RISK ASSESSMENT HANDBOOK VOLUME II: ENVIRONMENTAL

EM 200−1−4

31 December 2010

ENVIRONMENTAL QUALITY

RISK ASSESSMENT HANDBOOK

VOLUME II: ENVIRONMENTAL EVALUATION

ENGINEER MANUAL

AVAILABILITY

Electronic copies of this and other U.S. Army Corps of Engineers

(USACE) publications are available on the Internet at

http://140.194.76.129/publications/. This site is the only

repository for all official USACE engineer regulations, circulars,

manuals, and other documents originating from HQUSACE.

Publications are provided in portable document format (PDF).

Printed on Recycled Paper

CEMP-RT

Manual

DEPARTMENT OF THE ARMY

Army Corps of Engineers

Washington, D.C.

20314-1000

EM 200-1-4

No.

200-1-4

December 2010

Environmental Quality

RISK ASSESSMENT HANDBOOK

VOLUME

II:

ENVIRONMENTAL EVALUATION

Purpose. The overall objective of this manual is to provide risk assessors with the

recommended basic/minimum requirements for developing scopes of work, evaluating

Architect-Engineer prepared ecological risk assessments, and documenting risk

management options associated with Hazardous, Toxic, and Radioactive Waste

(HTRW) and Military Munitions Response

Program investigations, studies, and designs

consistent with principles of good science in defining the quality of risk assessments.

This EM is intended for use by

U.S. Army Corps

Engineers (USAGE) Project

Managers, technical personnel, and contractor personnel.

Applicability. This

applies to all HQUSACE elements and USAGE commands

responsible for HTRW projects and Military Munitions Response

Program projects.

References. References are listed

Appendix

Distribution. Approved for public release, distribution is unlimited.

Discussion. This manual is intended to provide USAGE risk assessors and

contractor personnel with supplemental guidance for performance and evaluation of risk

assessments under the Comprehensive Environmental Response, Compensation, and

Liability Act (CERCLA)

amended by the Superfund Amendments and Reauthoriza-

tion Act

(SARA)

1986, and the Resource Conservation and Recovery Act (RCRA)

amended by the Hazardous and Solid Waste Amendments (HSWA) of 1984. It

not

intended to replace the accepted guidance

the USEPA (e.g., Ecological Risk

Assessment Guidance for Superfund: Process for Designing

and

Conducting Ecological

Risk Assessment),

but should

used

conjunction with that document. Additional

information provided

this manual concerns presentation of the risk assessment

results for use in risk management and decision-making,

·concerns focusing on the

decisions, and criteria needed for decisions. Both risk and non-risk factors are

presented for consideration by the risk managers.

FOR THE COMMANDER:

2 Appendices

(See Table of Contents)

lonel, Corps

Engineers

Chief of

Staff

This manual supersedes

200-1-4, Volume

II,

dated

June 1996.

EM 200-1-4

31 Dec 10

-2-

THIS PA

GE INTENTIONALLY LEFT BLANK

DEPARTMENT OF THE ARMY EM 200-1-4

U.S. Army Corps of Engineers

CEMP-RT Washington, D.C. 20314-1000

Manual

No. 200-1-4 31 December 2010

Environmental Quality

RISK ASSESSMENT HANDBOOK,

VOLUME II: ENVIRONMENTAL EVALUATION

TABLE OF CONTENTS

Paragraph Page

Chapter 1. Introduction

Purpose and Scope ....................................................................................... 1.1 1-1

Applicability ................................................................................................... 1.2 1-1

Distribution Statement ................................................................................... 1.3 1-1

References .................................................................................................... 1.4 1-1

Introduction ................................................................................................... 1.5 1-1

Intended Audience and Use .......................................................................... 1.6 1-2

USACE Role in the HTRW Program ............................................................. 1.7 1-3

Overview of the HTRW Response Process ................................................... 1.8 1-3

Role of Ecological Risk Assessments in the HTRW Process ........................ 1.9 1-3

PA/SI or Other Preliminary Site Investigation Activities ................................. 1.9.1 1-3

RI or Other Additional Site Investigation Activities. ........................................ 1.9.2 1-4

FS, RD/RA or other Remedial Design and Implementation Activities ............ 1.9.3 1-4

Use of Ecological Risk Assessment in Special Studies ................................. 1.9.4 1-4

Ecological Risk Assessments as Decision Criteria in the HTRW Program .... 1.10 1-5

Concept of Risk Assessment and Good Science .......................................... 1.11 1-5

Chapter 2. Planning and Scoping an Ecological Risk Assessment

Introduction ................................................................................................... 2.1 2-1

Purpose of the ERA ...................................................................................... 2.2 2-2

Objectives of the ERA. .................................................................................. 2.3 2-2

Planning Considerations ............................................................................... 2.4 2-2

Process Discussions ..................................................................................... 2.4.1 2-3

Coordinating ERA and HHRA Planning ......................................................... 2.4.2 2-3

Planning for an ERA ...................................................................................... 2.4.3 2-4

HTRW Technical Project Planning Process .................................................. 2.4.4 2-5

Site-Specific Consideration ........................................................................... 2.4.5 2-6

Establishing the Level of Effort ...................................................................... 2.5 2-6

Approach and Level of Effort ......................................................................... 2.5.1 2-6

SLERA (Steps 1 and 2 of ERAGS) ................................................................ 2.5.2 2-6

BERA (Steps 3 through 8 of ERAGS) ........................................................... 2.5.3 2-7

Risk-Based Analysis of Remedial Alternatives .............................................. 2.5.4 2-8

Scoping an ERA ............................................................................................ 2.6 2-8

Guidance ....................................................................................................... 2.7 2-9

Department of Defense ................................................................................. 2.7.1 2-9

EM 200-1-4

31 Dec 10

Paragraph Page

U.S. Army ...................................................................................................... 2.7.2 2-9

USEPA (Superfund) ...................................................................................... 2.7.3 2-10

USEPA (Agency-wide) .................................................................................. 2.7.4 2-10

USEPA (Regional) ........................................................................................ 2.7.5 2-11

State ............................................................................................................. 2.7.6 2-11

Chapter 3. Problem Formulation

Introduction…................................................................................................. 3.1 3-1

Problem Formulation – SLERA ..................................................................... 3.2 3-1

Reconnaissance ............................................................................................ 3.2.1 3-1

Receptors ...................................................................................................... 3.2.2 3-2

Ecological Conceptual Site Model – SLERA ................................................. 3.2.3 3-2

Chemical Data Collection and Review – SLERA ........................................... 3.2.4 3-2

Problem Formulation – BERA ....................................................................... 3.3 3-3

Ecological Conceptual Site Model − BERA ................................................... 3.3.1 3-3

Management Goals ....................................................................................... 3.3.2 3-4

Selection of Key Receptors ........................................................................... 3.3.3 3-5

Ecological Endpoints ..................................................................................... 3.3.4 3-9

Chemical Data Collection and Review – BERA ............................................. 3.3.5 3-13

Selection of COPECS ................................................................................... 3.3.6 3-16

Level of Effort ................................................................................................ 3.3.7 3-27

Chapter 4. Analysis Phase

Introduction ................................................................................................... 4.1 4-1

Characterization of Exposure – SLERA ........................................................ 4.2 4-1

Characterization of Exposure – BERA .......................................................... 4.3 4-2

Exposure Setting ........................................................................................... 4.3.1 4-2

Exposure Analysis ......................................................................................... 4.3.2 4-3

Exposure Profiles .......................................................................................... 4.3.3 4-13

Ecological Effects Characterization – SLERA ............................................... 4.4 4

-25

Published Literature, Available Criteria and Information ................................ 4.4.1 4-25

Characterization of Ecological Effects – BERA ............................................. 4.5 4-26

Preliminary Toxicity Evaluation ..................................................................... 4.5.1 4-26

Bioassessment Tools and Techniques .......................................................... 4.5.2 4-27

Objectives ..................................................................................................... 4.5.3 4-27

Sources of Literature Benchmark Values ...................................................... 4.5.4 4-27

Selection of Literature Benchmark Values..................................................... 4.5.5 4-28

Development of Toxicity Reference Values ................................................... 4.5.6 4-30

Additional Considerations in Developing TRVs ............................................. 4.5.7 4-33

Special Chemicals ......................................................................................... 4.5.8 4-35

Chapter 5. Risk Characterization

Risk Characterization – SLERA..................................................................... 5.1 5-1

Refinement of the SLERA ............................................................................. 5.2 5-2

Risk Characterization – BERA ...................................................................... 5.3 5-2

Risk Estimation ............................................................................................. 5.3.1 5-2

EM 200-1-4

31 Dec 10

iii

Paragraph Page

Terrestrial Ecosystem Methodologies ........................................................... 5.3.2 5-3

HQ Method .................................................................................................... 5.3.3 5-3

Probabilistic Methodologies ........................................................................... 5.3.4 5-5

Aquatic Ecosystem Methods ......................................................................... 5.3.5. 5-6

Weight of Evidence ....................................................................................... 5.3.6 5-8

Risk Description ............................................................................................ 5.4 5-8

Factors Influencing Ecological Significance................................................... 5.4.1 5-8

Interpreting Site-Wide Ecological Significance .............................................. 5.4.2 5-8

Chapter 6. Uncertainty in Ecological Risk Assessments

Introduction ................................................................................................... 6.1 6-1

Characterization of Uncertainty – SLERA...................................................... 6.2 6-1

Characterization of Uncertainty – BERA ....................................................... 6.3 6-2

Objectives ..................................................................................................... 6.3.1 6-2

Sources of Uncertainty in a Risk Assessment ............................................... 6.3.2 6-2

Level of Effort ................................................................................................ 6.3.3. 6-7

Chapter 7. Alternative Evaluation

Introduction ................................................................................................... 7.1 7-1

Comparative Risk Assessment of Remedial Alternatives .............................. 7.2 7-1

Evaluation of Residual Ecological Threats .................................................... 7.2.1 7-1

Evaluation of Short-Term Ecological Threats ................................................ 7.2.2 7-2

Development of PRGs................................................................................... 7.3 7-3

Development of Remediation Levels ............................................................. 7.4 7-3

Chapter 8. Risk Management

Introduction ................................................................................................... 8.1 8-1

Determining Requirements for Action ............................................................ 8.2 8-6

Preliminary Assessment/Site Inspection ....................................................... 8.2.1 8-6

Remedial Investigation .................................................................................. 8.2.2 8-16

Feasibility Study/Remedial Design/Remedial Action ..................................... 8.2.3 8-20

Non-Risk Issues or Criteria as Determining Factors for Actions .................... 8.2.4 8-27

Design Considerations .................................................................................. 8.3 8-30

otential Risk Mitigation Measures ............................................................... 8.3.1 8-30

Risk Management; Degree Protectiveness ................................................... 8.3.2 8-32

Appendix A – References.....................................................................................................A-1

Glossary....................................................................................................................Glossary-1

EM 200-1-4

31 Dec 10

THIS PAGE INTENTIONALLY LEFT BLANK

EM 200-1-4

31 Dec 10

1-1

CHAPTER 1

Introduction

1.1. Purpose and Scope. This handbook provides technical guidance to USACE risk

assessors and risk assessment support personnel for planning, evaluating, and

conducting ecological risk assessments (ERAs) in a phased Hazardous, Toxic, and

Radioactive

1.1.1. Risk characterization is a similar process for both human health and ecological

risk assessments. The fundamental paradigm for human health risk characterization has

four phases: (1) hazard identification, (2) dose-response assessment, (3) exposure

assessment, and (4) risk characterization. Similarly, the fundamental framework for

ecological risk characterization includes four analogous phases: (1) problem formulation,

(2) exposure characterization, (3) ecological effects characterization, and (4) risk

characterization.

Waste (HTRW) response action and Military Munitions Response Program

(MMRP) projects with munitions constituents (MC). This handbook, a compendium to the

Risk Assessment Handbook: Volume I - Human Health Evaluation (EM 200-1-4, USACE

1999), encourages the use of "good science" within the framework of existing U.S.

Environmental Protection Agency ERA guidance.

1.1.2. This handbook encourages the concurrent assessment of human and

ecological risks so that data collection activities are coordinated and risk managers are

provided risk characterization results in a timely manner. Risk characterization results for

human and ecological receptors should be reasonable and communicated to the risk

managers in a clear and unbiased manner to facilitate the making of balanced and

informed risk management decisions.

1.1.3. For the purpose and intended use of this risk assessment handbook, the focus

is on the Defense Environmental Restoration Program (DERP) and Base Realignment and

Closure Program (BRAC) cleanup programs to address CERCLA- and RCRA-related

issues. EM 200-1-4 (USACE 1999), Risk Assessment Handbook, Volume I: Human

Health Evaluation, contains a complete discussion of the USACE HTRW program, which

will not be repeated here. The reader is referred to that document for details.

1.2. Applicability. This manual applies to all HQUSACE elements and USACE

commands responsible for HTRW and MMRP projects with MC.

1.3. Distribution Statement. Approved for public release, distribution is unlimited.

1.4. References. References are listed in Appendix A.

1.5. Introduction.

1.5.1. When this engineer manual was first published in June 1996, little detailed

guidance existed for performance of ecological risk assessments. The U.S.

Note that radioactive hazards, radioactive wastes, radiation generating devices and radioactively

contaminated materials are not addressed in this handbook.

EM 200-1-4

31 Dec 10

1-2

Environmental Protection Agency (USEPA) was in the process of drafting their guidance

(USEPA 1997a), as was the Army (Wentsel, et al. 1994). As time has passed,

USEPA’s guidance, Ecological Risk Assessment Guidance for Superfund (ERAGS):

Process for Designing and Conducting Ecological Risk Assessments (USEPA 1997a),

has become the industry standard, outlining the accepted ERA process. ERAGS is

based on the ERA paradigm first put forth in the document Framework for Ecological

Risk Assessment (USEPA, 1992a), which was expanded upon and replaced by the

Guidelines for Ecological Risk Assessment (USEPA 1998a).

1.5.2. The tiered approach originally put forth in this EM has fiscal and scientific

merit; however, it does not strictly follow the ERAGS process. This has caused district

risk assessors to avoid its use within the programs it was designed to augment

Installation Restoration Program (IRP), Base Realignment and Closure, and Formerly

Used Defense Sites (FUDS). Conversely, we have received positive feedback from

both within and outside the U.S. Army Corps of Engineers, citing the usefulness of

information contained herein. Therefore, this revision will discontinue the four-tiered

process and will not specify procedures conflicting with the ERAGS framework. It will,

however, contain all of the concepts and assessment techniques from the previous

issue, in a format that will facilitate application to ERAs conducted in accordance with

ERAGS.

1.6. Intended Audience and Use. This document was prepared for use by USACE

personnel responsible for scoping, preparing, directing, and reviewing ERAs at HTRW

response action sites. The engineering manual entitled, Technical Project Planning (TPP)

Process (USACE 1998) should be reviewed, particularly for understanding the process

described in Chapter 2 of this handbook on how to determine data quality objectives

(DQOs) to support an ERA.

1.6.1. The data collection, assessment, characterization of risk and uncertainty, and

the risk management decision-making aspects presented in this handbook are intended to

satisfy the Comprehensive Environmental Response, Compensation, and Liability Act

(CERCLA) and Resource Conservation and Recovery Act (RCRA) regulatory

requirements. The assessment of ecological risks under these two functionally equivalent

programs is essentially the same.

1.6.2. Response actions implemented in accordance with CERCLA or RCRA are not

legally subject to the National Environmental Protection Act (NEPA) and do not require

separate NEPA analyses. Therefore, this EM does not address NEPA directly. However

persons responsible for NEPA evaluations (i.e., environmental assessments,

environmental impact statements) may find many of the procedures in this handbook

useful in fulfilling NEPA requirements.

1.6.3. The concepts and assessment techniques presented in this handbook can be

used to optimize data quality design across regulatory program requirements (if

applicable) and justify or demonstrate that certain units or sites could be combined and

assessed as a single entity according to the concept of establishing a Corrective Action

Management Unit (CAMU) or temporary units (TU). If both regulatory programs are

EM 200-1-4

31 Dec 10

1-3

applicable at a site or unit, the ecological assessment components should be closely

coordinated to avoid duplication of effort. Where possible, the technical and risk

management approaches should be incorporated as specific language in agreements with

USEPA or states.

1.7. USACE Role in the HTRW Program. In the execution of USACE environmental

missions, the HTRW program is organized and staffed to respond to assignments for the

following five national environmental cleanup programs: (1.) USEPA Superfund Program

(a.k.a., CERCLA); (2) Defense Environmental Restoration Program, consisting of IRP,

FUDS, and Department of Defense and State Memorandum of Agreement/Cooperative

Agreement Program (DSMOA/CA); (3) BRAC; (4) Environmental Compliance

Assessment System (ECAS) (USACE 1992); and (5) HTRW environmental restoration

support for Civil Works projects and other federal agencies (Department of Defense (DOD)

and non-DOD).

1.8. Overview of the HTRW Response Process. HTRW response actions involve all

phases of a site investigation, design, remediation, and site closeout. The HTRW

response process is generally comprised of six executable phases or steps, once the

HTRW response site has been identified. They are: (1) Preliminary Assessment (PA);

(2) Site Inspection (SI), including the Screening-Level ERA (SLERA); (3) Remedial

Investigation (RI), including the Baseline ERA (BERA); (4) Feasibility Study (FS); (5)

Remedial Design/Remedial Action (RD/RA); and (6) Site Closeout.

1.9. Role of Ecological Risk Assessments in the HTRW Process. Performing an ERA is

an iterative process. Risk assessment information is continuously being collected during

the HTRW site response process, leading to the characterization of risks and uncertainties

qualitatively and/or quantitatively. Risk assessment information is used in various stages

of the HTRW site decision process as described below:

1.9.1. PA/SI or Other Preliminary Site Investigation Activities. In this phase of the

site process, risk assessment information is used to determine whether a site may be

eliminated from further concern, to identify emergency situations which may require

immediate response actions/interim corrective measures, to assess whether further site

investigations are required, to develop a data collection strategy, and to set site priorities,

e.g., to rank sites.

1.9.1.1. The SLERA (Steps 1 and 2 of ERAGS) developed during this phase should

be conducted using conservative scenarios as guided by the preliminary ecological

conceptual site model (ECSM), to ensure that any closeout decision at this stage is

protective of the environment. The SLERA is not to be confused with the Preliminary

Natural Resource Survey, which is a simple screening study, conducted by natural

resource trustees in conjunction with a natural resource damage assessment (NRDA). If

release of hazardous substances appears to have resulted in natural resource injury

(NRI), then Section 122(j) of CERCLA (as amended) requires federal natural resource

trustees to be notified. Section 122(j)(1) encourages federal natural resource trustees to

participate in response and remedy negotiations, so that data collected for an ERA can be

used by the trustees in carrying out their responsibilities.

EM 200-1-4

31 Dec 10

1-4

1.9.1.2. The Army has published guidance for coordination with the natural resource

trustees during CERCLA investigations. For FUDS, see FUDS Program Guidance to

Implement Army Interim Policy for Integrating Natural Resource Injury Responsibilities and

Environmental Response Activities (USACE 2003b). For active installations, see United

States Army Interim Natural Resource Injury Policy Guidance (USA 2005).

1.9.2. RI or Other Additional Site Investigation Activities. Data collected in this phase

should comprise those media and pathways identified in the SLERA, including background

data. If the data are useable and appropriate for the potential exposure pathways that are

considered to be complete, a BERA can be developed. The BERA can help identify the

potential for unacceptable ecological risks at the site.

1.9.2.1. Data Collection for Potential Ecological Risks. For assessing the potential

for ecological risks, data should be collected in the boundary or study area of ecological

concern and may need to be collected in reference areas as well. The study area may

necessitate combining solid waste management units (SWMUs) or operable units (OUs)

or developing a multi-site ERA if such combination is consistent with the ECSM for

assessing contamination and remediation options. Combined OUs or SWMUs should be

discussed with the regulators and identified in the agreements with agencies, the work

plan, or other decision documents.

1.9.2.2. The Army biological technical assistance group (BTAG) has authored a

position paper on how and when to combine sites for assessment of ecological risks.

See Integrating Multi-Site Ecological Risk Assessments for Wide-Ranging Receptors

(USA BTAG 2006).

1.9.3. FS, RD/RA or Other Remedial Design and Implementation Activities. The

BERA completed in the RI serves to identify the need for response actions and the relative

degree of response required. The potential environmental impacts posed during

remediation (short-term and long-term) and the residual risks after remediation are

evaluated during remedy selection.

1.9.4. Use of Ecological Risk Assessment in Special Studies. The following are

examples of ERAs used in special studies:

a. Applicable or Relevant and Appropriate Requirement (ARAR) Waiver — If a site-

specific remedial action objective (RAO), developed from the BERA is as protective as a

particular ARAR, an ARAR waiver request may be submitted under CERCLA Section

121(d)(2). The same process may be used to waive state ARARs.

b. Emergency Response — The effectiveness of a proposed removal action,

particularly for a time critical removal action, can be evaluated by the ERA in terms of the

ability of the action to reduce exposure or risks.

c. Biological Assessment of Endangered Species — The Endangered Species Act

(ESA) requires the preparation of a biological assessment if federally listed endangered or

threatened species or their habitat could be impacted by the contaminants or clean-up

actions (e.g., incinerator emissions) at hazardous waste sites. The ERA for the

EM 200-1-4

31 Dec 10

1-5

endangered or threatened species, and optional assessment of the Category 2 and rare

species, may satisfy the draft and final biological assessment requirements (Section 7

consultation) of U.S. Fish and Wildlife Service (USFWS) or other trustee agencies.

1.10. Ecological Risk Assessments as Decision Criteria in the HTRW Program.

1.10.1. The role of a risk assessment in the site decision-making process at

CERCLA and RCRA sites has been well defined by USEPA either through rule making or

program directive/guidance. Therefore, risk assessments have been used as decision

criteria in the USACE's HTRW program involving CERCLA and RCRA sites. For BRAC,

FUDS, or other HTRW work, which may not be on the national priorities list (NPL), risk

assessments should be similarly applied. Activities at these sites require the evaluation of

potential human health and environmental risks in order to return the property to conditions

appropriate for the current and planned future land uses. Therefore, a site-specific BERA

is an important decision tool to USACE customers. If cleanup is needed, the extent or

level of cleanup required will be based on results of the BRA (including the ERA and the

human health risk assessment(HHRA)), in addition to ARARs or other non-risk factors.

Therefore, risk assessment is used as a decision tool at all HTRW response action sites.

1.10.2. DOD and other federal agencies recognize the need for early input from all

stakeholders (broadly defined as the regulators, concerned citizens, environmental

groups, and other appropriate public and private interested parties) in order to facilitate risk

management decision-making. Establishing an early dialogue with stakeholders is

particularly important for ERAs to develop assessment strategies and preliminary RAOs.

1.11. Concept of Risk Assessment and Good Science.

1.11.1. Risk assessment can be qualitative and/or quantitative. It includes an

integration of hazard (chemical or non-chemical), exposure (scenario and pathways),

exposure-response (relationship between the magnitude of exposure and the resulting

ecological effects) and characterization of the risks and uncertainties. The risk assess-

ment process relies on strong fundamental scientific principles and representative data.

Despite this effort, there will be unavoidable data gaps and uncertainties where scientific

and professional judgment is needed to predict or infer certain outcomes under certain

scientific principles (Federal Focus Inc. 1994). The application of such judgment requires

that the risk assessor provide the rationale or basis for the judgment. This view is reflected

by the Policy for Risk Characterization (USEPA 1995a) and the National Academy of

Science (NAS) Science and Judgment in Risk Assessment (1993). Both USEPA and

NAS recognize the inherent uncertainties in the risk assessment methodologies, and the

need for making risk assessments more transparent, clear, consistent, and reasonable.

1.11.2. The fundamental principles of "good science" entail the thorough

understanding of: (1) site chemical data; (2) physical, chemical, and ecotoxicity

information associated with site chemicals; (3) fate and transport modeling; (4)

bioavailability and extent of uptake or bioconcentration; (5) the exposure-effects

relationship of site chemicals and underlying uncertainties/conservatism; (6) uncertainties

and limitations of the derived risk estimate; (7) the correct interpretation of previously

collected data, considering confounding factors, and making objective inferences or test

EM 200-1-4

31 Dec 10

1-6

hypotheses; and (8) unbiased presentation of findings and limitations or uncertainties

associated with the findings.

EM 200-1-4

31 Dec 10

2-1

CHAPTER 2

Planning and Scoping an Ecological Risk Assessment

2.1. Introduction. This chapter introduces the conceptual and technical objectives of an

ERA. The foundation for the present ERA approach is contained in ERAGS (USEPA

1997a). ERAGS prescribes an 8-step process, the first two steps constituting the

SLERA, and steps 3 through 8 making up the BERA.

2.1.1. The ERA is one component of overall site investigation and remedial

activities. It should be developed with recognition of how it is supported by preceding and

concurrent components of site activities, such as sampling and analysis and the HHRA

effort, and how it supports and shapes the subsequent components, such as remedial

design.

2.1.2. According to ERAGS, an ERA is defined as a process that evaluates the

likelihood that adverse ecological effects are occurring or may occur as a result of

exposure to one or more stressors. A stressor, as defined by USEPA, is any physical,

chemical, or biological entity that can induce an adverse ecological response. In the

Superfund program, an ERA entails the qualitative and/or quantitative appraisal of the

actual or potential impacts of a hazardous waste site on plants and animals other than

humans or domesticated species. Substances designated as hazardous under CERCLA

(see 40 Code of Federal Regulations (CFR) 302.4) are the stressors of concern

2.1.3. The approach to ERAs as presented in ERAGS is conceptually similar to the

approach used for human health, but is distinctive in its emphasis in three areas. First,

the ERA should consider effects beyond those to individuals of a single species and

should examine a population, community, or ecosystem (an exception is evaluation of

threatened or endangered species, where protection of the individual is mandated).

Second, no single set of ecological values to be protected can generally be applied to all

sites. Rather, these values are selected from a number of possibilities based on both

scientific and policy considerations. Finally, in addition to chemical induced toxic

stresses, ERAs may consider nonchemical induced stresses (e.g., loss of habitat)

qualitatively.

