may be reduced for many years as a result of
the spill.
The John Day River Basin is managed to maintain wild salmon populations, with no
„ , „ i.
enhancement through the release of hatchery stock. A management plan has been developed by
ODFW and the tribes to oversee this objective. Because of this management policy, no short-
term remedial actions, such as restocking with hatchery-reared fish, could be used to restore
resources lost during the acid spill. Recovery of damages in the form of habitat restoration
actions hi the John Day River Basin that are consistent with management plans for the area were
sought from the responsible party. Appropriate restoration actions will improve conditions in the
riyei; to promote fish and wildlife production lost due to the acid spill. In addition to the
i,!, •• i i ii, iJiii : • ';. . ;" • : : • ; «,. 'I}*., i /HI ••!;;" . . '.: ••:
mainstem North Fork, restoration efforts should be directed to the tributaries such as the Middle
Fork John Day River, Camas Creek, and Desolation Creek. Providing unproved habitat for fish
will aid in replenishing the injured resources, increase the survivability offish not killed during
the acid spill, and aid in replenishing the natural population by increasing productivity. In
addition, restoration projects will increase egg-to-smolt survival, increase smolt-carrying
capacity, provide more and better habitat for juvenile fish rearing, and increase pre-spawner
survival. Recovery of lost resources will not happen quickly; completion of restoration actions
and full recovery of the fish populations could take 10 or more years.
9.4.2.2. North Cape Oil Spill
On January 19,1996, the tug Scandia caught fire while it was towing the tank barge
North Cape hi the coastal waters off Rhode Island. The tug was abandoned and storm-force
winds grounded it and the barge off Moonstone Beach, approximately 3 miles west of Point
Judith, RL Approximately 820,000 gallons of number 2 fuel oil leaked from the damaged barge,
impacting coastal and marine habitats, including a national wildlife refuge and other sensitive
areas. Heavy surf hampered efforts to contain the spill, and officials reported that thousands of
lobsters, clams, and other invertebrates and hundreds of birds were killed. A 250-square-mile
ajea of Block-island Sound and seven coastal ponds were closed to fishing.
On January 20,1996, the EPA Environmental Response Team (ERTC) was activated to
ptovide technical assistance to the FOSC. EPA personnel from the Region 1 Emergency
Removal Program, the Region 1 Laboratory in Lexington, MA, and the Office of Research and
*; •:*: 9-12 ;'' , !
-------
Development (ORD) Laboratory in Narragansett, RI, also responded to provide support to the
U.S. Coast Guard, the National Oceanic and Atmospheric Administration, the State of Rhode
Island, and DOI to support assessment of the spill. EPA activities were focused on sampling and
analyzing sediments, the water column, and aquatic organisms to determine fate and effect of the
spilled oil while also providing technical support on methods to contain the released fuel.
Water quality concerns were addressed by a sampling effort conducted jointly by EPA
Region 1, ORD-Narragansett, and ERT. The natural resource trustees, both State and Federal,
also sampled biota and environmental media and rehabilitated oiled birds. The trustees formed
four technical working groups to evaluate injury and identify potential restoration opportunities.
Three groups investigated effects to natural resources, including marine communities, salt ponds,
and birds. A fourth group (economics) helped the other groups scale injury and restoration, as
well as determine economic losses.
There was an intense acute toxicological response in the benthic community hi the near-
shore area close to the grounding site. Severe weather conditions aggravated the toxicity of the
oil via complete entrainment and dispersion of the oil into the water column. There were
concentrations of 1 to 6 ppm total petroleum hydrocarbons throughout the water column to
depths of 20 meters. The trustees estimated that millions of lobsters, surf clams, crabs, and
amphipods were killed. In total, 26 species of finfish and large invertebrates were identified
among the dead organisms in beach standings. :
Seven salt ponds were exposed to North Cape oil and potentially impacted from the spill.
Extensive mortality of infauna (primarily amphipods) occurred in several of the ponds. Winter
flounder were exposed to levels of oil that cause sublethal effects and could potentially reduce
their reproductive output. Shellfish, including oysters, mussels, and soft-shell clams, were
exposed to oil, although acute mortality was limited. Brackish wetlands and salt marsh
communities were exposed to oil; however, no significant injury to salt marsh vegetation was
measured.
Investigators recovered about 400 dead birds (primarily waterfowl, loons, and grebes).