. These

definitions recognize that a risk does not exist unless: (1) the stressor has an inherent

ability to cause adverse effects, and (2) it co-occurs with or contacts an ecological

component long enough and at sufficient intensity to elicit the identified adverse effect(s).

2.1.4. The fundamental framework for ecological risk characterization includes four

phases: (1) problem formulation, (2) exposure characterization, (3) ecological effects

characterization, and (4) risk characterization. Problem formulation is a planning and

scoping process that establishes the goals, breadth, and focus of the ERA. Its end

product is an ECSM that identifies the environmental values to be protected (i.e.,

In very limited circumstances, the USACE may respond to substances that are not CERCLA

hazardous when they present an imminent and substantial endangerment. The appropriate legal

office should be consulted priot ro doing so.

EM 200-1-4

31 Dec 10

2-2

assessment endpoints), the data needed, and the analyses to be used. The second

phase develops profiles of environmental exposure and the third phase evaluates

ecological effects of the contaminants of potential ecological concern (COPECs) on the

receptors of concern. The exposure profile characterizes the ecosystem, in which the

COPECs may occur, as well as the biota that may be exposed. The exposure profile also

describes the magnitude and spatial and temporal patterns of exposure. The ecological

effects profile summarizes data (or in some cases, bioassessment results) on the effects

of the COPECs, on the receptors of concern, and relates them to the assessment and

measurement endpoints (see Section 3.3.3). Risk characterization integrates the

exposure and effects profiles. The potential for risks can be estimated using a variety of

techniques including comparing individual exposure and effects values, comparing the

distribution of exposure and effects, or using simulation models, and can be expressed as

a qualitative or quantitative estimate, depending on the available data.

2.2. Purpose of the Ecological Risk Assessment.

2.2.1. The main purpose of the ERA is to provide the necessary information to

assist risk managers in making informed site decisions. The ERA should provide an

objective, technical evaluation of the potential ecological impacts posed by a site, with the

risk characterization clearly presented and separate from any risk management

considerations. Although risk assessment and risk management are separate activities,

the risk assessor and risk manager need to work together at various stages throughout

the process to define decision data needs.

2.2.2. The risk assessor does not make decisions on the acceptability of any risk

level for protecting the environment or selecting procedures for reducing risk. The ERA is

used by the risk manager, in conjunction with regulatory and policy considerations, to

determine the appropriate response actions at the site. In the ERA, the risk assessor

needs to present scientific information in a clear, concise, and unbiased manner without

considering how the scientific analysis might influence the regulatory or site-specific

decision. The risk assessor is charged with:

a. Generating a credible, objective, realistic, and scientifically balanced analysis;

b. Presenting information on the problem, exposure, effects, and risks; and

c. Explaining confidence in the assessment by clearly delineating strengths,

uncertainties, and assumptions, along with impacts of these factors (USEPA 1995a).

2.3. Objectives of the Ecological Risk Assessment. The specific objectives of the ERA

are: (1) to identify and characterize the current and potential future threats to the

environment from a hazardous substance release; and (2) to establish remedial action

objectives that will protect those ecological receptors potentially at risk, if appropriate.

The ERA provides important risk management input at various project phases, identifying

ecological species or resources to be protected, as well as limitations and uncertainty.

2.4. Planning Considerations. Planning and problem identification are critical to the

success of the ERA and its usefulness with respect to remediation planning. To ensure

EM 200-1-4

31 Dec 10

2-3

that the scope of the ERA is sufficient for making appropriate risk management decisions,

the risk assessor must always be mindful of the question, "Do the data and ERA

approach support risk management decision-making?" The technical requirements of the

ERA should be considered early in the HTRW process to ensure that appropriate

information is gathered. It is important that the ecological risk assessor be involved in the

early planning stages of field investigations, including ECSM development, identification

of site media to be sampled, sampling plan design, data validation, compilation, and

interpretation. This will help ensure that the best possible and most relevant data are

available for use in the ERA.

2.4.1. Process Discussions. Throughout the planning process, the risk assessor

should strive to point out potential setbacks, problems, or difficulties that may be

encountered in a "real world" situation (see ERAGS, Step 5). Biological sampling

programs often entail scheduling constraints. For example, surveys for endangered

species (e.g., an orchid) should be conducted in the appropriate season (e.g., June, not

December). When special circumstances (e.g., lack of data, extremely complex

situations, resource limitations, statutory deadlines) preclude a full assessment, such

circumstances should be explained and their impact on the risk assessment discussed.

The risk assessor should also explain the minimum data quality considered to be

acceptable, how non-detects will be treated, and how medium-specific data will be

evaluated or compiled to derive or model the exposure point concentration in the risk

assessment.

2.4.2. Coordinating ERA and HHRA Planning. Planning for an ERA should be

conducted concurrently with that for a HHRA in that these two efforts often have similar

data needs, especially in the initial contamination characterization stages. Data needs for

the ERA, however, eventually focus on protection of ecosystem components, while the

HHRA focuses on protection of a single species, humans. The ERA format and process

is designed to be flexible. This allows for coordination with the HHRA in the chemical

sampling program, determination of the nature and extent of contamination,

characterization of site risk, and the overall site management decision process.

2.4.2.1. Coordinated planning efforts for the ERA and HHRA, particularly where

there is to be an expedited cleanup, should include consideration of the following:

a. Overlaps in information needs with regard to human and ecological food chain

issues;

b. Benefits of the cleanup and the effectiveness of presumptive remedies;

c. Ecological impacts from removal or remedial activities designed to protect human

health;

d. Identification of hot spots that may impact both human health and ecological

receptors;

EM 200-1-4

31 Dec 10

2-4

e. Identification of the key assumptions and criteria common to the human health

and eco risk assessments that may drive cleanup decisions and focus the decision

making process;

f. Early actions which may be taken at sites that could quickly and at a relative lower

cost reduce both ecological and human health risk;

g. Identification of areas of greatest concern that may be addressed as discrete

tasks, thereby allowing priority to be given to those actions (removal/remedial) that

achieve the greatest protection of the environment and human health for the capital

(dollars) spent; and

h. Activities common to both the ecological and human health risk efforts that

support DoD responsibilities as a Natural Resource Trustee or help coordinate between

multiple Natural Resource Trustees where jurisdictions or responsibilities overlap.

2.4.3. Planning for an ERA. The ERA should be developed, to some extent, with its

end uses in mind. Early interaction with risk managers and remedial designers is needed

to obtain information on the risk management options likely to be considered if remedial

action is required. This is not to infer that the ERA should be tailored to specific remedial

options, for that would compromise the objective nature of the assessment. However, if

the risk manager or remedial designer needs to know certain factors (for example, how

thick must the cap be to prevent on-site burrowing animals from being at risk), the risk

assessor should provide the basis that will allow him or her to answer this question.

2.4.3.1. Before initiating the ERA, project planning is generally conducted to help

set priorities and establish budget constraints. Early project planning establishes the

focus and complexity of the ERA. Planning includes a review of the available background

material and discussions to define the scope and critical aspects of the ERA. Spatial

boundaries such as the size of the site, extent of contamination, potential threats to onsite

and nearby ecosystems, and important ecosystem components (e.g., fisheries) greatly

determine the potential scope and design of the ERA. Any remediation or restoration

plans for the site should be considered in the planning stage. Data deficiencies should

also be recognized at this stage to the extent possible. Recognizing these planning

elements and articulating specific objectives early in the planning stage will drive the

design and focus of the subsequent ERA efforts.

2.4.3.2. In the risk planning process, when NRI are possible, it is important for the

project delivery team (PDT) to coordinate with natural resource trustees (e.g., DoD, the

State, National Oceanic and Atmospheric Administration (NOAA), USFWS, U.S. Forest

Service (USFS), and the Bureau of Land Management (BLM) at the earliest possible

stage. In this way, the trustee can be assured that potential environmental concerns are

addressed, and conclusion of action may be expedited. Coordination with natural

resource trustee agencies such as NOAA, provides for the exchange of ideas and issues

to ensure the technical adequacy of the RI/FS, and to ensure the protectiveness of the

selected remedy for trust resources. Coordination also allows DoD access to the

trustees' specific skills, information, and experience in ERAs. This interaction may occur

EM 200-1-4

31 Dec 10

2-5

through a variety of informal and formal forums, including but not limited to: preliminary

scoping and drafting of work plans, review of final work plans and subsequent data,

technical review committees, meetings, and public information meetings. The Army has

published guidance for coordination with the natural resource trustees, which should be

consulted whenever the potential for NRI is discovered (USACE 2003b and USA 2005).

2.4.4. HTRW Technical Project Planning Process. USACE recognizes the need for

cost-effective and efficient site investigation/response actions. To this end, EM 200-1-2

(USACE 1998), Technical Project Planning (TPP) Process, provides guidance on data

collection programs and defines DQOs for HTRW sites. The HTRW TPP process is a

four-phased process that begins with the development of a site strategy and ends with

the selection of data collection options. Using this process, the risk assessor will be able

to define minimum information requirements for risk evaluations in support of site

decisions. The project planning process should produce an outline for a site-specific ERA

that is credible, objective, realistic, and scientifically-balanced.

2.4.4.1. In identifying data needs for the ERA, the risk assessor must fully

understand the customer goals, the regulatory programs driving the HTRW project

execution, the study elements necessary for the relevant project phase, and the type of

ERA needed, based on the study elements. The concept of TPP is fully explained in EM

200-1-2, which emphasizes the need for the data users (in this case, the risk assessor) to

identify minimum data requirements for the tasks to be performed. The concept of

"minimum requirements" for the ERA is important in that it identifies certain minimum

requirements for data collection activities preceding the ERA to ensure that critical data

gaps are addressed.

2.4.4.2. DQOs define the project's data needs, data use, number of samples

required, the associated quality requirements (e.g., detection limits, blanks, split and

duplicate samples, etc.), and level of confidence or acceptable data uncertainty for the

requisite data. DQOs are generated at the final phase of the TPP process after the

customer has selected the preferred data collection program. The process includes

evaluation of previously collected data, and assessment of need for additional data to

support the study elements for the current or subsequent phases of the project. This

coordinated project planning effort is designed to satisfy the customer goals, applicable

regulatory requirements, and minimum technical data requirements for performing a site-

specific ERA.

2.4.4.3. Throughout the process, USACE HTRW personnel of various disciplines

and responsibilities work closely together to identify data needs, develop data collection

strategy, and propose data collection options. The TPP process implements the

USEPA's DQO process (USEPA 2000b), which is an iterative process applicable to all

phases of the project life cycle. The DQO development process is considered to be a

total quality management tool (USEPA 1989c). Incorporating the TPP process is key to

ensuring successful planning and performance of the ERA.

EM 200-1-4

31 Dec 10

2-6

2.4.5. Site-Specific Considerations. Site-specific data should be collected and

used, wherever practical, to determine whether or not a site release presents

unacceptable risks and to develop quantitative cleanup levels that are protective. Site-

specific data can include such things as plant and animal tissue residue data,

bioavailability factors, and population- or community-level effects studies.

The Army BTAG has prepared a position paper specifically for planning an ERA

(USA BTAG 2002a). This document applies the TPP process to ERA planning and will

assist the risk assessor in establishing the data required for the ERA.

2.5. Establishing the Level of Effort. The preliminary level of effort and nature of the ERA

are directly related to the study elements that need to be addressed. Boundaries need to

be set early in the scoping process, since the amount of information that could be

incorporated into an ERA is potentially limitless. Although often predetermined to a large

extent by schedule and budget constraints, these boundaries should be tied to the

objectives of the assessment and the site-specific nature of the potential risk.

2.5.1. Approach and Level of Effort. The approach and level of effort for an ERA

are based on DQOs developed during the TPP process. DQOs address data quality and

quantity requirements, as well as data use. DQOs are integral to the design and conduct

of cost-effective and efficient ERAs under current and future land-use scenarios. While

the overall framework for the conduct of the risk assessment should remain consistent

with the ERAGS paradigm, the risk assessor may apply a variety of approaches and

classification schemes in the conduct of the ERA. Two distinct approaches are generally

seen in ERAs: the criteria-based approach and the ecological effects-based approach,

which will be discussed in Chapter 3.

2.5.2 SLERA (Steps 1 and 2 of ERAGS). The SLERA constitutes the first two steps

in the ERAGS process and is intended to allow a rapid, inexpensive determination if the

site poses no or negligible risk. A SLERA may be performed for a PA/SI, or as the initial

step in the RI (if it was not done prior). The SLERA is generally a criteria- or chemical

concentration-based approach. Chemical criteria, such as state and federal ambient

water quality criteria (AWQC), sediment quality guidelines, or ecotoxicological risk-based

screening concentrations, similar to human health risk-based concentrations, are routinely

screened against in the SLERA. These chemical screening concentrations represent

conservative values that are designed to be protective of specific receptors or

ecosystems (aquatic, terrestrial, wetland), and should not be applied as cleanup levels at

a site. Screening concentrations, however, are not available for all chemicals and all

receptors.

2.5.2.1. In addition to conservative environmental criteria, exposure factors that are

used should also be conservative. ERAGS lists the following factors and provides

explanation relative to how these parameters are evaluated in the SLERA: area use

factor – 100%; bioavailability – 100%; life stage of receptor – most sensitive; body weight

and food ingestion rate – minimum body weight to maximum ingestion rate; dietary

composition – 100% of diet consists of the most contaminated dietary component. The

SLERA may help eliminate chemicals, pathways of exposure (e.g., soil ingestion),

EM 200-1-4

31 Dec 10

2-7

foraging guilds (e.g. small mammalian herbivores), and even entire sites (yet in practice

this is rare given the conservatively biased approach that is used). If the SLERA

indicates the potential for risk, project planning should occur to review the screening

results and define the scope and critical aspects of performing a BERA.

2.5.2.2. In certain circumstances, it may be worthwhile to refine some exposure

parameters to evaluate less conservative exposures. This decision would be based on

the likelihood that more realistic exposure parameters would bring the calculated

hazard quotients (HQs) below one. This step (Step 3A) would occur prior to beginning

problem formulation for the BERA (see Section 5-2). The decision to continue beyond

the SLERA does not indicate that risk is unacceptable or that risk reduction is necessary,

rather it indicates that a more focused evaluation and characterization of the potential for

risk and accompanying uncertainty is needed.

2.5.3. BERA (Steps 3 through 8 of ERAGS).

2.5.3.1. The ecological effects-based approach is more commonly applied in the

BERA. This approach is based on the detailed evaluation of site-specific conditions using

toxicity tests or actual biological measurements. This approach is commonly applied to

aquatic ecosystems, where standardized American Society for Testing and Materials

(ASTM) test methods may be used. This causal evidence approach allows for the

identification of biological or ecological impacts without specific accountability for the

chemical causative factors and is not constrained by the limitations of chemical analytical

techniques. Chemical concentration data are used primarily to establish general

accordance. As proof of causality is not a requirement for the ERA, the evaluation of

causal evidence is used to augment the risk assessment. Criteria for evaluating causal

associations have been suggested by Hill (1965) and are provided in USEPA's Guidelines

(1998a).

2.5.3.2. Unlike the SLERA, the focus of the BERA is to evaluate potential threats

using site-specific information wherever possible. Only those receptors, pathways and

COPECs remaining after the SLERA are evaluated, and exposure factors are adjusted to

more realistic levels. The BERA should provide an objective, technical evaluation of the

potential ecological impacts posed by a site. The process combines data from biotic and

abiotic media along with exposure and toxicity information to provide a determination of

environmental risk. The BERA should be clear about the approaches, assumptions,

limitations and uncertainties in the evaluation, to enable the risk assessor and manager to

interpret the results and conclusions appropriately. The BERA is used by the risk

manager, in conjunction with regulatory and policy considerations, to determine the

appropriate response actions at the site.

2.5.3.3. To evaluate the relationship between contamination and ecological effects,

the BERA requires evaluation of strategy objectives and data needs, based upon the

integration of three types of information:

a. Chemical: Chemical analyses of appropriate media to establish the presence,

concentrations, and variability of specific toxic compounds.

EM 200-1-4

31 Dec 10

2-8

b. Ecological: Ecological information to document potentially exposed ecosystems

and populations (or threatened and endangered individuals); to characterize the condition

of existing communities; and to observe whether any obvious adverse effects have

occurred or are occurring.

c. Ecotoxicological: Ecotoxicological information or testing to establish the link

between adverse ecological effects and known contamination.

d. Without these three types of data, other potential causes of the observed effects

on ecosystems unrelated to the presence of contamination, such as natural variability and

human-imposed habitat alterations, cannot be eliminated.

2.5.4. Risk-Based Analysis of Remedial Alternatives. Various types of ERAs may

be applied to conduct a screening evaluation of remedial alternatives or a more detailed

analysis of a selected alternative. Generally, the procedure used for the SLERA will be

sufficient in providing the risk inputs for selection of potential remedial alternatives or

corrective measures (including the no-further action alternative) or the need for

procedural changes or engineering controls to minimize short-term risks or residual risks.

The two prime objectives of this type of ERA are: 1) the development of remediation

goals to be applied to site cleanup, and 2) development of comparative risk assessments

between different remedial options. The first type is sometimes performed as a compo-

nent of the RI, but is distinguished in this section because of its use in the development of

remedial options. The second type of ERA is not as commonly performed, but it can be

useful in distinguishing between potential remedial options. When evaluating ecological

risks and the potential for response alternatives to achieve acceptable levels of

protection, the risk manager should characterize risk in terms of (1) magnitude, (2)

severity, (3) distribution, and (4) the potential for recovery of the affected receptors.

2.6. Scoping an Ecological Risk Assessment. For scoping purposes, it should be noted

that most ERAs are highly site-specific and often require unique investigative plans and

actions. The approaches and contents of the anticipated ERA should be explained or

discussed in the project planning stage in unambiguous terms. An iterative, tiered

approach to the risk assessment, beginning with the SLERA, is used to determine if a

more comprehensive assessment is necessary. The nature of the risk assessment

depends on available information, the regulatory application of the risk information, and

the resources available to perform the ERA. Informed use of reliable scientific

information from many different sources is the central feature of the ERA process

(USEPA 1995a,c). Therefore, establishing a scope of work for an ERA requires familiarity

with the site features/habitats, the COPECs, and the receptors of concern.

2.6.1 Engineer Pamphlet (EP) 200-1-15 (USACE 2001), Standard Scopes of Work

for HTRW Risk Assessments, was written to provide the USACE risk assessor with the

minimum information necessary to begin scoping an ERA, either screening-level or

baseline. The Scopes of Work (SOWs) are provided in Word® format to allow the risk

assessor to tailor the SOW to site-specific conditions.

EM 200-1-4

31 Dec 10

2-9

2.6.2. As part of their ECO Update series, USEPA published Developing a Work

Scope for Ecological Assessments (USEPA 1992k). Written to support Superfund, this

document will help ensure that the ERA accomplishes its objectives within reasonable

budget and schedule limitations.

2.7. Guidance. Principal guidance documents for conducting or managing ERAs within

the USACE HTRW program are listed below. The risk assessor, as well as the project

manager (PM), should be familiar with the procedures put forth in these documents.

2.7.1. Department of Defense.

2.7.1.1. The DoD Tri-Services Ecological Risk Assessment Work Group has

produced ERA guidance generally following ERAGS protocols. The Tri-Service

Procedural Guidelines for Ecological Risk Assessments (Wentsel et al. 1996) preceded

ERAGS but advocated the same 8-step process, and provides information on more

than 100 environmental models and test methods.

2.7.1.2. The Tri-Service Remedial Project Manager’s Technical Handbook For

Ecological Risk Assessment (Simini et al. 2000) provides the PM with information to

ensure the ERA stays focused while being timely and cost-effective

2.7.2. U.S. Army. The Army BTAG has published several position papers

applicable to ERAs. Designed to clarify and enhance the ERAGS process, these papers

are written for either technical personnel, or the PM.

2.7.2.1. Technical Document for Ecological Risk Assessment: Planning for Data

Collection, (USA BTAG 2002a). As noted above, this document applies the TPP Process

to ERA planning.

2.7.2.2. Technical Document for Ecological Risk Assessment: Selection of

Assessment and Measurement Endpoints for Ecological Risk Assessments, (USA BTAG

2002b). The purpose of this document is to provide general recommendations in regard

to selecting appropriate assessment and measurement endpoints for ERAs at Army

installations.

2.7.2.3. Technical Document for Ecological Risk Assessment: Screening-Level

Ecological Risk Assessments for Army Sites, (USA BTAG 2005a). This document is

designed to assist the risk manager in understanding the SLERA and how it applies to the

overall site investigation process.

2.7.2.4. Technical Document for Ecological Risk Assessment: Process for

Developing Management Goals (USA BTAG 2005b). This document directs how to

establish management goals that lead to selection of assessment and measurement

endpoints for ERAs.

2.7.2.5. Technical Document for Ecological Risk Assessment: Integrating Multi-Site

Ecological Risk Assessments for Wide-Ranging Receptors (USA BTAG 2006). This

EM 200-1-4

31 Dec 10

2-10

document is a technical guide addressing criteria for identifying when and why a multi-site

ERA may be appropriate and how it may be conducted.

2.7.3. USEPA (Superfund).

2.7.3.1. Ecological Risk Assessment Guidance for Superfund (ERAGS): Process

for Designing and Conducting Ecological Risk Assessments, Interim Final (USEPA,

1997a). This document provides guidance to site managers and RPMs who are legally

responsible for the management of a site on how to design and conduct technically

defensible ecological risk assessments for the Superfund program. This document

supersedes USEPA's 1989 Risk Assessment Guidance for Superfund, Volume 2,

Environmental Evaluation Manual (USEPA 1989a) as guidance under Superfund.

However, the Environmental Evaluation Manual contains useful information on the

statutory and regulatory basis of ecological assessment, basic ecological concepts, and

other background information that is not repeated in ERAGS.

2.7.3.2. Ecological Assessment of Hazardous Waste Sites: A Field and

Laboratory Reference, Final (USEPA, 1989b). This document provides information for

field and laboratory procedures for designing, implementing, and interpreting ERAs at

hazardous waste sites.

2.7.3.3. The Role of Screening-Level Risk Assessment and Refining

Contaminants of Concern in Baseline Ecological Risk Assessments. ECO Update,

Intermittent Bulletin (USEPA 2001a). This supplemental guidance is intended to

provide further clarification and direction regarding SLERAs, as described in ERAGS. It

also provides an approach for incorporating additional components into the Problem

Formulation phase of more detailed (i.e., “baseline”) ecological risk assessments,

particularly in Step 3.2, which discusses refining COPECs.

2.7.3.4. Role of the Ecological Risk Assessment in the Baseline Risk Assessment.

(USEPA 1994a) Memorandum from Elliott Laws, Assistant Administrator. August 12.

USEPA Office of Solid Waste and Emergency Response (OSWER) Directive No. 9285.7-

17. This document emphasizes USEPA’s interest in protection of the environment as well

as human health.

2.7.4. USEPA (Agency-wide). Guidelines for Ecological Risk Assessments, Final

(USEPA, 1998a). This document provides a flexible process for organizing and

analyzing data, information, assumptions, and uncertainties to evaluate the likelihood of

adverse ecological effects. ERA provides a critical element for environmental decision-

making by giving risk managers an approach for considering available scientific

information along with the other factors (e.g. social, legal, political, or economic) in

selecting a course of action. These guidelines will help improve the quality of ERAs at

USEPA while increasing the consistency of assessments among the Agency’s program

offices and regions. The Guidelines expand upon and replace the USEPA’s

Framework for Ecological Risk Assessment (USEPA 1992a).

EM 200-1-4

31 Dec 10

2-11

2.7.5. USEPA (Regional). Some USEPA Regions have supplemented the national

USEPA risk assessment guidance with their own guidance, policies and procedures for

use in conducting an ERA. These guidance documents, in the form of memoranda,

directives, or stand-alone documents, address a wide range of issues. Regional

guidance should be consulted and evaluated for applicability for any work within the

regional boundaries. Access to EPA Regions can be obtained through:

http://www.epa.gov/epahome/locate2.htm

2.7.6. State. Some states have established guidance for performing hazardous

waste risk assessments within the state. Some have also promulgated cleanup levels

and/or procedures to supplement the National Oil and Hazardous Substances Pollution

Contingency Plan (NCP). It should be noted that CERCLA has clearly defined ARARs

within the process, and such state guidance, policy or procedures may or may not

qualify. Nonetheless, the risk assessor should evaluate the applicability of any state

guidance/requirements when performing HTRW work within the state. Access to state

hazardous waste programs can be obtained through:

http://www.epa.gov/epaoswer/osw/stateweb.htm

EM 200-1-4

31 Dec 10

2-12

THIS PAGE INTENTIONALLY LEFT BLANK

EM 200-1-4

31 Dec 10

3-1

CHAPTER 3

Problem Formulation

3.1. Introduction. Planning and problem identification are critical to the success of an

ERA and its usefulness with respect to decision-making. The interface among risk

assessors, risk managers, and stakeholders at the beginning of the process, and the

communication of risk at the end of the ERA, is critical to ensure that the results of the

assessment can be used to support a management decision (USEPA 1998a).

3.1.1. The characteristics of an ERA are determined by agreements reached by

risk assessors and risk managers during the planning process. These agreements

include: (1) clearly established and articulated management goals (see Section 3.3.3.),

(2) characterization of decisions to be made within the context of the management

goals, and (3) agreement on the scope, complexity, and focus of the ERA, including the

expected output and the technical and financial support available to complete it

(USEPA 1998a).

3.1.2. The TPP Process delineated in EM 200-1-2 (USACE 1998) was developed

to focus on data needs and to design quality data collection options. The TPP Process

also encourages early refinements of potential data collection options as a means of

identifying the most cost-effective options for selection. The Army BTAG has authored

a position paper that applies the TPP process to determining data needs for an ERA.

This paper, Technical Document for Ecological Risk Assessment: Planning for Data

Collection (USA BTAG 2002a), should be consulted prior to initiating problem

formulation for any ERA, either screening-level or baseline.

3.2. Problem Formulation – SLERA. Problem formulation for the SLERA begins with a

compilation of readily available information on the environmental setting and potential

contamination problem. USEPA suggests use of their environmental checklist (Appendix

B of USEPA 1997a) in conjunction with a site visit by a qualified ecologist/biologist to help

determine the level of effort needed to assess ecological risk at a particular site.

Knowledge of the environmental setting and potential contaminant migration pathways

allows for an early determination of the presence or absence of complete exposure routes

and the potential for significant ecological impacts. State and federal laws (e.g., Clean

Water Act (CWA), ESA) designate certain types of receptors (endangered species) and

environments (critical habitats, wetlands) that require special consideration during the risk

assessment process or protection at the remediation stage. Knowledge of pertinent state

and federal laws pertaining to natural resources and sensitive environments at the site is

a key element of the problem formulation step and the identification of endpoints.

Ecological information on potentially impacted environments and components can be

derived from an installation’s Integrated Natural Resource Management Plan, installation

natural resource personnel, state natural heritage reports, and federal agencies such as

the USFWS.

3.2.1. Reconnaissance. During the reconnaissance, a checklist of biological

species should be developed. From this list, receptors of special concern will be

EM 200-1-4

31 Dec 10

3-2

identified. Depending on the contaminant sources and potential transport pathways,

these receptors could include major elements of the given food chain from plants to

higher trophic levels such as insects, reptiles, birds, and mammals. Aquatic

ecosystems, for example, can include aquatic plants, bottom fauna (e.g., insects,

mollusks), amphibians, turtles, piscivorous snakes, fish, wading birds or ducks, and

predatory raptors.

3.2.2. Receptors. Receptors are the components of ecosystems that are or may

be adversely affected by a chemical or stressor. In the SLERA, species, species

groups, functional groups (e.g., producer, consumer, decomposer), food guilds (i.e.,

organisms with similar feeding habits), and critical habitats are the focus of receptor

selection. Receptors can be any part of an ecological system, including species,

populations, communities, and the ecosystem itself. Toxicity of chemicals to individual

receptors can have consequences at the population, community, and ecosystem level.

Population level effects may determine the nature of changes in community structure

and function, such as reduction in species diversity, simplification of food webs, and

shifts in competitive advantages among species sharing a limited resource. Ecosystem

functions may also be affected by chemicals, which can cause changes in productivity,

or disruption of key processes (alteration of litter degradation rate). Because it is

difficult to assess potential impacts to all receptors, a smaller group of receptors of

concern (key receptors) are used to assess potential harm to all components of the

system. In the ERA, specific organisms or groups (e.g., small herbivores) are usually

selected as key receptors, depending on whether the ERA is screening-level or

baseline. The reader is directed to the Army BTAG position paper, Technical

Document for Ecological Risk Assessment: A Guide to Screening Level Ecological Risk

Assessment (USA BTAG 2005a) for additional information.

3.2.3. Ecological Conceptual Site Model – SLERA. A preliminary ECSM should be

developed during the problem formulation. The ECSM is a simplified, schematic, diagram

of possible exposure pathways and the means by which contaminants are transported

from the primary contaminant source(s) to ecological receptors. The exposure

scenario(s) usually include consideration of sources, environmental transport, partitioning

of the contaminants amongst various environmental media, potential chemical/biological

transformation or speciation processes, and identification of potential routes of exposure

(e.g., ingestion) for the ecological receptors. Because this is a screening effort and

knowledge of site-specific ecological receptors may be lacking, the ECSM should be quite

simplified, incorporating general categories (e.g., terrestrial or aquatic biota) in place of

site-specific ecological receptors. Reference EM 1110-1-1200, Conceptual Site Models

for Ordnance and Explosives (OE) and Hazardous, Toxic, and Radioactive Waste

(HTRW) Projects (USACE 2003a)(currently under revision), for additional information

on conceptual site models.

3.2.4. Chemical Data Collection and Review – SLERA. Appropriate data must be

used for the SLERA to meet its objectives. Data available from PA/SI activities are

EM 200-1-4

31 Dec 10

3-3

usually limited in number but should be broad in scope of chemical analysis and in the

number/type of abiotic media sampled.

3.2.4.1. Sampling should have been conducted in areas of suspected contamination

and background areas to distinguish site contamination from background levels and to

provide information on the "worst case." If sampling was not conducted in areas of

suspected contamination, the SLERA will not provide an adequately cautious assessment

of potential risk. Similarly, if a broad chemical analysis was not performed, or if data are

not available for all abiotic media of potential concern, the screening SLERA will be

limited and cannot be used to eliminate the site from further consideration.

3.2.4.2. The following are examples of minimum requirements for data applied to a

SLERA:

a. Chemical-specific analyses of appropriate abiotic media of potential concern (soil,

sediments, surface water); and

b. Data of good quality according to the analytical methodology applied.

3.3. Problem Formulation – BERA. An initial step in problem formulation for the BERA

may be the development of working hypotheses. Hypothesis development is essential

when statistical comparisons are anticipated (e.g., comparisons of on-site to off-site biotic

populations).

3.3.1. Ecological Conceptual Site Model – BERA. The ECSM, which should have

been established in the PA/SI project phase, presents all potential exposure pathways

(sources and release mechanisms, transport media, exposure points, exposure routes

and receptors) and identifies those pathways which are complete (significant or

insignificant) and incomplete. The ECSM helps the project team focus the data

collection effort to evaluate significant pathways and address project requirements. At

this time, data concerning potential existence and locations of sensitive environments,

endangered species, or valued resources should already have been collected.

3.3.1.1. The ECSM establishes the complete exposure pathways that are to be

evaluated in the ERA and the relationship between the measurement and assessment

endpoints (see Section 3.3.3). The ECSM forms the basic decision tool for evaluating

the appropriateness and usefulness of the selected measurement endpoints in

evaluating the assessment endpoints (See 3.3.4.). The ECSM is also used as a tool

for identifying sources of uncertainty in the exposure characterization (exposure point

chemical concentrations).

3.3.1.2. Initial formulation of the ECSM in the SLERA is based upon existing

information and assumptions regarding chemical presence and migration, which now

should be verified and refined with data collected during the RI. The ECSM is refined in

greater detail throughout the characterization of exposure portion of the BERA (see

Chapter 4). The risk assessor and project team members should review site data and

information collected in earlier project efforts (i.e., PA/SI) to establish or refine the

EM 200-1-4

31 Dec 10

3-4

ECSM (based on more complete background information or non-chemical data) and

assess potential early/immediate response actions, as appropriate. All existing data

should be reviewed for quality, usability, and uncertainty before defining new data

acquisition requirements. The information should be able to assist the risk assessor in

developing a more definitive ECSM, or multiple ECSMs if there are multiple OUs,

SWMUs, areas of concern (AOCs), or CAMUs/TUs (if appropriate). This information

should include:

a. COPECs (information concerning the source characteristics, media

contamination, and background chemicals, including those of anthropogenic origin, is

needed to identify COPECs);

b. Potential target media (groundwater, surface water, soil/sediment, and air);

c. Media parameters and characteristics;

d. Potential receptors in the target media;

e. Major exposure routes or pathways of concern (e.g., direct contact resulting in

soil or sediment ingestion or dermal absorption of contaminants in the media,

consumption of food chain crops or prey species, surface water ingestion, and

inhalation of contaminants in ambient air);

f. Migration and transport potential of site chemicals from the source, including the

effect of existing institutional controls or interim corrective measures or removal actions

(e.g., groundwater capture well systems to prevent migration to surface water);

g. Exposure areas or units with common COPECs, which also pose common

exposure pathways and threats to ecological receptors;

h. Potential secondary, tertiary, and quaternary sources of contaminants, and their

release/transport mechanisms;

i. Level of contamination when compared to available ARARs or benchmark

values, and relevancy of sample location/matrix;

j. Removal actions or interim corrective measures taken; and

k. Data usability based on quality assurance characteristics, parameter analyzed,

validation results, and the way the data were compiled that may severely restrict their

use in the risk assessment.

3.3.2. Management Goals. Prior to conducting an ERA, social and political

considerations are used with site information to develop management goals. These

management goals are the cornerstone of subsequent phases of the risk assessment.

The problem formulation phase of the BERA uses management goals to develop

ecological endpoints.

3.3.2.1. Management goals are defined as a general statement about the desired

condition of ecological values of concern (USEPA, 1998a). These goals may vary from

EM 200-1-4

31 Dec 10

3-5

"no unreasonable effects on bird survival” to "minimize surface water impacts" to

"reestablish a tall grass prairie". Since ERA management goals often come from

interpretations of law by regulators or the desires of the property owner and/or the local

community, it is critical to involve all stakeholders when planning an ERA to ensure

management goals are appropriate for the site and ecosystem of concern.

3.3.2.2. The Army BTAG has authored a position paper on when and how to

establish management goals for an ERA. See Technical Document for Ecological Risk

Assessment: Process for Developing Management Goals (USA BTAG 2005b).

3.3.3. Selection of Key Receptors. Receptors are the components of ecosystems

that are or may be adversely affected by a chemical or other stressor. Endpoints are

characteristics of an ecological component that may be affected by an environmental

stressor (e.g., chemical contaminant). Because it is difficult to assess potential impacts

to all receptors for all endpoints, ecological assessment methods select particular types

of receptors (key receptors) and endpoints to represent potential harm to all

components of the system.

3.3.3.1. Objectives. Grouping of species, organisms, habitats, or ecosystem

components under the heading of key receptors helps focus the exposure

characterization portion of the BERA on species or components that are the most likely

to be affected and on those that, if affected, are most likely to produce greater effects in

the onsite ecosystem. The focus of receptor selection process is on species, groups of

species (e.g., birds, benthic invertebrates), or functional groups (feeding guilds), rather

than higher organizational levels such as communities or ecosystems. Chemical-

specific toxicological parameters are also generally limited to the more common

organisms or species in the onsite environment and prey organisms that are likely to be

used more heavily than others. Although grouping species together for the purposes of

exposure and risk quantitation (model analysis) results in some uncertainty, this

uncertainty might be offset by the use of conservative criteria to select key receptors

with the greatest sensitivity (highest trophic level receptor or chemically sensitive) or

greatest opportunity for exposure.

3.3.3.2. General Considerations. The selection of key receptors is in part a

subjective decision based on species presence, dominance, judged importance in the

food chain, and societal or scientific value. Key receptors and are not only species, but

may include habitat or areas of special legal protection. Location-specific ARARs,

identified as part of the RI effort, may concern locations of natural resources, sensitive

ecological receptors or species protected under a number of resource protection

statutes. Some of these statutes were developed several decades ago, and their

requirements are very specific. Environmental statutes such as the ESA, Migratory Bird

Treaty Act, Eagle Protection Act, and Wetlands Protection Act, are used in conjunction

with other criteria to help identify (but not mandate) important receptors and select

appropriate ecological endpoints (see Section 3.3.4). These laws may also be applied

to risk management decision-making during the FS to evaluate the need for and extent

EM 200-1-4

31 Dec 10

3-6

of remediation and the potential effects of various remedial alternatives, based on risk

characterization performed in the ERA.

3.3.3.2.1. Primary criteria for key receptor selection generally include consideration

of the following:

a. Likelihood of contacting chemicals,

b. Abundance in the study area.

c. A key component of ecosystem structure or function (e.g., importance in the

food web, ecological relevance),

d. Listing as rare, threatened, or endangered by a governmental organization; or

critical habitat for such,

e. Sensitivity to chemicals, and

f. Recreationally valued species (e.g., game animals).

3.3.3.2.2. Additional criteria used in key receptor selection include habitat

preference, food preference, and other behavioral characteristics, which can determine

population size and distribution in an area or significantly affect exposure potential. Key

receptors may include mobile game species with large home ranges; or smaller non-

migratory species; or organisms that are sedentary or have a more restricted

movement. For chemicals that bioaccumulate, the effects are usually most severe for

organisms at the top of the food chain (e.g., top predators) like bass in aquatic

ecosystems or raptors in terrestrial ecosystems.

3.3.3.3. Likelihood of Contacting Chemicals. Information from the site

reconnaissance, biota checklist (if available), and other available literature are used to

compile a candidate list from which preliminary key receptors are selected. General

field guides and publications on local and regional fauna, including environmental

impact statements, provide good preliminary information. Regional natural resource

agencies, such as state fish and wildlife departments, should be consulted for more

detailed information. Site maps should be reviewed for information on general

physiography, ecosystems, and habitat types.

3.3.3.3.1. Potential key receptors should be evaluated with respect to their

likelihood for directly or indirectly contacting areas affected by chemical input. Key

receptor selection analysis includes an evaluation of the receptor's relation to COPEC

exposure through both direct contaminant accumulation from the abiotic environment

and bioaccumulation through the food chain. Habitat destruction and loss or absence

of the receptor from impacted habitats are additional considerations in selecting key

receptors.

3.3.3.3.2. Where sites are large and numerous species are likely to be present, the

preliminary receptors may be reduced into categories (e.g., small birds, small

EM 200-1-4

31 Dec 10

3-7

mammals, wading birds, semiaquatic mammals) or into groups of species that are more

toxicologically sensitive (i.e., demonstrate adverse effects to lower environmental

concentrations of the COPECs). The list may also be reduced by grouping species into

taxonomically related groups and/or feeding guilds, such as hawks or eagles that are

often top predators in terrestrial food webs. From the reduced list, representative

species can be determined on the basis of observations indicating which species are

common onsite and potentially most sensitive to the COPECs.

3.3.3.4. Sensitivity to Chemicals. Species differ in the ways that they uptake,

accumulate, metabolize, distribute, and excrete contaminants. Susceptibility of an

organism also varies with the manner in which organisms are exposed to chemicals in

their environment. When possible, key receptors and endpoints are selected by

identifying those that are known to be susceptible to chemicals at the site based on

published literature. This process ensures that a conservative approach is taken to

evaluate receptors (at the individual/population, community, or ecosystem level) and

endpoints likely to be adversely affected in combination with the potentially most

hazardous chemicals found at the site.

3.3.3.5. Threatened and Endangered Species. By definition, threatened and

endangered species are already at risk of extinction: the loss of only a few individuals

from the population may have significant consequences for the continued existence of

the species. While threatened and endangered species and/or habitats critical to their

survival, may not necessarily be an important functional component of the ecosystem,

they are generally selected as key receptors due to their significant social and scientific

value.

3.3.3.5.1. If a species is rare, but not legally designated as either threatened or

endangered, local ecologists or other experts should be consulted to determine the

importance of the species in the context of the site. Migratory birds may also require

special consideration.

3.3.3.5.2. Federal and state natural resource trustees or other specialists should

be consulted to determine the location of such species and their potential for exposure

to the COPECs. The major sources of information on rare, threatened, and

endangered species are field offices of the USFWS and NOAA, officials of state fish

and game departments and natural heritage programs, and local conservation officials

and private organizations.

3.3.3.6. Importance of the Food Web. The purpose of determining the food web is

to evaluate pathways from chemicals in soil, sediment, or water to the affected species.

Food web analysis is most important where toxicological data indicate that the

COPECs bioaccumulate or if the potential effects on organisms might alter population

levels of one or more species. Food webs for many sites can be quite complex.

Diagramming the complete food web, however, is rarely necessary. Based on the

preliminary list of important species at the site, a preliminary simplified food web can be

drawn.

EM 200-1-4

31 Dec 10

3-8

3.3.3.7. Food Web Construction. Food web construction requires general

knowledge on the food habits of species or species groups (e.g., waterfowl,

grasshoppers, zooplankton) potentially occurring on the site. Available data on feeding

relationships, such as the percent contribution of a prey species in the diet of a

predator, can be included to indicate the strength of the feeding relationship.

3.3.3.7.1. Depending on the particular site conditions, one may construct either

one or more simple food chains, a community food web, a sink food web, or a source

food web (Fordham and Reagan 1991). A food chain would be used to illustrate the

movement of chemicals through a series of organisms by progressive consumption.

a. A community food web includes the feeding relations of the entire community.

b. A source food web includes a designated food source (e.g., a particular plant

species), all of the organisms that consume the source, and all the species that

consume these organisms up to the highest trophic levels involved (Cohen 1978).

c. A sink food web is also a subset of the community food web and includes all the

types of organisms eaten by a designated sink species (e.g., bald eagle), the food of

these organisms (e.g., fish and small mammals), and so on to the lowest level of the

food web (e.g., primary producers) (Cohen 1978). Sink food webs are especially

important where threatened and endangered species are a designated key receptor

and the pathways by which chemicals biomagnify through various trophic levels to this

receptor are to be quantified.

3.3.3.8. Keystone Species. Species that may not appear to be important may

nevertheless play significant roles in the stability of an ecosystem. Certain rodents

(kangaroo rats, prairie dogs) in the arid southwest, for example, are considered

keystone species due to their importance as prey, their practice of managing vegetation

in such a way as to control species presence, and their importance in providing habitat

for other species like burrowing owls. Certain insect groups (both aquatic and

terrestrial) may also be regarded as keystone species because of their importance as

prey for a wide variety of receptors, the profound effects they can have on vegetative

communities, and their potential importance as vectors for contaminant transport.

Because of the specialized knowledge required to recognize keystone species and

other important receptors, ecologists play a central role throughout the design and

conduct of the BERA.

3.3.3.9. Reptiles and Amphibians. Consideration of reptiles and amphibians has

generally been avoided in BERAs due to limited knowledge about contaminant effects

and issues associated with variations in bi-phasic life history strategies. Where scope

is limited in a BERA, USEPA (1986b) suggests one means for evaluating reptiles and

amphibians is to assume that when birds and mammals are protected via the risk

criteria of the assessment, then reptiles and amphibians are also protected. While

some protection is afforded reptiles and amphibians by these criteria, the level of

protection is not known. As more toxicological information becomes available on such

organisms, it should be considered more accurately in the BERA.

EM 200-1-4

31 Dec 10

3-9

3.3.3.10. Recreationally and Commercially Valued Species. USEPA (1998a)

suggests that potential adverse effects be noted on species that are of recreational and

commercial importance (e.g., sport fish, game), although as key receptors they may not

be ecologically relevant. Species that are food sources and directly support these

important species, as well as habitats essential for their reproduction and survival,

should also be considered in the planning and assessment process.

Information on which species are of recreational or commercial importance in an

area can be gathered from state environmental or fish and wildlife agencies, federal

agencies such as NOAA, USFWS, USFS, and local conservation and fish and game

personnel. Commercial fishermen's and trappers' associations may also be valuable

sources of data.

3.3.4. Ecological Endpoints. Ecological endpoints are identified within the ERA

process to provide a basis for characterizing risks to the environment. Ecological

endpoints are the particular types of actual or potential impacts a chemical or other

environmental stressor has on an ecological component (typically a key receptor).

These endpoints are of two types:

a. Assessment Endpoints

. Explicit expressions of the actual environmental value

that is to be protected, operationally defined by an ecological entity and its attributes.

(USEPA 1998a).

b. Measurement endpoints. Measurable responses to a stressor that are related

to the valued characteristics chosen as assessment endpoints (USEPA 1998a)

3.3.4.1. ERAs typically address both assessment and measurement endpoints.

Assessment endpoints are the ultimate focus in risk characterization and the link to the

risk management process. Assessment endpoints most often describe the

environmental effects that drive decision-making, such as reduction of key populations

or disruption of biological community structure. Selected assessment endpoints should

focus on identifiable harm that may come to exposed receptors. Such harm includes

death or reproductive impairment. Appropriate measurement endpoints should also

focus on determining which pathways may be complete for site COPECs and receptors.

As in the PA/SI, measurement endpoints in the BERA are frequently based on toxicity

values from the available literature. Sometimes, however, measurement endpoints are

expressed as the statistical or arithmetic summaries of the actual field or laboratory

observations or measurements.

3.3.4.2. A BERA may include descriptive sampling and measurement of ecological

attributes such as tissue residue levels or biological diversity in the contaminated area

compared to a nearby reference area. Ecological attributes that can be adversely

The term “Measurement Endpoint” has been redefined to “Measures of Effect” in the Guidelines (USEPA

1998a), noting that the latter term is more specific and less confusing. However, since ERAGS (USEPA

1997a) uses the term Measurement Endpoint, this document will as well, in order to maintain consistency

with that document.

EM 200-1-4

31 Dec 10

3-10

affected by contaminants are numerous. Selection of which attributes to measure should

be well documented and based on the TPP process (USACE 1998). Comparison of

ecological attribute measurements made at the reference and contaminated sites can

provide a qualitative measure of the ecological similarity between the two sites.

Interpretation of the significance of differences in measurements between contaminated

and reference sites is not always straightforward, especially where there are a large

number of species present and the analyses become quite complex. The detection of

differences between on-site and reference communities does not necessarily indicate that

contaminants are exerting biological effects. When quantitative risk estimates are

available and the results indicate the potential for risk, conclusions from biological field

studies and bioassays can be used as confirmatory weight-of-evidence to support risk

conclusions and interpretation. Some additional abiotic sampling and analysis may also

be needed so that the biotic data collected can be related to the chemical and physical

habitat currently affecting the biota.

3.3.4.3. When possible, receptors and endpoints are concurrently selected by

identifying those that are known to be adversely affected by chemicals at the site based

on published literature. COPECs for those receptors and endpoints are identified by

drawing on the scientific literature to obtain information on potential toxic effects of site

chemicals to site species. This process ensures that a conservative approach is taken

to selecting endpoints and evaluating receptors that are likely to be adversely affected

by the potentially most toxic chemicals at the site.

3.3.4.4. The Army BTAG has authored a position paper that provides general

recommendations in regard to selecting appropriate assessment and measurement

endpoints. See Technical Document for Ecological Risk Assessment: Selection of

Assessment and Measurement Endpoints for Ecological Risk Assessments (USA BTAG

2002b).

3.3.4.5. Population Versus Individual/Community/Ecosystem Endpoints. The

toxicity of contaminants to individual organisms (receptors) can have consequences at

the population, community, and ecosystem level. Population level effects may

determine the nature of changes in community structure and function, such as

reduction in species diversity, simplification of food webs, and shifts in competitive

advantages among species sharing a limited resource. Ecosystem functions may also

be affected by contaminants, which can cause changes in productivity, or disruption of

key processes (alteration of litter degradation rate). Potential endpoints for ERAs at the

individual, population, community, and ecosystem level include the following (USEPA

1989b):

Level 1: Individual Endpoints:

Changes in behavior

Decreased growth

Death

EM 200-1-4

31 Dec 10

3-11

Level 2: Population Endpoints:

Increased mortality rate

Decreased growth rate

Decreased fecundity

Undesirable change in age/size class structure

Level 3: Community Endpoints

Decreased species diversity

Decreased food web diversity

Decreased productivity

Change to less desirable community

Level 4: Ecosystem Endpoints

Decreased diversity of communities

Altered nutrient cycling

Decreased resilience

Altered productive capability

3.3.4.5.1. Population-level assessment endpoints are generally recognized in

ERAs because: (1) responses at lower levels (i.e., organismal and suborganismal) may

be perceived as having less social or biological significance (actions may be taken to

protect individuals of endangered species but only because it is prudent in light of the

precarious state of the population); (2) populations of many organisms have economic,

recreational, aesthetic, and biological significance that is easily appreciated by the

public; and (3) population responses are well-defined and more predictable with

available data and methods than are community and ecosystem responses.

Populations are biologically relevant because of their role in maintaining biological

diversity, ecological integrity, and productivity in ecosystems; individuals are important

only in maintaining populations. Because the environmental values to be protected are

sustainability of species or characteristics at higher levels of ecological organization

(e.g., biological diversity), the individual level is not appropriate for assessment

endpoints evaluation, except where loss of one individual could impact the survival of a

threatened or endangered population.

3.3.4.5.1.1. Ecosystem responses are characterized by many of the same

measures as communities: species composition and diversity, nutrient and energy

flows and rates of production, consumption, and decomposition. Unlike community

measures, ecosystem structure and function include nonliving stores of materials and

energy along with animals, plants, and microbes that make up the biotic portion of the

environment.

3.3.4.5.1.2. There is a general consensus among ecologists that results of

community and ecosystem studies are complex and highly variable, and therefore

difficult to interpret. One reason for this difficulty is that contaminants exert their effects

on communities both directly and indirectly. Direct and indirect toxicity can cause

changes in community structure due to differences in sensitivity among species.

EM 200-1-4

31 Dec 10

3-12

Indirect effects such as resultant shifts in diversity, productivity, or predator-prey

interactions (as the outcome of competition) are extremely difficult to predict or

measure.

3.3.4.5.1.3. Indirect effects of chemicals are often cited as justification for testing at

higher level of organization. Implementation of such testing, however, tends to be

expensive, time-consuming, present great uncertainty, and may have limited relevance

to the risk management decisions. If ecological endpoints are not appropriate and

compelling, they will not provide information relevant to site remediation decisions.

3.3.4.6. Assessment Endpoints. Most ecological assessment methods focus on

population measures as endpoints, since population responses are more well defined

and predictable than are community and ecosystem responses. The latter responses

are often more difficult to measure and interpret, highly variable, and not diagnostic of

actual exposure. Population measures can also be used to model changes at the

community or ecosystem level. Where the population is protected and individuals are

important to the overall sustained success of the population, then assessment

endpoints focus on adverse effects at the individual level.

3.3.4.6.1. Assessment endpoints are identified by drawing on the scientific

literature to obtain information on the potential adverse effects of site conditions to

populations, communities, and ecosystem levels of ecological organization. Valued

ecological resources such as trees, fish, birds, and mammal populations are typically

selected as the focus of the assessment endpoints.

3.3.4.6.2. In ERAs, ecological entities that are valued (based on a combination of

societal and ecological concerns) and to be protected are first identified and then

investigated by directly measuring appropriate ecological parameters or responses

(measurement endpoints) that are related to the assessment endpoints

3.3.4.7. Measurement Endpoints. When assessment endpoints cannot be

measured directly, measurement endpoints are selected. Measurement endpoints are

those used to approximate, represent, or lead to the assessment endpoint (USEPA

1989b). Measurement endpoints should be selected so as to provide insights related to

the specific assessment endpoint. Toxicity reference values (TRVs)(e.g., median lethal

. Unlike human

health risk assessments which focus on risk to individuals, ecological risk assessments

usually address risk at the population, community, or ecosystem level of organization.