To account for birds that were never found because they sank; drifted out to sea, or were
scavenged, the trustees applied a multiplier of 6 to the total number of birds recovered, resulting
in an estimate of approximately 2,300 dead birds (nonwater birds were not included in the
multiplier). This multiplier was based on a qualitative analysis of factors influencing oil
spill-related bird mortality. These factors included the weather conditions, location of the spill,
oil characteristics and volume, and number of birds potentially exposed to oiling. The trustees
also assessed impacts to the federally threatened piping plover, which breeds on Moonstone
Beach. Piping plover productivity at Moonstone Beach dropped 37% in 1996 compared with
9-13
-------
iTt
1995. Declines in piping plover productivity at Moonstone were in contrast to other sites in
Rhode Island, where productivity increased 6% in 1996 compared with 1995.
9.4.2.3. Conoco Marine Terminal 1,2-Dichloroethane Spill
in i..< ' , Hi JlJl'MII , i' . , I , ^ , , , ,. , , , . ,|
In March 1994, Conoco discovered that a pipeline from its marine terminal on the
Clopney Loop of the Calcasieu River in Louisiana was leaking 1,2-dichloroethane (EDC).
Although the company undertook emergency response operations that it believed were
successful, gross EDC contamination was discovered in late May 1994 during routine sediment
sainpling. This discovery caused Conoco to notify the National Response Center (NRC) of the
additional amount of the release.
NOAA provided technical advice to the FOSC, conducted reconnaissance and ecological
ri£S assessment sampling, and worked with the responsible party (Conoco) to review the past
work that Conoco had undertaken and the proposed work plan. NOAA also worked to have
ecological investigations conducted in the Clooney Loop. These investigations determined that
benthjc communitigs probably had been impacted by the release, but that those impacts were
localized and not widespread and that the contaminant would attenuate naturally over time. This
work reinforced the State's decision to allow natural degradation to complete restoration of the
Clooney Loop rather than requiring extensive dredging. This resulted in a cost savings to
Conoco of approximately $20 million.
NOAA assembled the natural resource trustees concerned wifh the release to facilitate
de¥e lopment of a negotiated settlement subsequent to CERCLA removal. The trustees, led by
NOAA, began developing a cooperative strategy to assess injuries and acquire compensation.
NQAA continued tQ participate in the response by reviewing reports, providing advice to the
FOSC, and facilitating expeditious cleanup. Conoco presented a proposal at a meeting in March
1 995 to use habitat equivalency analysis to scale a restoration project 9 river miles from the
Clooney Loop. NOAA worked with the trustees and Conoco to refine and finalize the proposal.
As a result, all parties agreed to a project that cost-effectively would restore the natural resources
of the Calcasieu estuary harmed by the EDC release.
In June 1996, Conoco and the natural resource trustees signed an agreement that Conoco,
its agent Stream Management, Inc., would restore 41 acres of transitional wetland
habitat. Of thai; total, 4.5 acres will be maintained in perpetuity. This area will provide
compensation for the lost 20.7 acre-years of services. The remainder (36.5 acres) will be
maintained for 50 years as a natural buffer to protect the ecological functioning of the
compensation tract. Conoco and Stream Management have purchased the property, and the final
settlement document is in preparation. An environmental assessment, a requirement under the
9-14
-------
National Environmental Policy Act, was prepared, submitted for public comment, and finalized.
Due to NOAA's efforts, an ecologically protective cleanup was expedited, and cost-effective
natural resource restoration was begun quickly without resorting to time-consuming, expensive
litigation.
9.5. RISK ASSESSMENT METHODOLOGY AS APPLIED TO ACCIDENTAL
RELEASES
By evaluating the components of an ecological risk assessment, risk assessors and
managers can examine how each element may fit during an accidental release. The intent of the
ecological risk assessment is not only to show that there are adverse impacts but also to
determine at what levels contaminants can exist in the environment while the ecological system
remains protected. The heart of the ecological risk assessment is the problem formulation phase.
Problem formulation is composed of several elements, including selecting assessment endpoints,
developing a testable hypothesis and a conceptual model, and determining measurement
endpoints. Although presented here as a linear process, the problem formulation phase is
iterative, and each element affects the others in the process. These elements form the basis of a
logic tree that allows the investigator to make logical decisions regarding the remediation or
cleanup activity.