The exception to this is in the case of endangered or threatened species, where

individuals must be protected in order to preserve the population.

For a site where there are storage yard drums leaking to a nearby stream in which there are fish upon

which bald eagles (a federally protected species) are feeding, a likely assessment endpoint would be:

impairment of reproductive success in the bald eagle. The corresponding measurement endpoint could be

dose-response data for the COPEC in a related species (e.g., another member of the order Falconiformes or

family Accipitridae). Exposure characterization could require fish and abiotic media sampling to confirm the

contaminant transport pathway and modeling of fish tissue concentrations to bald eagle tissue

concentrations. Comparison of dietary (fish) eagle concentrations and modeled eagle tissue concentrations

to concentrations known to impair reproduction in the eagle generates the risk estimate.

EM 200-1-4

31 Dec 10

3-13

dose (LD

), lowest observed adverse effects level (LOAEL), no observed adverse

effects level (NOAEL)) obtained from the scientific literature are used as toxicological

endpoints (or surrogate measurement endpoints) for the purpose of risk characteriza-

tion. Where estimated exposure concentrations far exceed the effects levels, and

adverse effects are considered likely, additional confirmatory data may be needed in

the decision-making process. For wildlife, confirmatory data may be obtained on a

variety of measurement endpoints including chemical analyses of tissue samples from

potentially exposed wildlife or their prey, or from observed incidence of disease,

reproductive failure, or death. Several factors should be examined in the selection of

measurement endpoints, including: the sensitivity of the receptor; size comparability;

diet composition and quantity; home range size; abundance; resident versus migratory

species; and whether toxicity data are available (Hull and Suter 1993). Use of field

measurement endpoints may also require comparison to a reference area. Where

biological data are to be collected, the DQO process and guidance provided in USA

BTAG (2002a) should be followed.

3.3.5. Chemical Data Collection and Review – BERA. Planning, collection, and

review of chemical data constitute the initial and often the most substantial level of

effort in a BERA. Because of the importance for obtaining useable data to the end goal

of an acceptable ERA, the following paragraphs describe the data collection and review

process in detail.

3.3.5.1. Planning and Providing Input to Data Collection. The ecological risk

assessor can effectively contribute to the data collection process when he/she is

involved early on and has some information regarding the ecological setting and the

contamination history of the site. To effectively contribute to the overall data collection

and analysis process, the risk assessor should be knowledgeable and experienced with

the overall DQO process.

3.3.5.1.1. Data needs for the ERA are likely to overlap with those for the human

health risk assessment or other data users’ needs in specific physical areas of a site.

The potential for data need overlaps should be identified early on. Nearby surface

water bodies that are potentially linked to the source through chemical fate and

transport are typically sampled for human health purposes. Sediment samples may

also be required by the human health risk assessor, but human exposure points may be

different from ecological ones, so proposed sample locations should be reviewed. The

ecological risk assessor may need water and sediment samples from specific locations

such as where waterfowl are feeding or where effects on benthic communities are likely

to occur. Similar data needs should be determined early on by the human health and

ecological risk assessors for the elimination of unnecessary work or redundancies in

sampling.

3.3.5.1.2. Historical data collected for purposes other than the ERA may be

available from previous investigations, facility records, permit applications, or other

sources. Often, use of historical data sets is limited by the lack of information on

sample locations, analytical methods, detection limits, laboratory and quality

EM 200-1-4

31 Dec 10

3-14

assurance/quality control procedures, or scope of analyses. Data from historical

sources, therefore, may not be appropriate to use in the quantitative ERA; however, it

often can be used in a supportive, qualitative role. When evaluating historical or

purposely collected data, a number of factors need to be evaluated.

3.3.5.1.3. On the other hand, unique data needs may also be identified early on in

the investigation that would require purposive (biased) sampling in order to collect

abiotic samples from specific areas of contaminant or ecological concern. On-site

animal activity should be initially observed to best evaluate obvious activity patterns

relative to the contaminant source areas. For example, if receptors of special concern

are observed on site, it may be advisable to collect chemical sample(s) from their

specific habitat.

3.3.5.1.4. The need to detect contaminants at extremely low concentrations may

also be a unique data need for the ERA. For example, some polycyclic aromatic

hydrocarbons (PAHs) (naphthalene, benzo-a-pyrene, and phenanthrene) have reported

effects levels in sediments below the normal reporting limits for these chemicals. Also,

matrix interference in soil and sediment analyses often results in detection limits well

above ecological effects levels. While it may be desirable, it is not always possible to

have the reporting limits or detection limits lower than the effects levels. Such

considerations, however, are important to the data collection planning process, the data

interpretation, and resultant risk characterization.

3.3.5.1.5. The risk assessor's data needs for a site is the culmination of the

assessor's effort to conceptualize and develop a strategy for conducting the BERA,

based on available chemical and ecological information. Often, the ecological risk

assessor is invited to merely comment or advise on a sampling program that has

already been devised for other users, which is normally not an effective way to initiate

site investigations. Other times, the ecological risk assessor may be largely responsible

for design of the entire sampling program. The level of effort for this task may range

from minimal to large and complex.

3.3.5.1.6. Appropriate sampling and analysis methods should be identified, and

detailed work plans developed. If biological sampling is short-term, seasonality of the

species, population, or community to be sampled should be carefully considered, so that

representative biotic samples can be collected. For example, if an assessment endpoint

concerns adverse effects in nesting birds, then bird surveys should be conducted in the

summer; if, however, the assessment endpoint concerns migratory birds, more

appropriate seasons for surveys are spring and fall. Also, locations of biological sampling

should be chosen in view of any previous sampling of exposure point media and any

anticipated abiotic sampling and chemical analysis.

3.3.5.2. Evaluation of Existing Chemical Data. Care should be taken where data

collected during earlier site work (e.g., the PA/SI) are largely intended for use in the

HHRA, as detection limit needs can be different for the two assessments. For example

the drinking water criterion for copper is 1.3 mg/L, while the chronic aquatic life criterion

for copper at 100 mg/L CaCO

hardness is much lower (12 µg/L).

EM 200-1-4

31 Dec 10

3-15

3.3.5.2.1. Conversely, some of the listed carcinogenic organic compounds are

relatively non-toxic to aquatic life, but have extremely low human consumption criteria

limits. The PA/SI environmental media data should be evaluated to determine whether

chemical concentrations exceed ARARs or guidance criteria. Where data gaps are

identified (e.g., chemical data are not available for the location or media of ecological

interest), then planning for additional data collection should be undertaken.

3.3.5.2.2. The need to conduct biological sampling could be indicated by

exceedance of the toxicity benchmarks or other regulatory criteria or by the presence of

organic chemicals that biomagnify. Organic chemicals with bioconcentration factors

(BCFs) greater than 100 (on a 3% mean lipid content) or logarithm of the n-octanol

water partition coefficient (log K

) values greater than 3.5 are of greatest concern

(USEPA 1991b) due to their potential to biomagnify in ecological systems

3.3.5.3. Review of Analytical Data. The quality of an ERA depends directly on the

quality of the chemical data applied. Regardless of how well other components of the

ERA are performed, if data quality is poor or data do not accurately reflect site

contamination or the types of exposures assessed, the ERA will not provide an

adequate description of potential adverse ecological effects posed by the site.

Therefore, it is imperative that data types used in the assessment be carefully

evaluated and properly used.

. Organic

chemicals with BCFs greater than 300 are considered to be of significant concern in

aquatic ecosystems, while for terrestrial organisms, BCFs as little as 0.03 can be

significant. Chemicals with water solubilities less than 50 mg/L and potential for

significant partitioning into environmental media other than air and water would also be

of concern. The presence of chemicals that can biomagnify generally results in a

greater level of effort for characterizing risk or in the need to proceed to biological

sampling.

3.3.5.3.1. Planning for appropriate data acquisition is an important step in

obtaining the necessary, high quality data. During this planning stage, appropriate

location, number and types of samples, detection limits and analytical methods can be

specified as part of the DQO process. These and other minimum requirements for ERA

data should be specified prior to data collection by having the risk assessor involved in

early stages of site planning. Once available, a thorough review of the data is needed

to ensure that DQOs and minimum requirements have been met. This further ensures

that the most appropriate information is used in the ERA.

3.3.5.3.2. Numerous factors may potentially have to be considered when

identifying minimum data collection requirements for an ERA, or when reviewing

existing data to determine usability in an ERA. Relevant guidance on data usability in

ERAs is published in the following documents:

See discussion in Section 3.3.6.10 relative to the Great Lakes Water Quality Initiative (GLWQI; USEPA

1995b) as that reference considers a log K

of 3.0 as the demarcation between bioaccumulators and

non-bioaccumulators.

EM 200-1-4

31 Dec 10

3-16

a. Guidance for Evaluating Performance-Based Chemical Data (USACE 2005)

b. Guidance for Data Useability in Risk Assessments (Parts A and B) (USEPA

1992b,c)

c. Laboratory Data Validation Functional Guidelines for Evaluating Inorganics

Analysis (USEPA 1994b)

d. Laboratory Data Validation Functional Guidelines for Evaluating Organics

Analysis (USEPA 1994c)

3.3.5.3.3. An evaluation of data quality should examine the following five broad

categories:

a. Data Collection Objectives (discussed above),

b. Documentation,

c. Analytical Methods/Quantitation Limits,

d. Data Quality Indicators, and

e. Data Review/Validation.

3.3.5.3.4. Each of these categories contains other factors that should be

considered, as well. In some cases, portions of the evaluation are performed by

disciplines other than the risk assessor (for example, data validation is most often

performed by a qualified chemist); in other cases, the risk assessor must take the lead

in acquiring and reviewing the information. In either case, the risk assessor must be

aware of the important factors within each category to enable him or her to judge

whether the data are appropriate for inclusion in an ERA.

3.3.5.4. Data Presentation and Summary. Data that have been identified as

acceptable for use in the ERA should be summarized in a manner that presents the

pertinent information to be applied in the ERA. Any deviations from the DQOs or

minimum requirements should be identified, and the potential effect upon the ERA

described in the assessment. Any data that have been rejected as a result of the data

evaluation should be identified, along with a reason for their rejection.

3.3.5.4.1. At this point in the ERA, all appropriate site data identified as acceptable

by the data evaluation process should be combined for each medium for the purposes

of selecting preliminary COPECs for the site. However, this does not mean that all

available data are to be combined. "Appropriateness" of data should take into

consideration the area of exposure to be assessed.

3.3.6. Selection of COPECs. COPECs are those chemicals that can potentially

induce an adverse response in ecological receptors. Because not all chemicals found

at a site will have adverse effects on biota, the list of chemicals to be evaluated can be

narrowed. Chemical, physical, ecological, and toxicological criteria are used in

EM 200-1-4

31 Dec 10

3-17

evaluating preliminary COPECs. COPECs typically include: (1) chemicals that are not

laboratory contaminants (i.e., chemicals whose detection has not been flagged as a

result of laboratory contamination), (2) chemicals that occur at higher concentrations

than found at background or reference sites, (3) chemicals that have the potential

(qualitatively based on concentrations detected and toxicity) to cause acute or chronic

toxicity following exposure, (4) chemicals which have the potential to bioaccumulate or

biomagnify. Although the selection process for COPECs parallels that for the human

health risk assessment, the lists may differ somewhat based on chemical fate and

transport characteristics and species-specific toxicities.

3.3.6.1. Objectives.

3.3.6.1.1. The objective of selecting preliminary COPECs for the ERA is to identify

a subset of chemicals detected at the site that have data of good quality, are not

naturally occurring or a result of non-site sources, and are present at sufficient

frequency, concentration, and location to pose a potential risk to ecological receptors.

The selection of COPECs is a process that considers site-specific chemical data in

conjunction with the preliminary ECSM that describes potential exposure pathways from

chemical sources to ecological receptors. This selection process is needed for several

reasons:

a. Not all chemicals detected at a site are necessarily related to site activities.

Some may be naturally occurring, a result of anthropogenic activities, or a result of

chemical use in off-site areas.

b. Some chemicals may be a result of inadvertent introduction during sampling or

laboratory analysis.

c. Disparities as well as similarities exist in the selection process for COPECs and

chemicals of potential concern for human health.

d. Not all chemicals detected at a site are present at high enough concentrations

to pose a potential exposure or ecological threat. Additionally there may be trace

elements present at nutritionally required or ecologically-protective concentrations.

3.3.6.1.2. The chemical selection process is performed by evaluating the data that

have been identified as useable by the data evaluation process. Chemical selection

involves evaluation of these data using criteria to identify those chemicals that are not

appropriate to retain as COPECs. Through an exclusion process, the COPECs are

selected from the list of chemicals analyzed in site media. The outcome of the

selection process is a list or lists of chemicals in site media that will be assessed

quantitatively in the ERA.

3.3.6.2. General Considerations. The following general factors should be

considered before applying the chemical selection process. These factors allow the

assessor to select the most appropriate data to include in the assessment.

EM 200-1-4

31 Dec 10

3-18

What is the exposure area?

a. An exposure area can be defined as the area in which a receptor will be

exposed to a medium through one or more exposure pathways. The boundaries of the

exposure area depend on the available pathways for exposure, the home range of the

receptor and the extent of contamination within the habitat. For example, an exposure

area may be the entire site if chemical contamination is widely dispersed, or it may be a

small sub-section of the site if chemical contamination is localized. The exposure area

may be a downwind/downgradient area for air, soil, or surface water exposure.

Because the exposure area is a function of receptor foraging range as well as areal

extent of contamination, the exposure area may include portions of the site that have

not been impacted by specific chemicals that are being assessed. For example, if a

former tank area is being assessed within a larger site, soil samples from the general

tank area should be considered as a discrete exposure area, and should not be

combined with other site soils that are remote from the tank area. When unrelated

areas of the site are combined with impacted areas, detection frequency and exposure

point concentrations can be biased low. It would be appropriate, however, to include

samples from within the defined tank area that are reported as non-detected with the

contaminated samples from within the same area since these samples are within a

defined exposure area. Under some circumstances, however, inclusion of unrelated

areas may be acceptable where doing so provides a more realistic foraging-exposure

area for a receptor population of concern.

b. Not all chemical data collected from site media represent those to which

ecological receptors are necessarily exposed. When selecting COPECs, the potential

receptors, exposure pathways, and exposure routes identified in the preliminary ECSM

should be examined. The preliminary ECSM will identify how and where exposure is

expected to occur (i.e., through soil, sediment, or water ingestion, by direct contact or

indirect ingestion, etc.). This information is then used to help identify the media and

locations where assessments will be directed and COPECs need to be identified.

c. A distributional analysis of the chemicals present at a site should be conducted.

This examination would differentiate between impacted areas and non-impacted areas.

The distributional analysis may be a statistical or a qualitative evaluation. The

distributional analysis may identify the whole site as the exposure area or only sub-units

of the site as the exposure area. It should be noted that some sites are simply too

small to support adequate populations of animals to warrant evaluation of ecological

risks. Some states do not require any evaluation of ecological threats unless the site is

at least 2 acres in size.

d. Reference area locations should not be included with site samples when

defining an exposure area. Reference locations are selected to represent off-site

conditions and to help distinguish between chemicals and ecological conditions that are

site-related and those that are not. Reference samples may or may not be "clean",

depending on local background conditions, global atmospheric deposition, other

EM 200-1-4

31 Dec 10

3-19

anthropogenic sources, or upgradient sites (i.e., other non site-related sources of

chemicals may be present), but they should not be impacted by site conditions.

Reference samples should be collected from locations un-impacted by anthropogenic

inputs, to the greatest degree reasonably possible. Reference areas may be used to

establish background chemical concentrations, if appropriate criteria are used to select

the reference areas.

Are the chemical data appropriate?

Even with high quality, useable data, the form of the chemical or sampling

technique should be examined for useability and relevance for exposure. Federal

AWQC for metals are based on total recoverable metals; measurement of dissolved

metals levels would therefore not be directly comparable (although dissolved metals

measurements do have a place in ERAs)

Are the chemical data ecologically relevant?

. Filtered water samples are generally not

relevant for most wildlife exposures. To apply federal AWQC, site-specific factors

associated with metals availability (e.g., total organic carbon, pH) and toxicity to aquatic

life need to be collected (USEPA 1993b).

Soil and sediment samples from below a predetermined biologically relevant depth

are not typically included in the terrestrial assessment. The biologically relevant depth

is based on the ecology of the site and the depth to which small mammals or other

receptors of concern (birds or invertebrates) on the site burrow and may therefore be

exposed. Feeding habits of animals also determine the type of exposure. Data

composited from multiple locations over a large area are not relevant to exposures for

animals with a small home range or specific habitat preferences.

3.3.6.3. Selection Criteria/Methodology. Criteria that can be applied to determine

whether a chemical should be removed as a potential COPEC must be fitting to the

selected or anticipated ecological endpoints and the overall adequacy of the sampling

program. The process for selecting COPECs is not entirely standardized or

mechanistic, but employs a considerable amount of professional judgment throughout

the process. For example, the assessor should consider whether limited chemical

distribution or limited presence is an artifact of sampling inappropriate media or

locations. Could site-related COPECs potentially exert similar toxic action as

background "contaminants" or exacerbate the toxicity of the background

"contaminants"?

USEPA has published metals ratios so that comparisons can be made between dissolved and total metals

concentrations (see Water Quality Standards: States Compliance - Revision of Metals Criteria, Interim Final

Rule, 60 FR 22229 (USEPA 1995d)).

The decision to carry forward all detected compounds into the

exposure and effects characterization portions of the or BERA is sometimes made

depending on the number of chemicals detected and project scope. More often, risk

assessors chose to sequentially eliminate chemicals through the progressive

Contaminants, in this case, refers to naturally-occurring metals or organics or chemicals present as a result

of large, regional-scale contamination.

EM 200-1-4

31 Dec 10

3-20

application of screening criteria. Through this elimination process, the risk assessor

assures that all chemicals are addressed (not overlooked), but that only the relevant

chemicals are carried forward into the quantitative risk analysis. Examples of screening

criteria include the following:

a. Non-detection (use of appropriate detection limits);

b. Limited chemical distribution and limited presence in environmental media;

c. Comparability with screening criteria (AWQC, effects range – low (ER-L), lower

effects level (LEL), etc.);

d. Comparability with background concentrations (consideration of site-

relatedness);

e. Non-site-relatedness;

f. Role as an ecologically essential nutrient at site concentrations;

g. Low toxicity/bioconcentration screen; and

h. Low potential for bioaccumulation and biomagnification.

3.3.6.3.1. These criteria are typically applied sequentially to the available data.

Once a chemical is eliminated based on a screening criterion, it is not considered in

subsequent screenings or assessments. Each of the above criterion is discussed

further in the following sections.

3.3.6.3.2. The ECSM will often identify two or more ecological receptors of

concern, particularly where both terrestrial and aquatic ecosystems are present. In

these cases, the COPEC selection process is branched: one branch focuses on

aquatic receptors, the other branch focuses on terrestrial receptors. Within the

terrestrial COPEC selection process, further branching may occur in those cases where

the chemicals are known to bioaccumulate. Where there are migratory birds and higher

trophic level predatory raptors present, for example, one branch would focus on the

COPECs that may have acute or chronic effects on migratory birds, and the other

branch would focus on chemicals that bioaccumulate and may affect the top trophic

level receptors (e.g., raptors).

3.3.6.4. Non-detection. Chemicals analyzed for but not detected in any sample of a

site medium should not be included as COPECs for that medium. To be selected, a

chemical must be detected in at least one sample of the environmental medium of

interest (i.e., the results are not all reported as non-detect and qualified with a "U").

Where samples have an associated duplicate analysis, the mean of the two samples (if

both were detected) is usually presented ; if both the sample and the duplicate results

were not detected (ND), then the lower of the two censoring limits should be used; if

one result is detected and the other is ND, then the detected concentration is reported.

EM 200-1-4

31 Dec 10

3-21

3.3.6.4.1. Care must be taken when evaluating non-detects with very high

censoring limits as non-detects may mask the presence of a chemical if the risk-based

decision limit that is less than the censoring limit. Although a quantitative estimate of

the chemical's concentration value is unavailable in such a case, the chemical may

need to be assessed qualitatively if it is present in other site media.

3.3.6.4.2. Detection levels also need to be evaluated with respect to ARARs and

toxicity screening levels. For some PAHs and dioxins, quantitation limits below the

estimated toxicity effects level for a particular receptor of concern may not be possible.

For other chemicals, such as mercury, the detection limit (0.01 µg/L) is barely below the

AWQC (0.012 µg/L).

3.3.6.5. Chemical Distribution. The physical distribution and frequency of detection

of a chemical in a site medium or exposure area can be used to remove a chemical

from consideration as a COPEC. The premise behind this criterion is that a chemical

with limited presence in a medium or exposure area is unlikely to be contacted

frequently and, therefore, does not pose as great a potential ecological risk as do more

frequently detected chemicals.

3.3.6.5.1. The distribution of the chemicals present in a site or exposure area

should be examined by identifying where the chemicals were and were not detected

and their frequency of detection. If this evaluation indicates that the distribution of a

chemical is low, i.e., it is detected in only one or a few locations, it may be reasonable

to exclude it as a COPEC (assuming an appropriate sampling design was used), or to

select the chemical as a COPEC for a smaller exposure area of the site. Within the

smaller exposure areas, chemicals detected in five percent or fewer samples may also

be considered for elimination.

3.3.6.5.2. The following factors should be considered when applying this criterion:

a. The number of samples available

. In a small data set, a limited frequency of

detection of a chemical may be more a statistical artifact of a limited sampling design

rather that the infrequent presence of the chemical.

b. The quantitation limit and censoring limit. If the quantitation limit or censoring

limit exceeds a risk-based limit, it is typically not possible to determine whether or not

contamination is present above or below the risk-based threshold.

c. The sampling scheme. Biased sampling plans may over-represent the

occurrence of chemicals).

3.3.6.6. Comparability with Background Concentrations. In conducting a risk

assessment, it may be important to distinguish site contamination from anthropogenic

or naturally occurring background in order to determine the presence or absence of

contamination and to compare with background risk (USEPA 1992b,c). Some

chemicals detected in site media may be naturally occurring or present as a result of

ubiquitous or off-site chemical use. Therefore, it is appropriate to exclude them from

EM 200-1-4

31 Dec 10

3-22

the risk assessment. Background samples are kept separate from the site data for the

purposes of assessing exposures, and are used exclusively to identify non-site-related

chemicals.

3.3.6.6.1. The most appropriate measure of background quality is obtained by the

collection of background data from unaffected on-site areas or nearby, off-site areas, or

reference areas. The risk assessor should be involved in the selection of background

sample numbers, types, and locations as part of the ERA minimum data requirements,

to ensure that adequate data are collected. When selecting COPECs, the background

data collected should be reviewed to identify whether minimum requirements have been

met, or in the case of historical data, whether background measurements are adequate.

The following factors should be considered:

Are the locations of the background samples appropriate?

a. Appropriate background sampling locations vary with the media being

examined, but should generally be offsite; hydrologically upgradient for surface water

and sediments; upwind of the site at the time of measurement and under usual climate

conditions for air; and in areas remote from surface water drainage for soil.

Background samples should also be located away from other potential off-site sources

of contamination that would not impact the site, such as other sites, roadways, etc.

b. If off-site areas have the potential to contribute chemicals to the site being

assessed (for example, upgradient industrial facilities), part of the goal of identifying

appropriate background sample locations should be to obtain sufficient background

samples to identify potential chemical contributions from off-site sources.

Are the background samples comparable in type to the media being examined?

Background samples should be as similar as possible to the site samples being

evaluated. Background sampling locations should have similar habitat and soil

conditions to the onsite locations. Soil and sediment depths and stream characteristics

should be comparable. The type of analyses performed on site and background

samples (such as filtered versus unfiltered water, soluble versus total metals) should

also be comparable.

Are the number of background measurements sufficient?

a. Erroneous conclusions may be drawn if the number of background samples

collected is insufficient to adequately describe background. The number of background

samples should be specified as a minimum requirement during the project planning

stage. The development of statistically based data quality objectives (e.g., per USEPA

2000b) and sampling designs per USEPA QA/G-5S (USEPA 2002b) is typically the

most effective approach to help ensure that the sample size will be adequate for

background comparisons. The actual number of samples with data available should be

examined to determine if the minimum requirements have been met. For historical

EM 200-1-4

31 Dec 10

3-23

data, it is important to determine whether the nature (e.g., locations) and number of

background samples is appropriate or if additional samples are needed.

b. Sampling data from Superfund sites have shown that data sets with fewer than

10 samples per exposure area provide poor estimates of the mean concentration, while

data sets with 10 to 20 samples per exposure area provide somewhat better estimates

of the mean, and data sets with 20 to 30 samples provide fairly consistent estimates of

the mean. In general, the sample mean approaches the true (population) mean as

more samples are included in the calculation.

3.3.6.6.2. Acquisition of site-specific background information is always preferable

to regional or national values when examining site-relatedness and comparability to

background concentrations. Literature values describing regional or national

background ranges for chemicals in soil, groundwater, surface water, and sediments

may be used, but only if site-specific background is unavailable. Regional or national

ranges are relatively insensitive and can lead to the erroneous inclusion or exclusion of

a chemical as a COPEC. If historical data include National Pollutant Discharge

Elimination System data, they may be used in addition to any other regulatory-required

data acquisition.

3.3.6.7. Comparison to background. Determination of comparability with

background can be accomplished in several ways, depending on the amount of data

available. Statistical and non-statistically based evaluations have been historically used

to compare site and background concentrations. A statistical evaluation is best when

there are sufficient site and background samples to do statistical comparisons (e.g., to

test the null hypothesis that there is no difference between the site and background

mean chemical concentration at a defined level of confidence). This approach requires

quantitative data quality objectives (DQOs) to be established during project planning

and needs to be supported by statistically based sampling design. Prior sampling data

is also desirable to help estimate the number of sample size that will be required to do

the comparisons.

3.3.6.7.1. Several statistical tests are available with which to determine whether

the two data groups, background and site, are comparable. Texts on statistics, such as

Zar (1984), Ludwig and Reynolds (1988), or Gilbert (1987), should be consulted for

tests applicable for use in specific site conditions. Also refer to EM 1110-1-4014

(USACE 2008). Test selection depends upon data distribution (normal, non-normal),

whether non-detected values are included, if appropriate proxy values are used,

number of samples, and other factors. Statistical evaluations are often the most

reliable method for comparing site and background data.