Assessment endpoints are essential in the ecological risk assessment and can play an
important role both in preparedness and during the actual assessment following a release or spill.
Assessment endpoints are defined as "an explicit expression of the environmental value that is to
be protected." During a release or during the planning for such an event, the investigator should
identify what elements of ecological concern exist. The first priority during a spill or release is
to determine habitats that have not yet been impacted but are at potential risk. This process is
necessary in order to avoid additional impacts to unimpacted areas. These areas can then be
classified as sensitive or relatively resistant. A sensitive area might include a salt marsh system
whereas a resistant area would include a jetty or high-energy rock cobble beach. By identifying
these areas and their relative sensitivities, the investigator is initiating the process of developing
assessment endpoints. ;
In addition to establishing which areas may be at risk, it is also important to identify the
areas in a spill or release that have already been affected by the contaminant In evaluating these
areas through the development of assessment endpoints, the investigator can determine if these
areas warrant cleanup or remediation and can assess whether the cleanup is effective.
Therefore, through the process of developing assessment endpoints, the investigator will
evaluate habitats that have not been impacted as well as affected habitat areas, and determine
9-15
-------
.,!"." ' . ' i,r II' . ill I ' , , i,!1' ,,''!,, ,| !
assessment endpoints that will be used to determine the ecological function of those areas and
IS: . ! k. :, II I-':;'! ' ' ' ' , ,i • •'. - ' '' ,. I , ' li","i I • ! lit! ,» ', i'f I II „; l|, I , i «| i
W^?t^er ft *$ ^^sary to implement protective measures or remediation to ensure the protection
,;>i ! '. 'I!'1 .1 1 I • 'l>! , - ' . i • " . !';'i : '( ' I'. , , ' ,,
of those habitats.
Through OP A, assessment endpoints have been defined in many coastal areas in the form
of maps to determine the location of sensitive areas of special ecological concern. These areas
n" i . • • i'i i;;, !""!LJ!I ', ,.i :• • •, ' ' • ",'.. ,ir« • • i T, > ,,,ii .•,• • mi • • i • i| M,, • •*<•
represent assessment endpoints. This aspect of the process helps prepare for a spill or release by
Providmg iai^31 response personnel with critical information on the location of areas that should
be protected, this level of information does not formally exist for many inland areas, so
development of assessment endpoints for these parts of the country must be conducted on a site-
specific basis.
Testable hypotheses are specific risk questions that are based on the assessment
endpoints. Based on the mechanism of contaminant toxicity, the number of exposure pathways
that may exist for an assessment endpoint, or other factors, there may be more than one question
for each assessment endpoint. These questions must be answered and statistically evaluated to
reach conclusions relative to the assessment endpoints.
The conceptual model links the contaminants with the sensitive habitats that have been
identified through establishing the assessment endpoints. The conceptual model follows the
contaminailtsX?tressor) in *e environment through ecological (biological) compartments. In the
case of a spill, the conceptual model determines the linkage between air, waterp sediment, and
so*jto ^e recel)tor °f concern, an organism or its habitat. For example, if the contaminant of
fl'l'l •' ' li'JL'li'i l.'JW"1 , " .:, • . ii J'-llfli.
'i I
iltiilliiiiilfca^ ii iiiJIIi.i'liW' III ,ilj
-------
of exposure pathways and to allow establishment of a causal link between the exposure and the
effects. ',
9.6. NEXT STEPS
9.6.1. Ecological Risk Assessment Needs
• Standardize the ecological risk assessment methodology as applied to spills.
• Assess the process used to determine incidental media ingestion.
9.6.2. Contingency Planning
According to the NCP (Section 300.210 c3v4i), area contingency planning consists of
coordination and consultation among individuals, agencies arid businesses that could potentially
be responsible for an accidental release, the U.S. Fish and Wildlife Service, NOAA, and other
interested natural resource management agencies. Area contingency plans should provide for
coordinated immediate and effective protection, rescue, and rehabilitation and minimization of
risk of injury to fish and wildlife resources and habitats. Protection is extended to marine and
freshwater species as well as terrestrial wildlife and includes their habitats and food resources,
whether directly or indirectly affected. This information is utilized in the preparation of the area
contingency plans, which contain valuable information relative to ecological risk. Although the
format and presentation of the risk mformation is different from what may traditionally be used
in formal ecological risk assessment, many of the elements of problem formulation are present in
contingency plans. Thus, reevaluation and updates of area contingency plans to utilize a more
structured ecological risk process are recommended.
9.6.3. Research on Cleanup Methods
Engineering expertise should be applied to develop new equipment and methods for
safely and effectively removing oil and other chemicals from transportation vessels. This will
reduce the continuing release of material at the scene of an accident.
The use of dispersants at the scene of oil spills continues to be controversial, and there is
no consistent policy among Federal agencies as to where, when, or if they should be used.
Dispersants do not remove oil but retard the recoalescence of droplets into slicks, and thus make
it appear that less oil is present. The compounds are proprietary, and basic effects data needed
for risk assessment are not available. Their ecological Impacts are largely unknown, and further
research is needed before policy decisions can be made.
9-17
-------
i,1!'1
9.7. REFERENCES
I" : ! • V i ','5 ' •''''... i i
Bamthouse, LW; Suter, GW, II; Rosen, WE. (1990) Risks of toxic contaminants to exploited fish populations:
influence of life history, data uncertainty, and exploitation intensity. Environ Toxicol Chem 9:297-311.
Calabrese, EJ; Baldwin, LA. (1993) Performing ecological risk assessments. Chelsea, MI: Lewis Publishers.
Wi,. •' i • S.i'i, ,v , i,' .'! ' • ' 'i ' ', • /'. , Vlll'in , ' I 'I i | i 'I, ,,
Dougan, J. (1990) Field notes: chemical spill, North Fork John Day River, February 8, 1990. Umatilla National
Forest, U.S. Forest Service.
Gloss, SP; Schofield, CL. (1989) Liming and fisheries management guidelines for acidified lakes in the Adirondack
region. Biological Report 80(40.27), U.S. Fish and Wildlife Service.
Haines, TA; Baker, JP. (1986) Evidence offish population responses to acidification in the eastern United States
Water Air Soil Pollut 3 1 :605-629.
Oregon Department of Fish and Wildlife. (1990) North Fork John Day River hydrochloric acid (HC1) spill report
February 12, 1990. . . '
-:"! • ..... u ...... -i ' ' • ' ' •<• -i i
Oregon Department of Environmental Quality. (1990) Notice of civil penalty assessment, Thatcher Company May
1, 1990. Portland, OR.
Suter, GW, II: Vaughan, DS; Gardner, RH. (1983) Risk assessment by analysis of extrapolation error. A
demonstration for effects of pollutants on fish. Environ Toxicol Chem 1:369-377.
u-ss Jnyironmentil PrRlFtiop Agency. (1992) Framework for ecological risk assessment. Risk Assessment Forum,
Office of Research and Development, Washington, DC. EPA/630/R-92/001.
U-S, Environmental Protection Agency. (1994) A review of ecological assessment case studies from a risk
assessment perspective, volume II. Risk Assessment Forum, Office of Research and Development Washington DC
EPA/630/R-94/003. '
U.S. Environmental Protection Agency. (1998, May 14) Guidelines for ecological risk assessment. Federal Register
63(93):26846-26924.
' Lap°St, TW; Simini, M; Checkai, RT; Ludwig, D; Brewer, LW. (1996) Tri-services procedural
guidelines for ecological risk assessment, volume 1. U.S. Army, Edgewood Research, Development and
Engineering Center, Aberdeen Proving Ground, MD.
I/i'i! I . viR ' iiii'i'i ' •, - • ••,•''• to •:.'.'. ' ' if I • '::
Wiener, JG; Eilers, JM. (1989) Chemical and biological status of lakes and streams in the upper Midwest:
assessment of acidic deposition effects. Lake Reservoir Manage 3:365-378.
i':,:;
• • i •• i
9-18
-------
10. GLOSSARY (Adapted in part from U.S. EPA, 1996)
adverse ecological effects—Changes that alter valued structural or functional attributes of
ecological entities defined in assessment endpoints. An evaluation of adversity may include a
consideration of the type, intensity, and scale of the effect as well as the potential for recovery.
While risk assessors evaluate adversity, risk managers determine the acceptability of adverse
effects.
assessment endpoint—An explicit expression of the environmental value that is to be protected.
An assessment endpoint includes both an ecological entity and specific attributes of that entity.