3.3.6.7.2. A geochemical approach can be very effective for distinguishing

anthropogenic from naturally occurring metal concentrations, particularly when it

supplements traditional quantitative statistical evaluations. Geochemical evaluations

can identify naturally occurring metal concentrations that would be erroneously

identified as site-related by traditional statistical evaluations. The geochemical

approach can not only be used to determine whether a study area has been affected by

EM 200-1-4

31 Dec 10

3-24

anthropogenic metal contamination but can also identify the individual sampling

locations that are suspected to possess the elevated metal concentrations. However, it

is important to recognize the limitations of this approach. Its primary disadvantage is

that it is subjective, as it is predominately qualitative in nature. In particular, decision

errors are not quantified and well-defined criteria for distinguishing native from

anthropogenic metal concentrations are not specified. In addition, although the

approach distinguishes anthropogenic metal contamination from naturally occurring

concentrations, it does not distinguish site-related contamination from non-site-related

anthropogenic metal contamination. In other words, elevated contamination relative to

background identified by the geochemical approach may be consistent with

anthropogenic background. Statistical comparisons using a background study area

would typically be needed to distinguish site-related contamination from total

background metal concentrations (from anthropogenic and non-anthropogenic

sources). Lastly, an additional limitation of the approach is that it implicitly assumes

that, at most, only a portion of the site has been impacted by anthropogenic metal

releases. This assumption is typically reasonable but can be violated if the study area

is too small (i.e., is predominately limited to a “hot spot”). (Refer to USACE 2008 for

additional information on geochemical evaluations.)

3.3.6.7.3. Non-statistical numerical comparisons are often done when background

data are limited or when minimum requirements for ERA data collection have not been

met and less than optimal numbers of background sample results are available.

However, when possible these types of comparisons should be avoided, as they are not

as reliable as statistical and geochemical evaluations and can result in incorrect

decisions (e.g., may produce high false positives or false negative rates). Examples of

non-statistical numerical comparisons include the following:

a. Comparison of site and background arithmetic mean concentrations;

b. Comparison of range of detected concentrations in both data sets.

c. Comparisons of maximum detected values.

3.3.6.7.4. Sometimes some multiplicative factor of the mean is used to determine

“comparability.” As an example of this approach, site samples could be defined as

comparable if the mean concentration is less than or equal to two times the mean

background concentration. However, it is emphasized the criterion for "comparability" is

rather arbitrary for these types of comparisons, and may produce conclusions that are

either incorrect or inconsistent with statistical evaluations.

3.3.6.8. Determination of Site-Relatedness. Background sampling is conducted to

distinguish site-related contamination from naturally occurring or other non-site-related

levels of chemicals (USEPA 1989d). In some instances, comparison with background

is insufficient to identify chemicals that are derived from other sources, despite

appropriate planning of background sample locations. If such chemicals are not site-

related, however, they generally should not be included quantitatively in the ERA,

EM 200-1-4

31 Dec 10

3-25

although this decision requires professional judgment for reasons noted earlier, as well

as and policy and legal considerations.

3.3.6.9. Trace Element and Essential Nutrient Status. Some chemicals are

essential trace elements or nutrients in the diet of plants or animals, and may be

present in site media at nutritionally required concentrations or ecologically-protective

levels. The following chemicals can be evaluated with regard to essential trace element

or nutrient status:

If adequate and confirmable information is

available that identifies a different site as the source of a chemical, even in the absence

of background information, it may be appropriate to exclude that chemical as a COPEC.

The supporting information must be conclusive and presented in the report.

Calcium

Copper

Chromium (trivalent)

Magnesium

Manganese

Iron

Potassium

Selenium

Sodium

Zinc

3.3.6.9.1. Elements that serve as nutrients and are within the recommended

allowable dietary range for some receptors may be toxic to other ecological receptors at

the same concentration (McDowell 1992). For example, metals such as copper may

not be toxic to animals which drink the water, but may be toxic to aquatic organisms.

The toxicity of such chemicals should be evaluated in light of the potential site-specific

receptors. As a general screening tool, the nutritional requirements of domestic

animals (mammals and birds) can be used to assess whether site concentrations of

these elements are within acceptable ranges or are likely to pose a hazard to on-site

receptors. Nutritional requirements and limits for livestock and experimental laboratory

animals (e.g., small mammals, birds, fish) are well-established.

3.3.6.9.2. The evaluation of chemicals as trace elements or dietary requirements

may be made on a qualitative or quantitative basis. Elements such as calcium, iron,

magnesium, potassium, and sodium are rarely retained as COPECs, for example. It

should be noted in any case, however, whether the elements could be present at a site

as a result of site activities. If it is known that a particular element's occurrence is a

result of site activities, it may not be appropriate to remove it from the list of COPECs.

3.3.6.10. Preliminary Toxicity Screen. A toxicity screen to determine which

chemical concentrations exceed toxicity benchmarks may be performed for the

selection of COPECs. Various TRVs for water and sediment developed by USEPA

(1986a, 1993a, 1994d, 1995b,d, 1996) can be used. Oak Ridge National Laboratory

(ORNL)(1998) has also developed screening benchmark preliminary values for aquatic

Recent court cases, plus policies adopted by some states, suggest that "non-site-relatedness" is not an

appropriate criterion; mere presence of a potential COPEC may require a response, while the assessment or

assignment of liability for that response must be determined separately and is not to interfere with the

response assessment.

EM 200-1-4

31 Dec 10

3-26

and terrestrial ecosystems. Guidance values from NOAA (Long and Morgan 1991),

Washington State Department of Ecology (1991), Florida Dept. of Environmental

Protection (MacDonald 1994), and Canada (Long et al. 1995, Persaud et al. 1992,

CCME 1995) for marine and freshwater sediment threshold environmental effects levels

can be used directly in screening for COPECs in aquatic ecosystems with few or no

modifications. Additional toxicity benchmarks for aquatic ecosystems may be

developed using information provided in USEPA databases such as ECOTOX and

ASTER, available at http://www.epa.gov/med/prods_pubs.htm

3.3.6.10.1. Standardized values to perform a toxicity screen of chemicals in

terrestrial ecosystems are generally not available, although ORNL (1998) has published

toxicity benchmarks for a variety of species that can be used in a terrestrial toxicity

screen. Standardized values for screening terrestrial wildlife, the ecological soil

screening values (EcoSSLs) are currently under development by USEPA, and many

values have already been published (http://www.epa.gov/ecotox/ecossl/). Four water

quality criteria (mercury, p,p'-dichlorodiphenyl-trichloroethane (DDT), 2,3,7,8-

tetrachlordibenzo-p-dioxin (TCDD), and polychlorinated biphenyls (PCBs)) for the

protection of wildlife (birds and mammals) which feed on aquatic organisms are

published in the GLWQI Final Rule (USEPA 1995b). In a few cases, chronic federal

AWQC for chemicals that bioaccumulate are based on final residue values and the

protection of sensitive mammals (PCBs and mink) or birds (DDT and brown pelican).

Where such exposure pathways are appropriate, the GLWQI criteria and federal and

state AWQC should be used in screening water concentrations for COPEC selection. A

cautious approach should be used in COPEC screening as toxicity can differ among

similar receptor species due to differences in either physiology or exposure.

3.3.6.10.2. In terrestrial ecosystems, chemicals may be very limited in distribution,

but still present potential for acute toxicity for ecological receptors. For those chemicals

that are found at limited locations or in five percent or fewer samples and tend not to

bioaccumulate, the median lethal concentration (LC

) values (for plants or soil-dwelling

organisms) may be compiled from available ecotoxicological literature and compared to

the concentration in soil. The concentration term for each chemical in soil is the lower

of (1) the maximum detected concentration or (2) the 95% UCL of the mean.

3.3.6.10.3. Chemicals that have the potential to bioaccumulate or biomagnify

through the food web should be retained for consideration as COPECs, even where

distribution is limited or they might be eliminated based on the preliminary toxicity

screen. Chemicals that bioaccumulate include those that are taken up by an organism

either directly from exposure to a contaminated medium or by consumption of food

containing the chemicals (Rand and Petrocelli 1985). Chemicals that biomagnify are

those that are found in increasingly higher tissues concentrations in higher trophic

levels (i.e., concentrations increase across at least two trophic levels) (USEPA 1995b).

By definition, chemicals that tend to biomagnify also bioaccumulate. Chemicals with a

log K

of less than 3.0 or an organic carbon partition coefficient (K

) of less than 500

(i.e., logarithm of the organic carbon partition coefficient (log Koc) less than 2.7) are not

expected to bioaccumulate or biomagnify. A lengthy list of bioaccumulative(biomagnify)

EM 200-1-4

31 Dec 10

3-27

and nonbioaccumulative chemicals that are of potential concern is presented in the

GLWQI (USEPA 1995b)

3.3.6.10.4. The chlorinated pesticides are the most well known of the chemical

groups that tend to bioaccumulate and biomagnify. PCBs and dioxins/furans are also

strong bioaccumulators and biomagnifiers. Volatile organic compounds (VOCs) such

as tetrachloroethene, toluene, trichloroethene, 1,1,1-trichloroethane, and xylenes are

unlikely to bioaccumulate and biomagnify (Van Leeuwen et al. 1992; USEPA 1982).

Semivolatiles, including PAHs, tend not to bioaccumulate and show little tendency to

biomagnify because they are readily metabolized (Eisler 1987b, Beyer and Stafford

1993).

3.3.6.11. Presentation of COPECs. The chemical selection process results in a

select list of preliminary COPECs that will be quantitatively assessed in the BERA.

Tables should be developed identifying the COPECs selected for each medium and/or

exposure area. All chemicals that were removed from consideration should be

identified, with an explanation of the reason for the removal. A flow diagram illustrating

the COPEC selection process should be included to clearly illustrate the decision

process used.

3.3.7. Level of Effort. The BERA may include laboratory or field bioassays and/or

more detailed, sophisticated computer models or probabilistic methods. Quantitative

biological samples, as well as abiotic samples, as needed, may be collected to document

exposure, to assess bioaccumulation potential, or to determine dose-response of the

tested species or the selected receptors when exposed to site media. Limited field

investigations may be conducted to determine presence of specific receptors or to

estimate biodiversity. It also may include short-term toxicity tests or bioassays, standard

rapid biological field assessment protocols, or focused tissue residue analyses of key

receptors or their prey. As needed, semi-quantitative sampling of the contaminated area

and a reference site may be conducted to describe the identity and populations of biota in

both areas. If limited fate/transport modeling (e.g., one-dimensional analytical model) is

used, site-specific input values for key parameters of the model may be needed. This

data, when integrated with chemical data, should generally be adequate to provide

information on the significance of potential or observed ecological effects, the need for

remediation/removal actions, and the development of preliminary cleanup goals based on

ecological concerns and remedial action objectives.

The GLWQI table is based on chemicals that bioaccumulate and are of initial concern in the Great Lakes

because of their strong tendency to biomagnify. Chemicals listed in this table as "not of concern" are still of

considerable concern due to their bioaccumulation potential. Chemicals that bioaccumulate in lower level

organisms may still present a significant contaminant pathway and dietary hazard to higher trophic level

receptors, even if they don't biomagnify in the latter. For example, copper is bioaccumulated to very high

level by oysters, but does not biomagnify through food webs. PAHs are accumulated in invertebrates which

lack metabolic pathways for their excretion, yet are not accumulated in most vertebrates which have such

enzyme systems.

EM 200-1-4

31 Dec 10

3-28

3.3.7.1. A variety of ecological evaluation tools, techniques, or approaches may be

used to evaluate and estimate the magnitude and importance of the potential risk.

Such techniques vary in level of effort, sophistication, and cost, but the most

sophisticated or time-consuming techniques are not necessarily the most appropriate to

a given site. Assessment of chemical effects on key receptors is directly dependent on

the use of evaluation techniques appropriate for the assessment and measurement

endpoints. Decisions as to which techniques to use should be well documented during

problem formulation.

3.3.7.1.1. Each of the evaluation techniques has its own unique advantages and

disadvantages in terms of the data and information provided. Some of these tools are

useful to measure effects at the individual operable unit and species level; e.g., field

sampling of tissue residues. Tools, such as Habitat Evaluation Procedures (USFWS

1987) and Index of Biological Integrity (IBI) (Karr et al. 1986), can be used to quantify

adverse effects to biological resources at the community/ecosystem level by measuring

reductions in habitat quality. Others such as toxicity tests are used to characterize

cumulative hazards from multiple chemicals with no attempt to apportion chemical

contribution from the individual COPECs or to discern mechanisms of chemical

interactions. Tools such as probabilistic pathways analysis are most appropriate when

there is an endangered species at risk from chemicals that bioaccumulate. To measure

critical ecosystem functions such as nutrient cycling, tools other than those listed may

be needed.

3.3.7.1.2. Each technique has its own peculiarities in terms of the interpretation of

results, and many of these tools cannot account for such phenomena as biological

resistance and avoidance. Also, some of these tools are restricted as far as their

applicability (e.g., Wetland Evaluation Technique (WET) may only be used in wetlands).

No single species test, indicator parameter analysis, statistical procedure, or field

inspection review can address the complex nature and extent of contamination or risk in

biological systems. Impacts at one hierarchal level do not always translate easily into

effects at other levels, and emergent system-level properties cannot be studied at lower

levels of organization (Kimball and Levin 1985). Chains of influence are common

features of ecosystems, and indirect effects, which can be more important than direct

effects, often predominate in ecosystems (Kimball and Levin 1985, Johnson et al.

1991). To thoroughly evaluate ecosystem risk, multimedia (i.e., air, water, soil,

sediment, and biota) as well as different trophic and hierarchal (organism, community,

population, ecosystem) levels may all need to be addressed or measured.

3.3.7.1.3. The following paragraphs are intended to guide the risk assessor in

determining, on a site-specific basis, the level-of-effort necessary to address data gaps

and complete the BERA with an acceptable level of uncertainty.

3.3.7.2. Examples of some ecological evaluation techniques and tools (and

references where descriptions of the approach may be found) include:

EM 200-1-4

31 Dec 10

3-29

a. HQs,

b. Sediment-Water Equilibrium Partitioning (USEPA 2005a, Chapman 1997) or

Water Quality Approach (Long and Morgan 1991),

c. Screening Level Concentration Approach (Long and Morgan 1991)

d. Apparent Effects Threshold (AET) or Species Approach (Long and Morgan

1991),

e. Bioeffects/Contaminant Co-Occurrence Analyses Approach (Long and Morgan

1991),

f. Sediment Quality Triad Approach (Chapman 1989, Porebski et al. 1999),

g. Rapid Bioassessment Protocols (Barbour et al. 1999, USEPA 1989g),

h. Sediment Quality Criteria Approach (Chapman 1989),

i. Bioassay Approach (Toxicity Tests) (USEPA 1989b, USEPA 1994f, USEPA

1994g, ASTM 2005),

j. Diversity Indices (Pielou 1975),

k. Species Richness/Relative Abundance Indices,

l. Wetland Evaluation Technique (USACE 1987, Clairain and Smith 1988),

m. Index of Biological Integrity (Karr et al. 1986),

n. Habitat Evaluation Procedures (USFWS 1987),

o. Exposure Pathway Analysis (Fordham and Reagan 1991),

p. Probabilistic/Sensitivity/Uncertainty Analysis (MacIntosh et al. 1994, USEPA

2001b),

q. Linear Structural Modeling (Johnson et al. 1991), and

r. Linked Deterministic and Simulation Models.

3.3.7.2.1. Standardized protocol and detailed descriptions of some of the numerous

ecotoxicological investigative methods available are provided in various agency (USEPA,

ASTM, the U.S. Food and Drug Administration (FDA), the U.S. Army Edgewood Chemical

Biological Center (USAECBC), NOAA, the Water Environment Research Foundation

(WERF)) publications.

3.3.7.2.2. Biological-Based Assessments. Bioassays or measurements of biological

integrity, rather than chemical analyses, may be preferred, or even required under some

federal regulations (40 CFR, Part 227.13, Federal Regulations on Ocean Dumping of

Dredged Sediments, USEPA 1991c, USEPA 1998c) to determine whether a particular

EM 200-1-4

31 Dec 10

3-30

abiotic medium (sediment, soil, surface water) is toxic to biota or contains chemicals at

concentrations of ecological concern. Decisions as to which method to use depend on

project objectives, data needs, desired certainty level, and the suitability of each method

to meet these needs.

3.3.7.2.2.1. In addition to methods described in the field studies and laboratory

studies sections below, the following descriptions mention only a few of the numerous

field and laboratory methods that may be employed to better characterize risk or provide

a basis for remediation decision-making. The need for measuring additional

ecotoxicological endpoints should be carefully evaluated. The risk assessor should

consider the following specific criteria:

a. The biological response is a well-defined, easily identifiable, and documented

response to the designated COPECs (i.e., methodology and measurement endpoint are

appropriate to the exposure pathway);

b. Exposure to the COPEC is known to cause the biological response in laboratory

experiments or experiments with free-ranging organisms;

c. The methodology is capable of demonstrating a measurable biological response

distinguishable from other environmental factors such as weather or physical site

disturbance;

d. The biological response can be measured using a published standardized

laboratory or field testing methodology; and

e. The biological response measurement is practical to perform and produces

scientifically valid results (e.g., sample size is large enough to have useful statistical

power and small Type II error).

Significant Ecological Threats:

a. The questions the risk assessor must keep in mind are "Do any ecological

threats exist?" and "Are these ecological threats related to chemical contamination?”

Using the information discussed above, the risk assessor can begin to identify the

habitats potentially affected by contaminants at the site. Decisions can be partly based

on absence of biota where expected, especially if plant or animal life is absent along

likely contaminant exposure pathways. For example, if areas within the project

exposure pathways(s) are devoid of plant life or are obviously stressed, a significant

ecological threat probably exists. If there is a groundwater or surface water discharge

zone to a stream that is affected by site chemicals and depleted of biota, that would be

an obvious significant ecological threat. If effects are less obvious, then it may be

necessary to use a more sophisticated approach to determine any impacts, such as a

comparison of site biota diversity and relative numbers to an unaffected reference site

within or adjacent to the watershed. For specific models and methods that may be

employed, publications from USEPA (1989b), Wentzel, et al. (1996), WERF (1994), and

NOAA (1992) can be consulted.

EM 200-1-4

31 Dec 10

3-31

b. A relatively new alternative approach to ERA poses the essential question "Are

ecological receptors at the contaminated site displaying signs of stress or impact as a

result of having been chemically exposed?" With small rodents having been exposed

for tens (and in some instances, hundreds) of generations by the time ERA efforts

commence (owing to their relatively short life spans, and contaminant-release events

having occurred decades ago), every opportunity for toxicological effects to have been

elicited in on-site receptors has been afforded. For close to a decade the Army has

employed the patent-pending Rodent Sperm Analysis (RSA) method (Tannenbaum et

al. 2003; Tannenbaum et al. 2007) at terrestrial sites, noting that small rodents are for

all intents and purposes the maximally-exposed receptors (given their miniscule home

ranges and their non-migratory nature). The RSA method interprets an exceedance of

just one of its sperm parameter thresholds to mean that larger, wider-ranging, and

higher-trophic level site mammals (i.e., species for which a site cleanup could

realistically proceed) as also being reproductively compromised. Conversely, if

maximally-exposed small rodents are not found to be reproductively compromised,

there is no reason to suspect that the other mammals are experiencing reproductive

stress or impact. One key advantage of the RSA method over desktop approaches to

ERA is that the potentially exhaustive suite of site stressors is evaluated for its ability to

have compromised reproduction, and thereby overall health.

3.3.7.3. Short-Term Assessments. The following biological sampling methods are

simple, short-term and reasonably inexpensive. The resultant information can be used

to provide further quantification of ecological risk assessment and to improve risk

interpretation through additional weight-of-evidence and should reduce the uncertainties

of the BERA.

3.3.7.3.1. Field Studies.

a. Quantitative (semi-quantitative) descriptive sampling in contaminated and

reference areas to confirm the identity and quantity of potentially exposed biota or to

measure other ecological attributes such as biological diversity (Noss 1990, Debinski and

Brussard 1972). For example, data on vegetation community composition, structure

anddiversity can be collected using semi-quantitative methods such as relevé analysis

and Braun-Blanquet rating methods (Mueller-Dombois and Ellenberg 1974).

b. Tissue sampling of key receptor species or their dietary or prey items to document

exposure. Tissue residue studies are used to provide site-specific estimates of exposure

to higher trophic level organisms and to relate tissue residue levels to concentrations in

abiotic environmental media. Knowledge of the physiology and biochemistry of the

species to be sampled for residue analysis is important. Species vary in their ability to

metabolize various contaminants (e.g., fish can metabolize PAHs).

c. Rodent Sperm Analysis (RSA). Adult male rodents of one or more species that

occur at both the contaminated site and a nearby, habitat-matched non-contaminated

reference location are live-trapped. Sperm parameter (count, motility, morphology)

population means are compared, and the absolute differences (i.e., reduction in count,

EM 200-1-4

31 Dec 10

3-32

reduction in motility, increase in morphology) are compared to established thresholds for

reproductive effect in site rodents. Corroborative RSA population metrics include total

animal captures, number of species, sex ratios, and age distributions. Additional somatic

corroborative metrics include organ-to-body weight ratios and tissue histology.

d. One-time collection of exposure point media (e.g., surface water, sediment) for

use in short-term (acute) laboratory bioassays.

e. In situ acute bioassays, possibly using exposure point surface water and

upstream water for dilution, to determine the LC

contaminant concentration.

f. One-time confirmation surveys of federal or state-protected species to confirm

their presence or document their potential presence (or presence of suitable habitat) and

potential exposure to suspected COPECs. This is in keeping with the NCP directive to

"assess threats to sensitive habitats and critical habitats of species protected under the

ESA" (USEPA 1990a)

g. If needed, one-time collection of exposure point abiotic media (e.g., soils,

sediment, surface water) for additional chemical analysis to supplement existing chemical

data.

h. If needed, one-time collection of physical media from reference areas.

3.3.7.3.2. Laboratory Studies.

a. Laboratory analysis of biological samples (e.g., periphyton, benthic invertebrates,

plants), as needed for taxonomy.

b. Chemical analysis of collected tissue samples for COPECs that are known or

suspected of bioaccumulating or biomagnifying.

c. Acute bioassays using on-site exposure media to determine an LC

or LD

d. Additional chemical analysis of exposure point media for specific species of

COPECs (e.g., chromium [+6] instead of total chromium) or selected COPECs at

detection levels lower than TRVs for the selected ecological receptors.

e. If needed, chemical analysis of physical media collected from reference areas.

3.3.7.4. Longer-Term Assessments. There may be a need for longer-term field or

laboratory studies (1 year or more), and employment of more extensive (and more

expensive) tests to resolve issues presented by larger sites having complex ecosystems

and food webs. Depending on site conditions and complexity, these may be the most

appropriate type of additional investigation. The biological sampling may involve long-

term (chronic) bioassays or tissue analysis of additional organisms or for additional

analytes, and/or additional quantitative biological (i.e., population) sampling development.

Data from quantitative surveys of populations and comparisons with reference location

EM 200-1-4

31 Dec 10

3-33

population characteristics may also be obtained.

a. Results of these field and laboratory investigations can fill the data gaps

previously identified, and may supplement the results from studies conducted previously.

The combined results are used to present risk estimates with less uncertainty, and

provide a rationale for multiple year assessments (see Section 3.3.7.5), if needed.

Chemical analyses of abiotic

exposure media also may be appropriate in order to ensure areal and temporal

correlation with biological data. Ecosystem function or other field data may be collected,

including nutrient loss (amount of undecomposed litter), biomarkers, histopathological

examinations, or mesocosm studies (in-situ biomonitoring). Site-specific input values for

key parameters of the model are also needed, if more sophisticated fate and transport

modeling is planned. Biological modeling may include single species modeling to

evaluate exposure-response when that species is co-located with multiple contaminants,

or multiple-species pathway analysis to simulate bioconcentration or bioaccumulation

within the community food web.

b.. Population studies may be required in the event that there is an apparent decline

in a key receptor's population size that is deemed important in the presence of a low HQ,

or no apparent effect on population size in the presence of a high HQ. Population studies

are typically more long-term and complex, although simple, short-term population studies

may be performed. Population studies involve taking a census of the number of

individuals in each life stage at several points over the course of one to several life cycles

or seasons. These studies can be expanded by including observations of the health or

intoxication of individuals at different life stages for each time interval. The temporal

aspects of the study design are likely to provide insight into age-related or life-stage-

specific sensitivities of the organisms in question.

c. This may also include sampling for model development or pattern description.

Data may be collected to support single-species exposure models that employ Monte

Carlo analysis techniques or integrated fate, accumulation and effects models, such as

the pathways analysis model for estimating water and sediment criteria (Fordham and

Reagan 1991). More intensive sampling to describe spatial patterns in biota and the

extent of contaminant distribution in relation to these biological patterns may also be

conducted.

3.3.7.4.1. Field Studies.

a. Quantitative biota (population/community) sampling extending over multiple

seasons within one year to document seasonal variability of potentially exposed biota.

b. Quantitative biota sampling in reference areas employing the same methodology

These characteristics include abundance, age structure, reproductive potential and fecundity

proportion, productivity, standing crop or standing stock (total biomass), food web or trophic diversity,

species diversity and dominance, presence of pollution tolerant/absence of pollution intolerant species,

etc.

EM 200-1-4

31 Dec 10

3-34

used at the exposure points to provide sufficient data for statistical comparisons to the

data collected at exposure points.

c. Additional tissue sampling of the key receptor species or their diets or prey.

d. Collection of exposure point media (e.g., surface water, sediment) for use in

additional acute or chronic (long-term) laboratory bioassays.

e. In-situ acute or chronic bioassays to determine LC

, LOAEL, or NOAEL

contaminant concentrations.

f. Additional surveys of federal or state-protected species suspected of being

exposed to COPECs.

g. Additional sampling of abiotic exposure point media (e.g., soils, sediment, surface

water) to supplement existing chemical data and correlate with the biological samples.

h. Additional collection of abiotic media from reference areas for chemical analyses.