For example, salmon are a valued ecological entity; reproduction and population maintenance of
salmon form an assessment endpoint.
biological stressor—As used in this report, synonymous with nonindigenous species - a species
introduced (intentionally or unintentionally) beyond its natural range or natural zone of potential
dispersal. Biological stressors may also include genetically engineered organisms.
characterization of ecological effects—A portion of the analysis phase of ecological risk
assessment that evaluates the ability of a stressor to cause adverse effects under a particular set of
circumstances.
characterization of exposure—A portion of the analysis phase of ecological risk assessment that
evaluates the interaction of the stressor with one or more ecological entities. Exposure can be
expressed as co-occurrence or contact, depending on the stressor and ecological component
involved.
conceptual model—The conceptual model describes a series of working hypotheses of how the
stressor might affect ecological entities. The conceptual model also describes the ecosystem
potentially at risk, the relationship between measures of effect and assessment endpoints, and
exposure scenarios.
ecological entity—A general term that may refer to a species, a group of species, an ecosystem
function or characteristic, or a specific habitat. An ecological entity can be one component of an
assessment endpoint. !
ecological risk assessment—The process that evaluates the likelihood that adverse ecological
effects may occur or are occurring as a result of exposure to one or more stressors.
ecosystem—The biotic community aind abiotic environment within a specified location in space
and time.
ecosystem management—Management driven by explicit goals; executed by policies, protocols,
and practices; and made adaptable by monitoring and research based on our best understanding
of ecological interactions and processes necessary to sustam;ecosystem composition, structure,
and function (Christensen et al. [1996] ref. in Ch. 8). ;
10-1
-------
exposure—The contact or co-occurrence of a stressor with a receptor.
exP,!)Sure Profik—The product of characterization of exposure in the analysis phase of
ecojpgical risk assessment The exposure profile summarizes the magnitude and spatial and
temporal patterns of exposure for the scenarios described in the conceptual model.
exposure scenario—A set of assumptions concerning how an exposure may take place, including
assumptions about the exposure setting, stressor characteristics, and activities that may lead to
exposure.
hazard—As used in this report, hazard refers to the potential adverse ecological effects of a
stressor.
hazard quotient— A ratio of the predicted or estimated levels of a stressor divided by a predicted
or estimated level of the stressor causing a specific effect, e.g., for a chemical, the estimated
environmental concentration divided by the median lethal concentration.
secondary effects— An effect in which the stressoror acts on supporting components of the
ec°system» whi9h "} J^11 have an e£fect on me ecological component of interest.
lines of evidence— Information derived from different sources or by different techniques that can
boused to interpret and compare risk estimates. While this term is similar to the term "weight of
evi" it does not necessarily imply assignment of quantitative weightings to information.
lowest observed adverse effect level (LOAEL)— The lowest level of a stressor evaluated in a
test that causes statistically significant differences from the controls.
measure of ecosystem and receptor characteristics— A measurable characteristic of the
ecosystem or receptor that is used in support of exposure or effects analysis.
mefsure of eff|gt— 4 measurable ecological characteristic that is related to the valued
characteristic chosen as the assessment endpoint.
" , ''"!;; ! " :1:" :"""; "•'',' • ; ;"" ••• ' !: •' '!!:,,' ;
measure of exposure— A measurable stressor characteristic that is used to help quantify
exposure.
„,,,,, , j
!.'.!' . ..... j .....
measurement endpoint— See "measure of effect."
,
median lethal concentration (LC50)-A statistically or graphically estimated concentration that
is expected to be lethal to 50% of a group of organisms under specified conditions (ASTM,
1990).
nonEd%enous sPecies— the condition of a species being beyond its natural range or natural zone
of potential dispersal; includes all domesticated and feral species and all hybrids except for
naturally occurring crosses between indigenous species (OTA, 1993).