3.3.7.4.2. Modeling Studies.

a. Single-species modeling, which is a toxicity model based on a well documented

exposure-response relationship between a mixture of chemicals and a single species,

can be run using Monte Carlo simulations to produce a cumulative distribution of

projected ecological risk and can be run using various exposure scenarios representative

of different remediation alternatives.

b. Multiple-species pathways analysis modeling, which simulates contaminant

trophic transfer potential through community food webs.

3.3.7.4.3. Laboratory Studies.

a. Laboratory analysis of biological community samples (e.g., periphyton, benthic

invertebrates, plants), as needed for taxonomy.

b. Chemical analysis of collected tissue samples for COPECs that are known or

suspected of bioaccumulating or biomagnifying.

c. Acute or chronic bioassays using on-site exposure media in order to determine

s, LOAELs, or NOAELs.

d. Acute or chronic bioassays using doses of COPECs suspected of presenting a risk

in order to determine LD

s, LOAEL, or NOAEL doses.

e. Chemical analysis of exposure point abiotic media for the COPECs, specific

species of COPECs, or selected COPECs at detection levels lower than TRVs for the

selected ecological receptors.

f. Chemical analysis of physical media collected from reference areas.

3.3.7.5. Multiple Year Assessments. Some studies are reserved for the largest and

EM 200-1-4

31 Dec 10

3-35

most complex sites requiring multiple year sampling or modeling programs and is only

appropriate where data and an BERA with the highest degree of certainty is required for

the FS/RD-RA. Complex sites are those with complex chemical interactions among

numerous COPECs and exposure matrices, wide-spread contamination or numerous

contamination sources, and sites requiring the examination of potential risk reduction over

time (e.g., Rocky Mountain Arsenal (USEPA 1993e)). This includes biological studies of

long duration and great expense (e.g., multi-year population and community level studies)

or complex exposure modeling.

The effort may require additional abiotic sampling and/or tissue residue sampling to

establish correlation of cause-effect and or verification of a model.

3.3.7.5.1. Field Studies.

To execute these

models, a detailed understanding of the life history and population dynamics of species

studied is required. Complex, mathematical ecosystem models, which describe the

mechanisms of action to address exposure processes and pathways and toxic effects,

are applied. Methods for linking laboratory-derived toxicity data to fish population models

may be applied (Barnthouse et al. 1990). Other models, which address ecosystem

functions (energy and nutrient cycling), may also be developed.

a. Quantitative biota (population/community) sampling extending over multiple

seasons and years to document long-term variability or trends of potentially exposed

biota.

b. Quantitative biota sampling in reference areas during selected seasons to provide

sufficient data for statistical comparisons to the data collected at exposure points.

c. Additional surveys of federal or state-protected species suspected of being

exposed to COPECs.

d. If needed, collection of exposure point media for additional chemical analysis to

support the biological sampling and modeling results.

e. If needed, collection of abiotic media samples from reference areas.

3.3.7.5.2. Ecosystem Modeling Studies. Complex, mathematical ecosystem models

addressing such attributes as energy flow, material cycling, and food web assembly (Hull

and Suter 1993).

3.3.7.5.3. Laboratory Analysis.

a. Laboratory analysis of biological samples (e.g., periphyton, benthic invertebrates,

plants), as needed for taxonomy.

All these models are likely to require high costs and biological monitoring/field validation efforts

involving multi-year and multi-seasonal studies. These population and community models are often data

intensive.

EM 200-1-4

31 Dec 10

3-36

b. If needed, chemical analysis of exposure point media for the COPECs or specific

species of COPECs.

c. If needed, chemical analysis of reference area physical media for the COPECs.

EM 200-1-4

31 Dec 10

4-1

CHAPTER 4

Analysis Phase

4.1. Introduction. This chapter will address the two parts of the Analysis phase of the

ecological risk assessment framework contained in ERAGS (USEPA 1997a). These two

parts are characterization of exposure and characterization of ecological effects.

Additionally, each part will be further divided to address procedures for screening-level

and baseline ERAs.

4.2. Characterization of Exposure – SLERA. The two primary objectives of the

characterization of exposure for the SLERA are 1) identification of the ecological receptor

group(s) and 2) selection of appropriate exposure pathways and exposure point

estimates. Because it is impossible to account for all species in the ecosystems

potentially impacted, a few representative receptor groups are typically chosen for

evaluation in the SLERA. Ecological receptor groups (feeding guilds) with the highest

potential for exposure and/or high sensitivity to the COPECs should be identified.

Development of a preliminary ECSM (see Chapter 3) in conjunction with the preliminary

ecological site characterization can be used to identify these groups.

4.2.1. Evaluation of potential exposure pathways is one of the primary tasks of the

characterization of exposure for the SLERA. Most ecotoxicological information is

currently directed toward the quantification of exposure levels for terrestrial flora (uptake)

and fauna (ingestion) and for direct contact with water by aquatic organisms. While other

routes may be important (e.g., inhalation and dermal absorption by mammals), they are

typically not addressed in a SLERA. The SLERA focuses on those pathways with

maximum expected exposure potential based on professional judgment.

4.2.1.1. In the absence of sound site-specific information, preliminary exposure

estimates are usually based on conservative assumptions such as:

a. Area use factor – 100%;

b. Bioavailability – 100%;

c. Life stage of receptor – most sensitive;

d. Body weight and food ingestion rate – minimum body weight to maximum

ingestion rate; and

e. Dietary composition – 100% of diet consists of the most contaminated dietary

component.

4.2.1.2. The screening assessment should specify which contaminants are of

particular concern from an ecological perspective. This is generally done by comparing

the highest detected chemical concentrations to the screening criteria, or by comparing

estimated intake to a health-based TRV. If enough data are available, the 95% UCL on

the mean may be used (see Chapter 3). The range of chemical concentrations detected,

EM 200-1-4

31 Dec 10

4-2

as well as the number of samples collected, should be reviewed to evaluate which

approach is most appropriate.

4.2.1.3. Additional information on exposure assessment for the SLERA can be

found in the Army BTAG position paper, Technical Document for Ecological Risk

Assessment: A Guide to Screening Level Ecological Risk Assessment (USA BTAG

2005a).

4.3. Characterization of Exposure – BERA. This section discusses the development of

the characterization of exposure portion of a BERA. The purpose of the

characterization of exposure is to estimate the nature, extent, and magnitude of

potential exposure of receptors to COPECs that are present at or migrating from a site,

considering both current and plausible future use of the site. Several components of

the characterization of exposure have previously been evaluated during the SI and

SLERA for the purposes of developing the ECSM and focusing investigative activities.

These components include identification of COPECs, key receptors and food webs,

exposure media, and preliminary exposure pathways and areas (see Chapter 3).

These preliminary characterizations were based upon early and often incomplete

information that now must be clarified in light of the information obtained during the RI

field effort.

4.3.1. Exposure Setting. The objective of describing the exposure setting is to

identify the site physical features that may influence exposure for both current and

future scenarios. Most of this information should have been collected earlier (see

Chapter 3), however, the description of the site setting in the characterization of

exposure for the BERA should involve obtaining more specific, in-depth information

than was obtained previously. While each site will differ in the factors that require

consideration, some of the more common factors are listed below and discussed briefly.

Examples of how the factors may influence exposure also are provided. The

description should be supplemented by data collected during site investigations.

a. Geology. The land type and forms may influence exposure in various ways.

For example, the topography of the area can influence the direction and rate of

movement of chemicals to offsite areas.

b. Hydrology. The possible connection of surface water bodies with groundwater

should be evaluated where there are surface waters or wetlands. The potential

presence of groundwater seeps should also be evaluated. The presence and character

of surface water bodies or wetlands may affect potential exposures in aquatic

ecosystems.

c. Climate. The temperature and precipitation profiles of the area limit the types

of receptors present, feeding habits, frequency of exposure (e.g., frozen surface water

bodies) as well as influence the extent of chemical migration (e.g., surface water runoff

and erosion, infiltration).

d. Meteorology. Wind speed and direction influence the entrainment of soil

particles and the extent of transport and dilution of air contaminants.

EM 200-1-4

31 Dec 10

4-3

e. Vegetation. The nature and extent of vegetation influence the fauna that are

present and their potential for direct exposure and exposure through the food chain.

f. Soil type. The type of soil (e.g., grain size, organic carbon, clay content)

influence soil entrainment, the degree of chemical binding, leaching potential,

bioavailability, and the potential for unique vegetation types to be present. Soil

characteristics also influence erosion and the resultant vegetative communities.

g. Land Use. The types of receptors likely to have contact with site media and

COPECs depends, in part, on current and planned future land use. The appropriate

current and reasonable future land uses should be identified.

4.3.2. Exposure Analysis. Exposure analysis combines the spatial and temporal

distributions of the ecological receptors with those of the COPECs to evaluate

exposure. The exposure analyses focus on the chemical amounts that are bioavailable

and the means by which the ecological receptors are exposed. The focus of the

analyses depends on the ecological receptors being evaluated and the assessment and

measurement endpoints. A brief discussion on pertinent factors for generic exposure

routes is presented below. When performing the characterization of exposure, these

potential exposure routes should each be examined and a decision made regarding the

exposure route and pathway completeness. Consideration of exposure routes and

pathways for aquatic versus terrestrial receptors requires somewhat different

perspectives. Methods for quantifying exposure for these receptors are also quite

different. The approaches for assessing exposure in aquatic and terrestrial receptors

are thus presented separately in the following text.

4.3.2.1. Exposure Pathways Identification. Exposure pathways should be

identified for both current land use and reasonable future land use, which may or may

not be the same. The following factors should be considered when identifying exposure

pathways for current and future scenarios:

a. What is the current and future land use?

Land use at and surrounding the site

is used to identify the way in which the site is used and the types of exposure pathways

that may be present and complete. Besides the current land use, the reasonably

expected future land use should also be assessed and pathways of exposure identified.

b. What is the exposure area? If relevant, specific portions of the site or off-site

areas that may be contacted by potential receptors should be identified. These may be

source areas or secondary and tertiary media impacted by the source areas. The

plausibility of the entire site being contacted or posing a potential exposure hazard

should be examined. It should be noted that some sites are simply too small to support

adequate populations of animals to warrant evaluation of ecological risks. Some states,

such as Massachusetts, do not require any evaluation of ecological threats unless the

site is at least 2 acres in size.

c. In which media are COPECs presently contained? If COPECs are not present

in a medium sampled during the site investigation, and are not anticipated to be in that

EM 200-1-4

31 Dec 10

4-4

medium during the plausible exposure period for current or future receptors, exposure

to the medium does not need to be assessed.

d. For what period of time are the COPECs expected to remain in the medium?

By examining the chemical's likely fate, it should be determined whether depletion

or reduction of the chemical concentration needs to be considered, and whether the

exposure pathway is self-limiting.

e. Into which media are the COPECs anticipated to enter within the exposure

period? For example, accumulation of chemicals into animal and plant species over

time. Is predictive modeling needed?

f. What types of contact with the impacted media are possible? This

determination is based upon uses of the medium and types of contact made with the

medium. In general, direct contact (aquatic systems), direct uptake (plants), ingestion

(animals), inhalation (animals), and dermal contact (animals) are the possible types of

exposure/intake pathways assessed. Inhalation and dermal contact, however, are

typically not assessed in terrestrial ERAs as these routes are not well studied for

wildlife. Most wildlife also have protective features such as fur or feathers which

typically result in dermal contact being a negligible exposure pathway.

4.3.2.2. Exposure Routes for Aquatic Receptors. A complete exposure pathway

typically consists of four elements -- a source and release of COPECs, a transport

medium, an exposure point with receptors, and an exposure (uptake) route. In the

aquatic habitat (fresh water, estuarine, or marine), organisms exposed to COPECs are

principally the aquatic organisms (e.g., algae, plants, invertebrates, fish, marine

mammals) or their terrestrial consumers and predators (e.g., shore birds, waterfowl,

piscivores). Exposure of terrestrial receptors is discussed in Paragraph 4.3.2.8.

4.3.2.2.1. The aquatic ECSM serves a very useful purpose -- it enables the risk

assessor to visualize where and how COPECs may be moving from the source to the

ultimate receptors of concern, through the various release mechanisms, secondary

sources, uptake mechanisms, and primary receptors. The aquatic ECSM shows which

pathways may be significant and what measurement endpoints should be considered.

4.3.2.2.2. From the primary source of COPECs, chemicals move toward the

exposure points via the actions of direct discharge, leaching, infiltration, and erosion.

Leaching and infiltration to groundwater is the most common contaminant route to

aquatic receptors since many chemical releases are from tanks, pipelines, or other

spills to site soils and from there to groundwater. Groundwater itself is only rarely an

exposure medium for aquatic receptors, but it may be a primary pathway to surface

water, where chemical concentrations are rapidly diluted, and to sediment.

Volatilization of organic COPECs and dust generation from the primary source can

occasionally be release mechanisms through the air to water and sediment, but the air

pathway is rarely quantifiable except in cases of emissions from stacks or cooling

towers.

EM 200-1-4

31 Dec 10

4-5

4.3.2.2.3. Once in surface waters, chemicals are affected by a wide variety of

physical and chemical processes that can change their chemical configuration, physical

location, bioavailability, and toxicity within the aquatic environment. Chemicals can be

lost from the water through volatilization. Chemicals in water can move into the bottom

or suspended sediments via sorption or complexation with sediments or through

precipitation and settling, which can be caused by an increase in the pH of the water.

As indicated in the aquatic ECSM, chemicals move between water and sediment, with

the sediments often serving as a source of chemicals that have been sequestered from

past releases. Sediments are critical factors in aquatic ERAs because many COPECs

accumulate to elevated concentrations in sediments, and therefore act as sources of

chemicals to the interstitial (i.e., pore) water and overlying surface waters.

4.3.2.2.4. Aquatic receptors are, by definition, in continuous contact with the water.

They are also in contact with sediments, either bed sediments covering the bottoms of

the lakes, streams, and estuaries or suspended sediments that are in the water column.

Aquatic receptors can be exposed to sediments through incidental ingestion while

feeding or through contact of sediment with permeable membranes. The extent of

exposure to chemicals in sediment varies with several factors, including bioavailability

of COPECs, sediment type, sediment and water movements, organism life stage and

location in the water column, migratory movements, and feeding strategies.

4.3.2.2.5. Aquatic receptors can also be exposed to COPECs by ingesting prey

organisms that have bioaccumulated chemicals, typically organic compounds such as

pesticides or PCBs. Evaluation of the potential for risk through exposure of aquatic

receptors to COPECs is increasingly complex for the three exposure media -- water,

sediment, and prey. Because of this increasing level of complexity in assessing the

potential for exposure and risk, water is the exposure medium often evaluated first, by

screening against established water quality criteria and standards or laboratory

bioassay results. Contaminant concentrations in sediment can be compared to

sediment standards, guidelines, or COPEC sediment levels that are back-calculated

from water criteria using chemical-specific K

values in an equilibrium partitioning

approach. Finally, potential risk from ingesting contaminated prey can be evaluated by

using food ingestion models that consider all three pathways.

4.3.2.3. Exposure Route Modifying Factors for Aquatic Receptors. Numerous

factors modify the extent of exposure in the aquatic environment. Although factors

generally fit into physical, chemical, and biological categories, the factors act in

combination with each other to affect the exposure of aquatic receptors to COPECs,

bioavailability of the COPECs, and the toxicity of the COPECs.

4.3.2.3.1. Physical Factors. Physical factors affect the release mechanisms that

move COPECs from the source along a transport medium to the exposure point;

physical factors also can influence the movements of receptors and their presence at

the exposure point. These physical factors include discharge, leaching, infiltration,

erosion, dilution, settling, and resuspension on the physical media.

EM 200-1-4

31 Dec 10

4-6

4.3.2.3.1.1. An example can serve to illustrate the physical factors that influence

the presence and concentration of COPECs at the exposure point. COPECs in

contaminated soils can move into groundwater through leaching from contaminated

soils. Groundwater then moves toward surface waters at a given rate that when

multiplied by a COPEC concentration in ground-water, results in a loading rate to the

surface water. Groundwater typically moves through the interstices of the sediment

where the COPECs can accumulate in the sediment or can be diluted when mixed with

the surface water. Grain size and shape of the sediment particles affect the tendency

of COPECs to adsorb onto the sediment, thereby reducing their mobility in the aquatic

environment. Throughout, chemical factors such as pH, oxidation-reduction potential

(Eh), and presence of other chemicals interact with the physical factors described and

affect the presence, concentration, and form of the COPECs at the exposure points

(sediment and surface water).

4.3.2.3.1.2. Physical factors can also influence the movement and location of

aquatic receptors, thus affecting their exposure to COPECs. In an interactive scenario

analogous to that described above for physical and chemical factors, physical factors

interact with biological factors that also affect exposure of the receptors. Physical

factors such as current velocities, water temperature, and water salinity can influence

seasonal migratory movements and rates of growth that, in turn, can influence the

location of the receptors relative to COPEC concentrations.

4.3.2.4. Chemical Factors. Chemical factors can affect the chemical and physical

form of the COPEC, the bioavailability, and ultimately, the toxicity to receptors. In fresh

water, pH, Eh, hardness, and the presence of dissolved and particulate organics affect

the form and availability of many metals. The overall effect of these confounding

natural factors on toxicity of metals is reflected in the water effect ratio (WER), which is

based on the relative toxicities of a COPEC when tested in a dilution series using

laboratory water versus the same COPEC tested using upstream natural water for

dilution.

4.3.2.4.1. Some of the same chemical factors influencing exposure of receptors to

COPECs in water also affect exposure to COPECs in sediments. Two other chemical

factors, total organic carbon, and acid volatile sulfide (AVS) strongly affect exposure of

receptors to COPECs in sediments. Increased levels of organic carbon in sediments

tends to bind non-polar organics to the sediment. This effect is reflected in the

chemical-specific K

4.3.2.4.2. AVS affects the binding of metals to sediments by providing additional

binding locations for metals. The metals primarily affected include cadmium, copper,

lead, nickel, and zinc. These metals replace iron in iron sulfide complexes. If the

concentration of AVS exceeds the combined concentration of these five metals as

determined through a procedure referred to as simultaneously extracted metals

(SEM)(i.e., SEM/AVS ratio is greater than 1.0), the mobility of the metals is decreased

due to the abundance of binding locations. If the AVS level is lower than the SEM level

(i.e., SEM/AVS < 1.0), there may be a lack of binding locations, and the five SEM

EM 200-1-4

31 Dec 10

4-7

metals are more available (and potentially toxic) to receptors. The results of the AVS

and SEM analyses should be interpreted on a weight-of-evidence basis because of the

confounding influence of other chemical and physical factors.

4.3.2.5. Biological Factors. Several biological factors affect exposure of aquatic

receptors to COPECs in the water and sediment. Similar factors also affect the

exposure of prey organisms to COPECs that can bioaccumulate in the prey tissues,

thus contributing to the overall exposure of receptors to bioaccumulative COPECs.

Some of the more important biological factors affecting exposure to COPECs are life

stage, feeding strategy, and migratory movements of the receptors.

4.3.2.5.1. In a typical exposure scenario, COPECs are found in sediments and

water but are at higher concentrations in the sediments. Several benthic invertebrate

species (e.g., oysters), have larval stages that are planktonic (floating) and adult life

stages that are sessile (attached to a substrate). If that substrate or the surrounding

sediment has elevated COPEC concentrations, the adult is likely to be exposed to

COPECs, whereas the larval stage is less likely to be exposed since it is not directly

associated with the sediment.

4.3.2.5.2. Feeding strategy can also directly influence exposure to COPECs.

Receptors that feed in or along the sediment are apt to be exposed to COPECs through

ingestion of prey organisms that have accumulated COPECs as well as incidental

ingestion of sediment. If a receptor feeds higher in the water column, it is likely to be

exposed to lower levels of COPECs, as incidental ingestion of sediments would not be

a valid pathway. If a receptor is an upper-level predator (e.g., black drum), it is only apt

to be exposed to bioaccumulative COPECs through ingestion of prey that have

elevated levels of COPECs in their tissues.

4.3.2.5.3. Migratory movements of receptors can directly affect exposure to

COPECs. The effect of migratory movements is readily illustrated through a

comparison of a fish that follows anadromous migratory patterns (i.e., moves from the

ocean through an estuary into fresh water to spawn and then returns to the ocean) to a

resident species of the estuary. If the estuary and its sediments have elevated levels of

COPECs, the resident species is exposed throughout its life, while the anadromous

species is only briefly exposed. In the case of the migratory species, although its year-

round exposure cannot be confirmed, it often is assumed that the species is exposed to

the COPECs only while it is in the vicinity of the contaminated sediment or other

exposure medium.

4.3.2.5.4. The manner in which several of these biological factors may affect the

exposure characteristics of receptors to COPECs provides an emphasis for going

beyond mere listing of species present, which are formulated during the initial site

description and/or reconnaissance. A functional evaluation of how the species present

actually use the habitat is necessary. Uses such as spawning grounds, nursery

grounds, or adult food foraging should be distinguished so that significant biological

factors influencing exposure may be integrated in any evaluation of exposure routes.

EM 200-1-4

31 Dec 10

4-8

4.3.2.6. Exposure Routes for Terrestrial Receptors. Similar to the aquatic ECSM,

the terrestrial ECSM enables the risk assessor to visualize where and how COPECs

may be moving from the source to the ultimate receptors of concern, through the

various release mechanisms, secondary sources, uptake mechanisms, and primary

receptors. The three principal potential exposure routes for terrestrial (animal)

receptors are: dermal absorption, inhalation, and ingestion. Exposure route for plants

include both root uptake and foliar absorption.

4.3.2.6.1. Dermal Contact with Soil, Sediment, Water, and Air. Dermal contact with

soil, sediment, or water is a potentially significant exposure route for soil-dependent

terrestrial animals (e.g., invertebrates and microbes) or animals, which spend

considerable time submerged in surface water (e.g., muskrat, beaver). Wildlife may

receive indirect dermal exposure by brushing against surface-contaminated vegetation.

However, dermal absorption is generally an insignificant intake route for terrestrial

wildlife, as such, receptors are largely protected by their fur, feathers, or scales. Soils

that are covered by pavement are unlikely or impossible to contact, and the assessment

should account for this accordingly. Further discussion of the assessment of dermal

exposure is presented in Paragraph 4.5.5.3.

4.3.2.6.2. Inhalation Exposure to Air. Inhalation exposure by terrestrial receptors

could occur to both vapor phase chemicals and particle phase chemicals. Quantitative

methodologies for evaluating this exposure route in terrestrial fauna are not well

established, but have been developed in order to evaluate wildlife exposure to herbicide

sprays (USDOI 1991). Consideration should be given to the chemical form applied,

degree of chemical absorption, methods for estimating exposure point concentrations,

and toxicity values where there is the potential for this to be a significant pathway.

Further discussion of the assessment of inhalation exposure is presented in Section

4.5.5.2.

4.3.2.6.3. Ingestion of Water. Ingestion of water by terrestrial wildlife should be

examined where there is a significant water source. Analysis of unfiltered surface water

samples best represents chemical concentrations to which a terrestrial receptor may be

exposed. Potential exposure of biota to chemicals in small, temporal, surface water

puddles is typically not evaluated (unless concentrations are extremely toxic) as the

exposure is likely to be insignificant compared to exposure from other pathways.

4.3.2.6.4. Ingestion of Soil or Sediments. Ingestion of soil or sediment should be

considered for all exposure scenarios that provide direct access to soil. Many wildlife

species ingest soil while feeding, but ingestion rates are known for only a few species.

Soil ingestion rates have been measured for certain livestock in order to estimate

pathways for human exposure (USEPA 2005b). Similar estimates of soil ingestion

rates for grazing wildlife may also be used.

4.3.2.6.4.1. Except for earthworms and some other soil invertebrates, most

terrestrial animals do not "eat" dirt, but ingest only a limited amount of soil incidental to

feeding (typically less than 10 percent of food intake). Deliberate ingestion of soil may

occur under some circumstances, such as for sodium (salt licks) or calcium content, or

EM 200-1-4

31 Dec 10

4-9

for grit. Soil intake may also be a result of incidental (direct) ingestion from soil adhered

to the surface of food/prey items or from grazing, preening/cleaning or burrowing

activities. Under certain site conditions, the soil in the gut of earthworms may be an

important exposure medium for animals that eat these organisms (Beyer et al. 1993).

4.3.2.6.4.2. The sandpiper group is generally thought to have the highest rate of

soil/sediment ingestion (7 to 30 percent) due to their diet of mud-dwelling organisms.

Relatively high rates are also reported for wood ducks (11 percent), raccoon (9.4

percent) and woodcock (10.4 percent), which feeds extensively on earthworms, and

Canada goose (8.2 percent) (Beyer et al. 1994). Soil ingestion rates for small rodents

are reported at less than 2 percent (Beyer et al. 1994).

4.3.2.6.5. Ingestion from Diet. Bioaccumulative COPECs tend to increase in

concentration within some organisms relative to their concentration in environmental

media and dietary sources due to sequestration in certain body tissues. It should be

noted that bioaccumulative COPECs can be present at a concentration in

environmental media that is protective for direct exposure, but that can pose indirect

risk to higher trophic level receptors (TNRCC 2001). Bioaccumulation can occur in an

organism any time a chemical is taken up and stored faster than it is eliminated, and it

represents the combined accumulation from diet and direct uptake from abiotic media.

4.3.2.6.5.1. Biomagnification is a special case of bioaccumulation whereby the

concentration of a chemical increases at each successive level in the food chain.

Predator species at the top of the food web are the most vulnerable to chemicals that

biomagnify. In general, long-lived and larger species (that accumulate fat) have a

greater opportunity to accumulate these compounds as well. Also, higher trophic level

species, particularly bird species, may be more sensitive to the COPECs than the

animals on which the birds prey. Persistent chemicals (i.e., half-life greater than 30

days), chemicals with a BCF of greater than 1000, or those chemicals with a log K

value greater than 4.2 tend to bioaccumulate (USEPA 2000a).

4.3.2.6.5.2. The Texas Natural Resources Conservation Commission (TNRCC) has

identified COPECs that are able to pose substantial risk due to bioaccumulation,

utilizing information from the USEPA, Environment Canada, the United Nations

Economic Commission for Europe and the North American Commission for

Environmental Cooperation. The table of bioaccumulative COPECs, below, is taken

from TNRCC (2001).

EM 200-1-4

31 Dec 10

4-10

CASRN

COPEC

Applicable Media

Metals

7440-43-9

Cadmium

Sediment, Soil

7440-47-3

Chromium

Soil

7440-50-8

Copper

Sediment, Soil

7439-92-1

Lead

Soil

7439-97-6

Mercury

Water, Sediment, Soil

744-02-0

Nickel

Sediment, Soil

7782-49-2

Selenium

Water, Sediment, Soil

7440-28-0

Thallium

Water

688-73-3

Tributyltin

Sediment

7440-66-6

Zinc

Sediment, Soil

Organochlorine Pesticides

309-00-2

Aldrin

Sediment, Soil

57-74-9

Chlordane

Sediment, Soil

72-54-8

DDD

Water, Sediment, Soil

72-55-9

DDE

Water, Sediment, Soil

50-29-3

DDT

Water, Sediment, Soil

60-57-1

Dieldrin

Sediment, Soil

72-20-8

Endrin

Sediment, Soil

76-44-8

Heptachlor

Sediment, Soil

1024-57-3

Heptachlor epoxide

Sediment, Soil

8001-35-2

Toxaphene

Sediment, Soil

Other Pesticides and PCBs

2385-85-5

Mirex

Sediment, Soil

3980-114-4

Photomirex

Sediment, Soil

1336-36-3

PCBs

Water, Sediment, Soil

Other Semi-Volatiles

None

Dioxins

Water, Sediment, Soil

None

Furans

Water, Sediment, Soil

118-74-1

Hexachlorobenzene

Water, Sediment, Soil

608-73-1

Hexachlorocyclohexane

Sediment, Soil

29082-74-4

Octachlorostyrene

Water, Sediment, Soil

4.3.2.7. Plant Uptake. The soil-plant system is an open system subject to inputs,

contaminants and fertilizers, and to losses, through plant consumption, leaching,

erosion and volatilization. Factors affecting the contaminant amounts absorbed by a

plant are those controlling (Alloway 1990):

a. concentration and speciation of the contaminant in the soil solution;

b. movement of the contaminant from the bulk soil to the root surface;

c. transport of the contaminant from the root surface into the root; and

EM 200-1-4

31 Dec 10

4-11

d. translocation from the root to the shoot.

4.3.2.7.1. Plant uptake is dependent on both the total quantity of the contaminant in

soil as well as the root mass present. Terrestrial plant uptake of contaminated water

can be a potentially significant pathway if the plant is a wetland species or a

phreatophyte (plants that depend on groundwater for their moisture). The uptake route

for water is generally insignificant for xerophytic (plants structurally adapted for life and

growth with a limited water supply) and mesophytic (plants that grow under medium

conditions of moisture) plants which have more shallow root systems and depend on

surface water from rainfall.

4.3.2.7.2. In addition to the root absorption, plants can absorb contaminants

through their foliage. Foliar absorption of contaminants (in the form of solutes) depends

on the plant species, its nutritional status, the thickness of its cuticle, the age of the leaf,

the presence of stomata guard cells, the humidity at the leaf surface, and the nature of

the solutes (Alloway 1990). The uptake route from air to terrestrial plants can be a

potentially significant pathway for vapor phase and particulate phase COPECs. While

chemical concentrations found in the air pathway generally pose only a minimal risk to

animal species, lichens, in particular, and trees can be especially sensitive to airborne

contamination. In ERAs conducted near forested areas, air may be an important

environmental transport medium for certain plant groups.

4.3.2.8. Exposure Route Modifying Factors for Terrestrial Receptors. Numerous

factors influence the spatial distribution and abundance of a population of animals

relative to the spatial extent of contamination. Exposure modifying factors such as

home range, mobility, and life-cycle attributes (breeding seasons, longevity) should be

evaluated in the characterization of exposure. Normalizing factors (e.g., body weight,

growth rate) for the various receptors are also to be considered during exposure

quantitation.

4.3.2.8.1. Area Use. Home ranges and feeding territories should be considered as

they may greatly influence potential exposure. The size and spatial attributes of a

home range often are determined by foraging activities, but also might depend on the

location of specific resources such as dens or nest sites. Home ranges depend on

habitat quality (e.g., carrying capacity), with home range sizes generally increasing as

habitat quality decreases to a condition beyond which the habitat does not sustain even

sparse populations. Home ranges can also vary by sex, season, and life stage.

Population density (the number of organisms per unit area) also influences potential

exposure.

4.3.2.8.1.1. The mobility of a receptor is usually expressed in terms of the average

foraging range of the key receptor (or similar species) under consideration. Mobile

receptors typically include the larger vertebrates and grazing species (deer, elk,

antelope), predators (fox, coyote), migratory birds (robin) and predatory birds (hawk,

eagle, falcon). The foraging areas of these transitory species are likely to be several

square miles. Smaller mammals and birds constitute a category of mobile receptors

whose foraging areas range from a fraction of an acre to several acres. Plants, soil

EM 200-1-4

31 Dec 10

4-12

organisms, and most flightless invertebrates can be considered to be stationary due to

the small area within which they live their lives.

4.3.2.8.1.2. In each case, to quantify chemical intake for the key receptor, an area

use factor should be applied to account for the foraging range of the key receptor, as

compared to the areal extent of the contaminated area. The area use factor is defined

as the ratio of the area of contamination (or the site area under investigation) to the

home range of the receptor (or feeding/foraging range).

4.3.2.8.2. Exposure Frequency. Exposure frequency is another type of modifying

factor that can be used to adjust exposure and chemical intake for a key receptor.

Resident species, rather than migratory species, should be evaluated first (when they

are present), due to the longer exposure duration potential of the resident species.

Migratory species should be evaluated where there is the potential for acute toxic

effects from infrequent exposure or where exposure pathways present a greater

exposure potential. Magnitude and frequency of exposure should be taken into

consideration where the assessment endpoint and toxic effect are based on chronic

exposure duration in the test organism.

4.3.2.8.3. Seasonal Activity Patterns. Many seasonal or life-cycle attributes affect

an animal's activity and foraging patterns in time and space and their exposure

potential. For example, many species of mammals, reptiles, and amphibians hibernate

or spend a dormant period in a burrow or den during the winter months. Longevity and

mortality rates also influence exposure potential and are important in determining

potential for chronic exposures.

4.3.2.8.3.1. Seasonal variability may also affect the interpretation of ecological data

and should be considered in the design of any sampling plan. Data obtained during

any short period could be accurate, but only for that period. For example, pinyon mice

apparently suffer substantial winter mortality (Morrison 1988). Trapping only in fall or

spring would falsely indicate a relatively high or low population size, respectively. A full

year of sampling is generally required to adequately characterize an ecological

population.

4.3.2.8.3.2. Some vertebrate population cycles, however, can take much longer;

e.g., a 23-fold difference between peaks and low numbers in snowshoe hares was

described in one 15-year study (Keith 1983), and it took 12 years for a relationship

between conifer seed crop and red squirrel abundance to be repeated (Halvorson

1984).

4.3.2.8.4. Dietary Composition. Dietary composition varies seasonally and by age,

size, reproductive status, and habitat. Dietary composition is an important

consideration for higher trophic level organisms indirectly exposed to chemicals that

bioaccumulate or biomagnify.

4.3.2.8.5. Habitat Preferences. Many wildlife species have habitat preferences that

may increase or decrease their potential exposure to contaminants. Woodcocks, for

EM 200-1-4

31 Dec 10

4-13

example, will remain longer feeding in fields with tall cover than in those with short

vegetation (Hull and Suter 1993). Robins, on the other hand, prefer fields or lawns

maintained by regular mowing.

4.3.2.8.6. Foraging Style. Animals with different foraging styles may also have

different morphologies and activity patterns that ultimately influence exposure to

contaminants. Piscivorous avian species, for example, can be classified into three

general types of foraging styles: raptorial predators (bald eagle), diving and swimming

predators (common merganser), and wading predators (great-blue heron).

4.3.3. Exposure Profiles. Using information obtained from the exposure analysis,

the exposure profile quantifies the magnitude and spatial and temporal patterns of

exposure. The exposure profiles developed for the ecological receptors and COPECs

serve as input to the risk characterization.

4.3.3.1. Quantitation of Exposure. For wildlife, chemical intakes are estimated for

exposures occurring from complete exposure pathways for each receptor group. The

exposures are quantified with respect to the magnitude, frequency and duration of

exposure, to derive an estimate of chemical intake. Chemical intake by wildlife is

estimated by combining two general components: the chemical concentration

component and the intake/exposure factors component. In the following subsections

the estimation of the exposure point concentrations, discussion of the selection of

intake and exposure factors, and the specific methods of combining them

mathematically are presented.

4.3.3.2. Determining Exposure Concentrations (Aquatic and Terrestrial Scenarios).

Exposure concentrations represent the chemical concentrations in environmental media

that the receptor will contact. Exposure concentrations may be derived from either data

obtained from sampling or from a combination of sample data and fate and transport

modeling, both of which are described below.

4.3.3.2.1. For current (and perhaps some future) exposure scenarios where current

site data are anticipated to be reasonably reflective of exposure concentrations over the

exposure period, the exposure point concentration can be directly derived from site

data. For future (and perhaps some current) exposure scenarios, where current site

conditions are not anticipated to be reasonably reflective of exposure concentrations

over the exposure period, some form of fate and transport modeling or degradation

calculations can be applied. However, these too will be based upon current site

conditions as a starting point. The available data need to be examined critically to

select the most appropriate data in each medium to describe potential exposure.

These data sets can vary depending on the receptor-specific exposure factors. For

example, soil data for soil-dependent organisms (earthworms) and burrowing mammals

would include samples from greater depths (up to 5’ below ground surface) than direct

soil exposure for large herbivores (surface soils).

EM 200-1-4

31 Dec 10

4-14

4.3.3.2.2. Since the exposure point concentration used in the assessment is a value

that represents the most likely concentration to which receptors may be exposed, a

value that reflects the central tendency of the data is appropriate to use. In order to

account for uncertainties in the ability of the measured data to reflect actual site

conditions, the concentration relating to the 95% UCL of the arithmetic mean is usually

used as the exposure point concentration (USEPA 2002a). In cases where the 95%

UCL concentration exceeds the maximum detected value (which can occur in small

data sets or data sets with a large variance), the maximum value has been historically

used

4.3.3.2.3. EPA has worked with its contractor, Lockheed Martin to develop a

software package, ProUCL, to perform many of the calculations. The most recent

version of this software, the ProUCL Version 4.0 Technical Guide and the ProUCL

Version 4.0 User Guide are available at:

. However, the exceedance of the UCL by the maximum value is frequently

indicative of an inadequate sampling design or a problem with the calculation of the

UCL. The maximum value is not a representative estimate of the mean (it tends to over

estimate the mean for large data sets and can under estimate the mean for small data

sets). Prior to using the maximum detected value, the nature of the data should be re-

examined; additional sampling may be required.

http://www.epa.gov/esd/tsc/TSC_form.htm.

The software uses a number of different statistical methods to calculate a set of UCLs

and subsequently recommends the “best” (i.e., most representative) UCL. The guides

contain in-depth discussions and recommendations for UCL calculations that are

beyond the scope of this guidance. However, several salient issues are discussed.

4.3.3.2.4. Often in data sets, a number of data points for a given chemical in a

given medium will be reported as non-detect or “less than values”. These results are

often referred to as “censored data.” The numerical limit to which a non-detect is

reported is some times called the “censoring limit.” Common errors in reporting and

handling these data can occur and include: (1) omission of the censoring limit, (2)

failure to define the censoring limit, or (3) use of an inappropriate high or low censoring

limit. For example, chemical results are often censored to the method detection limit

(as defined in 40 CFR Part 136), a value that is typically too low to minimize false

negatives. The appropriateness of the censoring limit should be reviewed by a qualified

chemist as the censoring limits are part of the data set that will be used for UCL

calculations. The most recent version of ProUCL requires separate entries for the

detected results and the censoring limits for the non-detects.

4.3.3.2.5. It should be noted that, historically, some multiple of the censoring limit

(usually one half) was substituted for each non-detect. This method is commonly

referred to as the “substitution method.” However, the substitution method can distort

data sets (e.g., producing false positives or false negatives) and is now considered

obsolete; as stated in the ProUCL User Guide, “…for data sets with NDs, the DL/2

Reasons for the 95% UCL value exceeding the maximum values are numerous. Such a circumstance

may be indicative of incomplete site characterization. This circumstance may also reflect high variance due

to biased, purposive sampling rather than random sampling.

EM 200-1-4

31 Dec 10

4-15

[one half the censoring limit] substitution method has been incorporated in ProUCL 4.0

only for historical reasons…It is well known that the DL/2 method…does not perform

well…”

4.3.3.2.6. In certain situations, an unusually high or elevated censoring limit may be

assigned to a non-detected result because of matrix interferences, high concentrations

of target chemicals in the sample (e.g., requiring the samples to be diluted), presence

of blank contamination or other factors. These results should be evaluated to

determine whether or not they are appropriate for inclusion in the data set for the UCL

calculation (e.g., it may be desirable to calculate the UCLs with and without these

results to make this determination). However, the most recent version of ProUCL uses

methods to treat non-detects that are relatively robust relative to the substitution

method (which produces UCLs that are strongly dependent on the censoring limits).

4.3.3.2.7. Sample size influences the magnitude of the statistical confidence of the

mean, as demonstrated by high 95% UCL concentrations for small sample sets. The

reliability coefficients (the "H" or "t" value used in calculating the UCL concentration,

obtained from statistical tables) are a function of the number of samples, and increase

with a decreasing number of samples. The overall effect, then, of a small sample size

upon statistical confidence is to increase the UCL concentration. In data sets in which

minimum requirements have been set prior to sampling, the risk assessor should

ensure that an adequate number of samples have been collected to minimize this

problem. UCLs should be calculated using at least 10 – 15 samples (though a sample

size of at least 20 – 30 would be preferable for the statistical calculations and is

recommended).

4.3.3.2.8. Exposure point concentrations are also sometimes derived from a

combination of measured data and the application of environmental fate and transport

modeling. For the most part, measured data points are preferred over modeled data;

where data are modeled, some level of validation and ground-truthing is required

(exceptions include ERAs for proposed incinerator emissions/deposition). Common

instances in which modeling may be used to predict exposure point concentrations

include:

a. When the potential exposure point is at a location other than those for which

monitoring data are available (e.g., in off-site areas or locations in-between those which

have been described);

b. When the potential exposure is anticipated to occur in the future (e.g., proposed

incinerator emissions);

c. When the chemical concentrations are anticipated to change with time;

d. When the potential exposure is in a medium other than those sampled (e.g.,

exposure to air impacted by contaminated soil, when only soil was analyzed);

Additional references appropriate for evaluation of exposure point concentrations are Shultz and Griffin

(1999) and Singh et al (1997).

EM 200-1-4

31 Dec 10

4-16

e. When the potential exposure point concentration is anticipated to increase with

time (as with bioaccumulation into animal or plant species); and

f. When the bioavailable portion of the chemical concentrations are anticipated to

change with time (e.g., seasonal AVS fluctuations, fluctuations between fresh and

saline water either with migration downstream or tidal influence).

4.3.3.2.9. Many fate and transport models are available with which to predict

exposure point concentrations from existing site data. These models are presented in

other references and on various web sites, including the following:

a. The Adaptive Risk Assessment Modeling System (ARAMS™). ARAMS is not a

model per se, but is a computer-based, modeling- and database-driven analysis system

for estimating human and ecological health impacts and risks associated with military

relevant compounds and other potential contaminants of concern. ARAMS is based on

a widely accepted risk paradigm that integrates exposure and effects assessments to

characterize risk. http://el.erdc.usace.army.mil/arams/

b. Guidelines for Exposure Assessment (USEPA 1992l)

c. Exposure Assessment Tools and Models Web Site,

www.epa.gov/oppt/exposure/index.htm

d. The Center for Exposure Assessment Modeling Web Site,

www.epa.gov/ceampubl/

e. The Center for Subsurface Modeling Support Web Site,

www.epa.gov/ada/csmos.html

f. The USACE Groundwater

http://chl.erdc.usace.army.mil/chl.aspx?p=s&a=Articles;585

Modeling System Web Site,

g. The USACE Surface Water Modeling System Web Site,

http://chl.erdc.usace.army.mil/CHL.aspx?p=s&a=Software;4

h. The AFCEE Subsurface Models Web Site,

http://www.afcee.brooks.af.mil/products/techtrans/models.asp (This includes a link to a

model selection chart, as well as links to specific models).

i. Superfund Exposure Assessment Manual (USEPA 1988a),

j. Air/Superfund National Technical Guidance Study Series (Volumes I - V)

(USEPA 1989e,f; 1992f, 1993c; 1995e),

k. A Workbook of Screening Techniques for Assessing Impacts of Toxic Air

Pollutants (USEPA 1988b),

Although groundwater modeling is typically not required for ecological receptors, there are times when

groundwater discharges to the surface and modeling may be required or deemed beneficial.

EM 200-1-4

31 Dec 10

4-17

l. Selection Criteria for Mathematical Models Used in Exposure Assessments:

Groundwater Models (USEPA 1988c),

m. Selection Criteria for Mathematical Models Used in Exposure Assessments:

Surface Water Models (USEPA 1987),

n. Rapid Assessment of Exposure to Particulate Emissions from Surface

Contamination Sites (USEPA 1985),

o. Methodology for Assessing Health Risks Associated with Multiple Pathways of

Exposure to Combustor Emissions (USEPA 1998b), and

p. Assessment and Control of Bioconcentratable Contaminants in Surface Water

(USEPA 1991b).

4.3.3.2.10. The type of model and level of effort to be expended in estimating

exposure point concentrations with models should be commensurate with the type,

amount, and quality of data available. In general, it is best to begin with a model that

employs simplified assumptions (i.e., a "screening level" approach) and determine

whether unacceptable ecological risks are posed by the exposure point concentration

estimated by this approach. If so, a more complex model that applies less conservative

assumptions can be used.

The validity of the estimation provided by the model will strongly depend on the

variables that are input to the models. Efforts should be taken to ensure the use of

input variables that best reflect site conditions and that are not overly conservative.

4.3.3.2.11. Initial abiotic sampling designs are often not established with sampling

for the selected key ecological receptors in mind. Often, biased sampling designs are

selected in order to best characterize potential hot-spot conditions and the nature and

extent of contamination. Calculation of a 95% UCL or averaging of these point

concentrations tends to result in an overestimation of the exposure concentration (and

risk) for larger mobile animals (deer, antelope) that do not forage onsite or at any

particular spot for extended periods of time. Where the receptor's home range is

greater than the contaminated area, area use and exposure frequency factors can be

used to modify the area-wide intake concentration. Where the receptor's home range

lies within the contaminated area, alternate methods of removing the bias from the

area-wide exposure concentration (e.g., weighted average, Theissen polygons) data set

can be used, but may result in an over or underestimate of exposure.

4.3.3.2.12. Probabilistic methods can also be used for developing more appropriate

exposure concentrations, where factors such as area use need to be considered

(USEPA 1997b). For mobile receptors such as fish, large herbivores, and predators,

determination of dietary exposure concentrations should be "area" (i.e., feeding range)

based rather than "point" (i.e., fixed location) based. Using probabilistic uncertainty

analyses methods to create models that simulate random walks, probable exposure

conditions for mobile receptors can be estimated under different time scenarios (daily,

EM 200-1-4

31 Dec 10

4-18

weekly, monthly, yearly). Two such models, developed for USACHPPM, are Terrestrial

Wildlife Exposure Model (TWEM) and Spatially Explicit Exposure Module (SEEM),

discussed below.

4.3.3.2.13. TWEM is a software tool that allows scientists to estimate exposure of

terrestrial wildlife receptors to organic and inorganic chemicals in soil, water, and biota.

These exposure estimates may then be used in screening-level ecological risk

evaluations. SEEM is a tool by which to complete a wildlife exposure assessment

guided by wildlife habitat preferences and foraging behaviors. SEEM incorporates

habitat suitability indices and two foraging strategies to capture exposures that are

more realistic than assuming uniform exposure across a heterogeneous site (i.e. site-

wide averages). The modeling approach and level of complexity are user defined. As a

result, SEEM can be employed to assess a variety of wildlife exposure problems.

These models can be downloaded from the USACHPPM web site at: http://chppm-

www.apgea.army.mil/tox/HERP.aspx

4.3.3.3. Calculating Intake for Terrestrial Wildlife. The following discussion of

terrestrial wildlife intake focuses on the oral ingestion route only. Oral intake (ingestion)

of three environmental media (food, water, soils/sediment) are the principal routes

evaluated in a terrestrial ERA, as they typically represent the most significant exposure

pathways. Quantitative data and methodologies by which to calculate inhalation and

dermal contact rates for various terrestrial wildlife are generally lacking; limited

guidance on these intake routes are provided by USEPA (1998b, 1993d), and USDOI

(1991). For each receptor, the following four exposure factors are considered in the

calculation:

a. Food Intake (FI) - These rates can vary by age, size and sex and by seasonal

changes (ambient temperature, activity levels, reproductive activities, and the type of

diet consumed). Food ingestion rates are available in the published literature for a

limited number of wildlife species. Methods for estimating food ingestion rates are

provided below. Food ingestion rates are typically expressed on a wet-weight basis.

Where results from wildlife laboratory studies are expressed on a dry weight basis, this

difference may be ignored as the moisture content of most laboratory studies is typically

less than 10 percent water (Beyer and Stafford 1993).

b. Dietary composition (DC) - Dietary composition varies seasonally and by age,

size, reproductive status, and habitat. Dietary composition is typically expressed as

percentage of total intake on a wet-weight basis.

c. Water Intake (WI) - Water consumption rates depend on body weight,

physiological adaptations, diet, temperature, and activity levels. Some species (e.g.,

deer mouse) can meet most of their daily water requirement with only the water

contained in their diet. Water ingestion rates can be estimated using allometric

equations below.

d. Soil/Sediment Intake - Soil or sediment intake is usually expressed as a percent

of dietary intake. Data quantifying soil/sediment intake are limited; values for selected

EM 200-1-4

31 Dec 10

4-19

wildlife species are presented in the Wildlife Exposure Factors Handbook (USEPA

1993d). As noted earlier, soil/sediment intake rates of up to 30 percent of diet are

reported for some wildlife.

4.3.3.3.1. Intake Equations. Exposure or chemical intake by terrestrial wildlife is

reported as "average daily dose" on a body weight basis, i.e., milligrams chemical per

kilogram body weight per day (mg/kg-bw/d). It is fundamental that exposure, chemical

intake, and toxicity benchmark determinations be adjusted to account for body weight

and dietary intake of the organism, to account for the differences in food intake relative

to body weight of the various organisms being compared. Exposure evaluations (and

toxicity benchmark selection) based on a comparison of dietary chemical

concentrations (i.e., milligrams chemical per kilogram food, mg/kg) amongst wildlife

receptors (e.g., deer and rabbits) are sometimes mistakenly attempted in an ERA as a

means to "simplify" the quantitation process.

4.3.3.3.1.1. The following equations for chemical intake were taken from the Wildlife

Exposure Factors Handbook (USEPA 1993d). Additional equations for pathways (e.g.,

inhalation) and receptors (e.g., amphibians) not addressed here can also be obtained

there.

Daily Intake (mg/kg-bw/d) = [(C x FI) + (C x WI)] x EMF

where:

C = Chemical concentration in food or water (i.e., mg/kg, mg/L, ppm)

FI = Food Intake rate (kg-food/day)

WI = Water Intake rate (L-water/day)

BW = Body weight of receptor (kg)

EMF= Exposure modifying factors (default value is 1.0) (unitless)

Birds For birds, Nagy (1987) developed the following equations for calculating

food ingestion (FI) rates (in grams dry matter per day):

FI (g/day) = 0.648 Wt

0.651

(g), or all birds

FI (kg/day) = 0.0582 Wt

0.651

(kg) all birds

FI (g/day) = 0.398 Wt

0.850

(g) passerines

FI (g/day) = 0.301 Wt

0.751

(g) non-passerines

FI (g/day) = 0.495 Wt

0.704

(g) seabirds

where Wt is the body weight (wet) of the animal in grams (g) or kilograms (kg) as

indicated.

Mammals For placental mammals, Nagy (1987) developed the following equations

for calculating FI rates (in grams dry matter per day):

EM 200-1-4

31 Dec 10

4-20

FI (g/day) = 0.235 Wt

0.822

(g), or all mammals

FI (g/day) = 0.0687 Wt

0.822

(kg) all mammals

FI (g/day) = 0.621 Wt

0.564

(g) rodents

FI (g/day) = 0.577 Wt

0.727

(g) herbivores

EPA (1988d) also provides the following equations for this calculation:

FI (kg/day) = 0.056 (Wt)

0.6611

(kg) laboratory mammals

FI (kg/day) = 0.054 (Wt)

0.9451

(kg) moist diet

FI (kg/day) = 0.049 (Wt)

0.6087

(kg) dry diet

WATER INTAKE RATES

Birds Calder and Braun (1983) developed the following allometric equation for

drinking water ingestion (WI) for birds:

WI (L/day) = 0.059 Wt

0.67

(kg) all birds

To estimate daily drinking water intake as a proportion of an animal's body weight

(e.g., as g/g-day), the WI rate estimated above is divided by the animal's body weight

in kg:

WI (g/g-day) = WI (kg/kg-day), or

= WI (L/day)/Wt (kg)

Mammals Calder and Braun (1983) developed the following allometric equation for

drinking water ingestion (WI) for mammals:

WI (L/day) = 0.099 Wt

0.90

(kg) all mammals

where Wt is the average body weight in kilograms (kg). Additional sources of water

not accounted for in this equation (i.e., metabolic water and water contained in food)

help to balance the animal's daily water losses.