10-2
-------
no observed adverse effect level (NOAEL)—The highest levbl of a stressor evaluated in a test
that does not cause statistically significant differences from the controls.
primary effect—An effect in which the stressor acts on the ecological component of interest
itself, not through effects on other components of the ecosystem (synonymous with direct effect;
compare with definition for secondary effect).
problem formulation—The initial stage of an ecological risk -assessment where the purpose of
the assessment is articulated, assessment endpoints and a conceptual model are developed, and a
plan for analyzing and characterizing risk is determined. ;
receptor—The ecological entity exposed to the stressor.
recovery—The rate and extent of return of a population or community to a condition that existed
before the introduction of a stressor. Because of the dynamic nature of ecological systems, the
attributes of a "recovered" system must be carefully defined. :
risk analysis—The process that includes both risk assessment and risk management.
risk assessor—An individual or team with the appropriate training or range of expertise
necessary to conduct a risk assessment (SETAC, 1997) ;
risk characterization—A phase of ecological risk assessment that integrates the exposure and
stressor response profiles to evaluate the likelihood of adverse ecological effects associated with
exposure to a stressor. The adversity of effects is discussed, including consideration of the
nature and intensity of the effects, the spatial and temporal scales, and the potential for recovery.
risk manager—An individual, team, or organization who can make decisions or take action
concerning alternatives for addressing risks. In some situations, risk managers may include a
wide range of interested parties or "stakeholders." (Adapted from SETAC, 1997)
risk mitigation—Actions taken to reduce or eliminate exposure to or effects of stressors.
risk quotient—See "hazard quotient."
secondary effect—An effect in which the stressor acts on supporting components of the
ecosystem, which in turn have an effect on the ecological component of interest (synonymous
with indirect effects; compare with definition for primary effect).
source—An entity or action that releases to the environment or imposes on the environment a
chemical, physical, or biological stressor or stressors.
stressor—Any physical, chemical, or biological entity mat can induce an adverse response
(synonymous with agent). :
10-3
-------
stressor-response profile—The product of characterization of ecological effects in the analysis
phase of ecological risk assessment. The stressor-response profile summarizes the data on the
effects of a stressor and the relationship of the data to the assessment endpoint.
uncertainty—"A lack of confidence in the prediction of a risk assessment that may result from
natural variability in natural processes, imperfect or incomplete knowledge, or errors in
conducting an assessment." (SETAC, 1997)
weight of evidence—See "lines of evidence."
10-4
-------
Abstract
The report entitled "Ecological Risk Assessment in the Federal Government" was prepared by an
interagency work group under the auspices of the Committee on Environment and Natural
Resources (CENR). The objective of the work group was to write a document on the major uses
of ecological risk assessment by Federal agencies. Eight task groups were formed with a total of
32 scientists from 9 Federal agencies. The task groups addressed eight topics: the Federal
Insecticide, Fungicide, and Rodenticide Act (FIFRA), the Toxic Substances Control Act (TSCA),
the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA),
nonindigenous species, ecological assessments hi ecosystem management, agricultural
ecosystems, the endangered/threatened species, and oil spills (accidental releases). The task
groups provided examples of current ecological risk assessment areas (established uses),
potential uses where components of ecological risk assessment are used, and related ecological
assessments and other scientific evaluations that might benefit from the use of ecological risk
assessment methodologies. Recommendations were made to improve the science, enhance
information transfer, and improve risk management coordination.
For additional copies or information contact:
Executive Secretary
Committee on Environment and Natural Resources
National Oceanic and Atmospheric Administration
Office of Policy and Strategic Planning
U.S. Department of Commerce
Washington, DC 20230
(202) 482-5916, fax 202-482-1156
Also available on the NSTC Home Page via link from the OSTP Home Page at:
http://www.wWtehouse.gov/WH/EOP/OSTP/htrnl/OSTP_Home.html
and the CENR Home Page:
http ://www.nnic.noaa.gov/CENR/cenr.httnl
-------
iiiiiiii
•IIII 111 I 111 111 II 111 t^mt Illllll IIIIII III I Illllllllllll Illllllllllll
iiiiiiiiiiiiii iiiiiiii i iiiiiiiiiiii i
Mill
IjIV
Ill 111 I 111 II I 1111 111 I Hill 111 Illllllllllll I
iiiiiiiiiiiiiiiiiiiiiiiiiiii in ill inn i nil iiiiiiiiii 11 i 1,111111111111 iiiiiii iiiiiiii iiiiiiiiiiiiiiiiiiiiiiiii
ill in iiiiiiiiiiii ii iiiiii
iiiiiiii in ii iiiiii iiiiiiiiiiiiiiiiiiiiiiiii
i i
EXECUTIVE OFFICE OF THE PRESIDENT
Office of Science and Technology Policy
Washington, B.C. 20502
IIII III III
illllllll
------- |