EPA (1988d) also provides the following equations for this calculation:

WI (L/day) = 0.10 (Wt)

0.7377

(kg) laboratory mammals

WI (L/day) = 0.009 (Wt)

1.2044

(kg) mammals, moist diet

WI (L/day) = 0.093 (Wt)

0.7584

(kg) mammals, dry diet

To normalize drinking water intake to body weight (e.g., as g/g-day), the WI rate

estimated above is divided by the animal's body weight in kg:

NWI (g/g-day) = WI (kg/kg-day), or

= WI (L/day)/Wt (kg)

4.3.3.3.1.2. Selection of appropriate intake and exposure modifying factors is a

critical component of the assessment, for these values largely determine the overall risk

estimates. The Wildlife Exposure Factors Handbook (USEPA 1993d) presents

exposure profiles for selected species of birds, mammals, and reptiles and amphibians.

Each species profile provides a series of tables presenting values for normalizing (body

EM 200-1-4

31 Dec 10

4-21

weight) and contact (intake) rate factors, exposure modifying factors (home range),

dietary composition, population dynamics, and seasonal activity patterns. Additional

information on wildlife exposure factors can be found in the published literature

including ORNL's (1998) screening benchmark reports, the USACHPPM life history

database (USACHPPM 2004) and food requirements of wild animals: predictive

equations for free-living mammals, reptiles, and birds (Nagy 2001). In an ERA, all

exposure and intake factors applied to the assessment should be identified in tabular

form, with the source of the value identified and a rationale for the use of the value

presented.

4.3.3.3.1.3. If C and FI vary over time, they may be averaged over the exposure

duration. However, it is not always appropriate to average intake over the entire

exposure duration: For example, a given quantity of a chemical might acutely poison

an animal if ingested in a single event, but if that amount is averaged over a longer

period, effects might not be expected at all. Similarly, developmental effects occur only

during specific period of gestation or development. C, FI, and BW should be selected

so as to be comparable to the specific TRV that is used.

4.3.3.3.1.4. Wildlife can be exposed to contaminants in one or more components of

their diet and different components can be contaminated at different levels. For

example, the diet of the deer mouse, an omnivorous key receptor commonly assessed

in ERAs, primarily consists of invertebrates and terrestrial plants. The daily intake for

the deer mouse is thus expressed as [(chemical concentration in invertebrates x %

ingested) + (chemical concentrations in terrestrial plants x % ingested) x daily food

intake] / deer mouse body weight.

4.3.3.3.1.5. In order to describe a range of potential exposures presented by a site,

the ERA may assess more than one potential exposure scenario. Use of a single

expression of potential ecological risk does not provide information on the possible

range of ecological risks, and may not allow the risk manager to evaluate the

"reasonableness" of the estimate. Current risk assessment guidance for human health

suggests the strategy for determining the exposure point concentration for soils should

depend on spatial contaminant distribution. If a contaminant is widely distributed

throughout the site, the exposure point concentration should be based on the 95% UCL

of the arithmetic average for all site samples, including non-detects. However, if the

contamination is unevenly distributed, i.e., "hotspot" areas exist, these areas should be

evaluated by determining exposure concentrations in these areas. A percentage of

time that the receptor spends on the site in these "hotspot" areas should be factored

into the intake equation. Presentation of these and other scenarios (e.g., central

tendency) provide information about the range of potential risks to the ecological

receptors.

4.3.3.3.2. Intake Variables. To develop a "high end" assessment, USEPA

recommends identifying the most sensitive parameters and using maximum or near

maximum values for one or a few of these variables, leaving other variables at their

mean values. Adopting maximum values for all intake and exposure parameters will

EM 200-1-4

31 Dec 10

4-22

virtually always result in a risk estimate that is above that experienced by the most

exposed receptor and is, therefore, inappropriate. According to USEPA (1992e) human

health guidance, the chemical concentration relating to the 95% upper confidence limit of

the mean is applied as the exposure point concentration term for both the average and

the reasonable maximum exposure (RME) scenarios. Although an upper bound value,

this concentration is descriptive of the mean, and accounts for the uncertainty associated

with measurements of the "true" mean.

4.3.3.3.2.1. The average or central tendency exposure (CTE) is derived by applying

average values for all intake and exposure (e.g., area use) parameters. Although

description of an average exposure is not particularly useful when exposure varies

greatly across all potentially exposed populations, it can provide information on the

extent of impact of the exposure parameters that were maximized in the high end

exposure. Use of a median value for exposure parameters, such as a geometric mean

rather than arithmetic mean, is more meaningful since it represents a midpoint value

(i.e., half the population above and half below).

4.3.3.3.2.2. Contaminants may enter terrestrial food chains directly from

soil/sediment, water, or air or indirectly through the consumption of plants (producers)

or animal prey (consumers). The following sections discuss means for determining

chemical concentrations in plants and prey.

4.3.3.3.3. Estimating Chemical Concentrations in Plants. The three principal

mechanisms by which contaminants can bioaccumulate in plants include: uptake by

roots, direct deposition on exposed plant tissues, and air-to-plant transfer of vapor-

phase contaminants. The relative importance of each pathway to the wildlife consumer

depends on the specific plant, the contaminant, site-specific physicochemical

conditions, and the preference of the wildlife receptor for the particular plant.

4.3.3.3.3.1. The plant-soil bioaccumulation factor (BAF

plant

) or transfer coefficient is

a measure of a contaminant's ability to accumulate in plant tissue and is defined as the

chemical concentration in the plant (dry weight) divided by the chemical concentration

in soil (dry weight). Bioaccumulation factors may be derived differently for inorganic

and organic chemicals, but they are generally dependent on the bioavailability of the

chemical in the soil or soil solution. Information and data on chemical transfer from

soils, particularly sludge-amended soils, to a variety of crop species are available in the

published literature (USEPA 1983, USDA 1983, USDOE 1984).

4.3.3.3.3.2. A number of models are also available for determining plant uptake of

contaminants from soil (Kabata-Pendias and Pendias 1984, Briggs et al. 1982, Topp et

al. 1986, and attachment 4-1 to the Eco-SSLs at:

http://www.epa.gov/ecotox/ecossl/SOPs.htm). Root uptake of numerous contaminants,

however, is inefficient and much of the contaminant concentrations found in plants

results from volatilization and leaf uptake (Suter 1993). Some methods for calculating

chemical concentrations in plant tissue due to root uptake and air to plant transfer are

published by USEPA (1998b). Other methods are available in the published literature.

Quantitative structure activity relationship (QSAR) models for determining combined

EM 200-1-4

31 Dec 10

4-23

root and leaf uptake of organic chemicals in soils are presented by Topp et al. (1986)

and Travis and Arms (1988).

4.3.3.3.4. Estimating Chemical Concentrations in Animal Prey. The animal prey

that higher trophic level predators usually consume as food, take up contaminants from

the food chain by ingesting soil-dependent organisms (plants, soil invertebrates), lower

trophic level consumers, or soil and water directly. Methods for determining BAFs or

biotransfer factors to livestock tissue are available for a variety of chemicals in plants

such as grain (corn, oats, wheat, etc.), forage (pasture grass, hay) and silage (USEPA

1998b). Similar methods for wildlife tissue are generally not available and thus the

livestock factors are sometimes used.

4.3.3.3.4.1. Models for determining the uptake and transfer of chemicals through

various food chains are becoming more numerous in the literature (Winter and Streit

1992, Fordham and Reagan 1991). BAFs can oftentimes be estimated for a receptor of

interest based on food chain data presented in the published literature or in studies

conducted at Superfund sites where tissue sampling was performed.

4.3.3.3.4.2. Studies on the accumulation of elements by earthworms, as well as

direct toxic threshold levels, are becoming more abundant due to the close association

between soil contamination and earthworms and the wide variety of earthworm

predators (Beyer 1990, Beyer and Stafford 1993). Several authors have published

models for determining the uptake of organic chemicals by earthworms (Wheatly and

Hardman 1968, Van Gestel and Ma 1988, Connell 1989).

4.3.3.3.5. Bioavailability. The intake equations used in ERAs typically do not

contain a factor to account for bioavailability or bioassimilation and therefore may

predict an intake higher than one that would occur in actual circumstances. By not

including a factor to consider bioavailability, it is assumed that 100% of the chemical

detected in the medium is bioavailable (when combined with toxicity values, the risk

associated with the absorption of the chemical in the animal study is derived).

Modifications may sometimes be made to these intake equations to account for this

factor, if the appropriate information is available.

4.3.3.3.5.1. Bioavailability refers to the ability of a chemical to be "available" in the

body to interact and have an effect. There are many aspects to bioavailability;

however, the type most of concern to ERAs is the ability of the chemical to be absorbed

into the body. Although the medium on which the chemical is contained may be

contacted, the chemical may not be absorbed for a number of reasons, including the

chemical form, competition with other factors (e.g., food in the stomach), damage of the

organ (e.g., stomach, lung), effect of the medium in which the chemical is contained,

and others. While many of these cannot be reliably addressed in an ERA, chemical

form and effect of the medium can be addressed.

4.3.3.3.5.2. The form of the chemical can affect the degree of absorption into a

body. This factor is most important for chemicals that form compounds (such as metals

and cyanide) and chemicals that can exist in different valence states (again, some

EM 200-1-4

31 Dec 10

4-24

metals). For example, soluble compounds of metals (e.g., barium sulfate) are readily

absorbed through the stomach whereas insoluble forms (e.g., barium carbonate) are

minimally absorbed. Usually, when environmental media are analyzed, chemicals are

reported as an isolated entity (e.g., barium), and no information is provided on the form

that existed in the medium. However, if the form of the chemical used at the site is

known, and information on the absorption of that chemical form is available, the intake

equation can be modified to account for a lesser absorption. Defensible information

should be available to make this modification.

4.3.3.3.5.3. The medium in which the chemical is contained also can affect the

degree of bioavailability. This is most pronounced in media that demonstrate an ability

to bind chemicals (such as soil and sediments). When ingested by wildlife, a

competition occurs between retention of the chemical on the medium and absorption of

the chemical into the body. Therefore, some of the chemical may be excreted from the

body without having been absorbed and some may have been absorbed and available

to exert an effect. Many factors can influence the degree to which the medium will bind

the chemical, most of which cannot be reliably predicted (for example, nature of the

medium [organic carbon or clay content, particle size], other chemicals being absorbed,

pH, organ condition, etc.). In some instances, information may be available on the

degree to which a particular medium affects specific absorption routes. If the

information justifies modifying the intake equations, such a modification may be made

(Drexler et al 2003, Henningsen 2003, NAVFAC 2000, Ruby 2003).

4.3.3.3.5.4. In most assessments, it is generally assumed that environmental

conditions are reasonably static and chemical concentrations remain constant over

time, often for as long as 30 years. Such assumptions may be unreasonable.

Chemical concentrations are usually reduced over time by degradation, migration,

dilution, volatilization, or other removal processes. If these processes are known and

can be quantified, a concentration that decreases over time can be derived for

assessing intakes. If no allowances are made to decrease concentrations over time,

risks will most likely be overestimated.

4.3.3.4. Exposure Characterization Summary. At the conclusion of the

characterization of exposure, the estimated chemical intakes for each exposed receptor

group under each exposure pathway and scenario should be presented in tabular form.

This presentation should include an identification of all pertinent factors (basis of

exposure point concentration, use of models, if applicable, assumptions made

regarding exposures, etc.). These intake estimates are combined with the COPEC

toxicity values, discussed in the following section, to derive estimates and characterize

potential ecological risk.

4.3.3.4.1. Uncertainties associated with the estimation of chemical intake should be

summarized at the conclusion of the characterization of exposure. The basis for each

uncertainty should be identified (e.g., use of a default parameter, propagation of error

through multiple layers of exposure modeling), the degree of the uncertainty

qualitatively (low, medium or high) or quantitatively estimated, and the impact of the

uncertainty qualitatively (overestimate and/or underestimate) or quantitatively stated.

EM 200-1-4

31 Dec 10

4-25

Description and presentation of uncertainties are discussed further in Chapter 5 (Risk

Characterization) and Chapter 6 (Uncertainty in ERAs).

4.4. Ecological Effects Characterization – SLERA. Screening level risk assessments

may be largely qualitative, using simple comparisons of abiotic media concentrations to

readily available screening "effects" criteria for these media, or they may employ a more

quantitative investigative approach that incorporates a threshold level or dose-response

assessment. In the more quantitative approach, screening level ecotoxicity values

(reference diet, dose, tissue, threshold levels) are developed for the principal receptors of

concern based on the complete exposure routes. For these complete exposure routes,

the lowest exposure level (e.g., concentration in abiotic media, or in diet [ingested dose])

shown to produce no adverse effects (e.g., reduced growth, impaired reproduction,

increased mortality) in the receptor of concern is identified. Where NOAELs are not

available, the LOAELs or other available toxicity values should be used and this

procedure should be discussed in the uncertainty discussion. The mode of toxicity

represented by the screening criterion should match the mechanism of toxicity for the

contaminant in question. For example, dioxins do not exhibit acute lethality as much as

they inhibit successful reproduction. Therefore the criterion for dioxins should be a

reproductive measure.

4.4.1. Published Literature, Available Criteria and Information. Sources for obtaining

ecotoxicity benchmarks in a screening assessment are generally limited to published

literature and readily available criteria and information such as the following. It should be

noted that selection of appropriate benchmarks should be done during the TPP session

prior to field work. This upfront agreement insures that problems are not encountered

during review of the ERA.

a. Federal AWQC. EPA's compilation of national recommended water quality

criteria is presented as a summary table containing recommended water quality criteria

for the protection of aquatic life and human health in surface water for approximately

150 pollutants. These criteria are published pursuant to Section 304(a) of the CWA

and provide guidance for states and tribes to use in adopting water quality standards.

b. State AWQC.

c. USEPA (USEPA 1996, NOAA (Long and Morgan 1991, Long et al. 1995), and

Ontario (Persaud, et al. 1992) sediment criteria,

d. SQuiRTs. NOAA has developed a set of Screening Quick Reference Tables, or

SQuiRTs, that present screening concentrations for inorganic and organic contaminants

in various environmental media. The SQuiRTs also include guidelines for preserving

samples and analytical technique options. The SQuiRTs were designed as a set of

double-sided, color cards, organized into the following sections: Inorganics in Solids

(freshwater and marine sediment, plus soil); Inorganics in Water (groundwater and

surface water); Organics in Water and Solids; Analytical Methods for Inorganics;

Analytical Methods for Organics; and Guidelines for Sample Collection & Storage.

EM 200-1-4

31 Dec 10

4-26

e. USEPA EcoSSLs. EPA's Superfund program has issued ecological soil

screening levels (Eco-SSLs) for twelve metals (aluminum, antimony, arsenic, barium,

beryllium, cadmium, chromium, cobalt, copper, iron, lead, manganese, nickel, silver and

vanadium), DDT, pentachlorophenol, and total PAHs, frequently found in soil at

Superfund sites. Eco-SSLs for additional contaminants, including values for selenium,

zinc, and dieldrin, are pending as of this writing.

f. USEPA ECOTOX Database. The ECOTOX database provides single chemical

toxicity information for aquatic and terrestrial life. ECOTOX is a useful tool for

examining impacts of chemicals on the environment. Peer-reviewed literature is the

primary source of information encoded in the database. Pertinent information on the

species, chemical, test methods, and results presented by the author(s) are abstracted

and entered into the database. Another source of test results is independently

compiled data files provided by various United States and International government

agencies.

g. ORNL ecological benchmarks. Screening ecological benchmarks are used to

identify chemical concentrations in environmental media that are at or below thresholds

for effects to ecological receptors. The Environmental Sciences Division of ORNL

developed and compiled a comprehensive set of ecotoxicological screening

benchmarks for surface water, sediment, and surface soil applicable to a range of

aquatic organisms, soil invertebrates, and terrestrial plants. Links to supporting

technical reports from which the benchmarks were obtained are also provided.

h. USACHPPM Wildlife Toxicity Assessments (WTAs) The WTAs address a

primary wildlife and habitat risk evaluation component--subsequent and related toxicity

associated with exposures to a specific chemical--through rigorous investigation and

evaluation of toxicity data, and logical derivation of terrestrial wildlife Toxicity Reference

Values.

i. USACHPPM Terrestrial Toxicity Database (TTD) An assemblage of chemical-

specific soil screening levels (SSLs) and TRVs that assist in plant, invertebrate and

wildlife risk identification.

4.5. Characterization of Ecological Effects – BERA. The ecological effects

characterization (toxicity assessment) includes a summary of the types of adverse

effects on biota associated with exposure to site-related chemicals, relationships

between magnitude of exposures and adverse effects, and related uncertainties for

chemical toxicity, particularly with respect to site biota. Ecological receptor health

effects are characterized using critical toxicity values, when available, in addition to

selected literature pertaining to site- and receptor-specific parameters.

4.5.1. Preliminary Toxicity Evaluation. The preliminary toxicity evaluation provides

toxicological profiles centered on health effects information on site biota. The profiles

cover the major health effects information available for each COPEC. Data pertaining

to site-specific species are emphasized, and information on domestic or laboratory

EM 200-1-4

31 Dec 10

4-27

animals are used when site-specific biota data are unavailable. Adequacy of the

existing database should also be evaluated as part of this task.

4.5.2. Bioassessment Tools and Techniques. Typically, the ecological effects

characterization is based on a desk-top HQ approach. Numerous bioassessment

tools,

a. Demonstrate whether the COPECs are bioavailable,

however, are available to the risk assessor to employ for directly measuring or

investigating toxicity, or even risk. Bioassessment techniques offer several advantages

over the HQ or model approaches to toxicity estimation including:

b. Evaluate cumulative impacts due to exposure to multiple COPECs,

c. Evaluate toxicity of COPECs for which no TRVs can be found,

d. Characterize the nature of the toxicity,

e. Integrate media variations and spatially characterize toxicity,

f. Monitor impacts before and after remediation,

g. Develop remedial levels in terms of toxicity and then monitor effectiveness and

success of remedial actions.

4.5.3. Objectives. The characterization of effects in the BERA fulfills two specific

objectives. First, available toxicological literature is reviewed to identify appropriate

literature benchmark values to use, forming the basis for developing summaries of the

potential toxicity of the COPECs for inclusion in the risk assessment. Second,

appropriate TRVs are developed using literature benchmark values to estimate

potential ecological risks associated with key receptor chemical exposure. This is

accomplished by reviewing the available information on COPEC toxicity and

summarizing the factors pertinent to the exposures being assessed. At this time, it is

only rarely necessary to search the literature for toxicity information. The toxicity of

most contaminants has been evaluated and TRVs have been calculated for use in

ERAs. The information that follows is intended to assist when it is necessary to make

the calculations. Reference USACHPPM (2000) for recommended procedures.

4.5.4. Sources of Literature Benchmark Values. The sources that should be

consulted for literature benchmark values will vary with the type of organisms being

used as ecological receptors (e.g., aquatic, terrestrial) and the level of effort. If the level

of effort (time and money) is limited, then documents that summarize available

ecotoxicological information will suffice. If a higher level of certainty in the data is an

objective in the compilation of literature benchmark values, then the primary

toxicological literature should be sought so that details of the toxicity test conditions can

be reviewed, validity of the test results confirmed, and applicability to site conditions

determined.

An in-depth discussion of topics related to the use of bioassessment approaches in ERAs is available in

the September 1994, Volume 2 series of Eco Updates (USEPA 1994f, 1994g, 1994h, 1994i)

EM 200-1-4

31 Dec 10

4-28

4.5.4.1. Toxicologic information on chemicals in aquatic ecosystems is fairly

plentiful, while that for terrestrial ecosystems is somewhat more limited. Most of the

available toxicological information for soil-based exposures has been generated using

soil-dependent biota. ORNL (1998), however, has published benchmark values for

plants, sediment associated biota, and terrestrial wildlife. Compilations of toxicological

data for soil-dependent organisms (plants, invertebrates, and microbes) are available in

the open literature (Hulzebros et al. 1993, Kabata-Pendias and Pendias 1984, USFWS

1990, Overcash and Pal 1979, Gough et al. 1979, Callahan et al. 1994).

4.5.4.2. Toxicity data and information for developing wildlife TRVs also may be

obtained from many of the same sources used for human health toxicity information,

particularly where data on small mammals (rats and mice) are needed. Regional

USEPA BTAGs (for NPL sites) and the Tri-Service Environmental Risk Assessment

Working Group (TSERAWG) persons can also be contacted for assistance. Other

sources for aquatic and terrestrial laboratory data are presented below:

a. USEPA Criteria Documents. Includes ambient water criteria documents,

proposed sediment quality criteria documents, drinking water criteria documents, air

qual effects assessment documents.

b. USFWS Contaminant Hazard Reviews. (Eisler 1985-1999). This is a series of

reports reviewing the hazards of over 25 metals and organic compounds to fish, wildlife,

and invertebrates. All are available in portable document format (PDF) files on-line.

c. Oak Ridge National Laboratory (ORNL 1998). Toxicological screening

benchmarks for ERAs. This series of reports includes benchmarks for terrestrial

wildlife, terrestrial plants, sediment-associated biota, and aquatic biota, and soil and

litter invertebrates and heterotrophic processes.

d. Toxicological Profiles developed by the Agency for Toxic Substances and

Disease Registry (ATSDR).

e. Aquatic and terrestrial toxicological data (and in some cases, literature citations).

Available in public or on-line databases such as TOXNET (contains ChemIDplus,

HSDB, Toxline, CCRIS, DART, GENETOX, IRIS, ITER, Multi-Database, Tri, Haz-Map,

TOXMAP), BIOSIS, ECOTOX (contains AQUIRE, TERRETOX and PHYTOTOX),

ASTER, Toxicity/Residue Database, Ecological Abstracts, Biological Abstracts, Current

Contents, Duck data (USFWS). ity criteria documents, and health

f. National Academies of Sciences publications such as Mineral Tolerance of

Domestic Animals (1980).

4.5.5. Selection of Literature Benchmark Values. Laboratory animals (rat and

mouse) studies are generally classified by the U.S. Dept. of Health and Human

Services (USDHHS) according to exposure duration: chronic (>365 days), intermediate

or subchronic (15 - 364 days), and acute (<14 days). In aquatic bioassay tests, test

durations for acute toxicity tests are typically 48 hours for invertebrates and 6 hours for

fish. Definitions of the terms chronic, subchronic, and acute, however, are often

EM 200-1-4

31 Dec 10

4-29

inconsistent, and depend on the organism being tested. Suter (1993) and USEPA

(1995b) arbitrarily consider chronic to be 10 percent of the organism's lifespan.

According to USEPA's health effects testing guidelines, chronic toxicity tests should

involve dosing over a period of at least 12 months. The organisms studied and study

duration should be reported when compiling literature benchmark values.

4.5.5.1. In selecting data to be used in the derivation of the TRV, the nature of the

observed endpoints is the primary selection criterion. Literature benchmark values

which best reflect potential impacts to wildlife populations through resultant changes in

mortality and/or fecundity rates should be used. Toxic responses such as elevated

enzyme levels (e.g., elevated blood aminolevulinic acid dehydrase from exposure to

lead) or increased tissue concentrations, while they may serve as good biomarkers

indicative of an organisms' exposure, are not useful endpoints insofar as being relevant

and indicative of adverse impacts to key receptor populations. Relevant intermediate

and chronic endpoints are those which affect organismal growth or viability, or

reproductive or developmental success or any other endpoint which is, or is directly

related to, parameters that influence population dynamics. The toxic effect manifested

at the lowest exposure level is (generally) selected as the critical effect. For some

ERAs, however, the lowest acute level also is selected for use in determining an acute

TRV. Where the toxicity database is large enough, a dose-response curve may be

generated and used as the basis to select a literature benchmark value or determine

the TRV.

4.5.5.2. The following factors should be considered when selecting literature

benchmark values and developing TRVs for use in the risk assessment:

a. Literature benchmark values should be obtained from bioassays having test

conditions as similar as possible to on-site conditions. For example, water hardness,

which affects the toxicity of many metals, should be the same in order to have the

bioassay results applicable to site conditions.

b. The literature benchmark values and TRV should correspond with the exposure

route being assessed; in ERAs, this is most typically the oral exposure route (dermal

exposure may be assessed using modified oral toxicity values).

c. The TRV should be appropriate for the key receptor and toxicity endpoint being

assessed, e.g., assessment of reproductive and developmental effects in mammals and

birds would require at least two, but possibly four, TRVs. TRVs for different toxicity

endpoints in different receptors or receptors groups may need to be developed.

d. The literature benchmark value and TRV should correspond to the appropriate

exposure duration period: subchronic (two weeks to 1 year) or chronic (greater than one

year).

e. The literature benchmark value and TRV should correspond to the chemical form

being assessed (only applicable to some chemicals, but especially metals such as

chromium [trivalent or hexavalent] and mercury).

EM 200-1-4

31 Dec 10

4-30

4.5.5.3. The process for selecting benchmark toxicity values is flexible so that site-

specific considerations can be incorporated. Careful consideration should be given to

the development of benchmark toxicity values, as they may provide the preliminary

information used to set the target cleanup levels at sites where remedial action is

anticipated. In the HQ approach, the TRV is essentially the measurement endpoint and

the hazard ratios calculated are inherently no more protective than the nature of the

toxic mechanism described by the TRV. Caution should be taken in the assessment

and selection of the TRV. For example, if the TRV were based on "acute" lethality, it

would not be protective of chronic exposure conditions.

4.5.6. Development of Toxicity Reference Values. Determination of TRVs for

terrestrial and aquatic organisms is dependent on both life style and life stage.

Literature benchmark values and TRVs for organisms in aquatic ecosystems (e.g.,

benthic macroinvertebrates and fish) are generally concentration-based, but can be

dosed-based for amphibians and higher trophic level receptors (waterfowl and aquatic

mammals). Amphibian exposure is perhaps the most difficult to quantify, as they have

both concentration-based aquatic life stages and dose-based terrestrial life stages.

Terrestrial TRVs can also be either concentration-based (e.g., flora and soil

invertebrates) or dose-based (e.g., vertebrate fauna).

4.5.6.1. Federal AWQC are frequently used as the equivalent of a TRV for aquatic

organisms. On some sites, AWQC may be judged to be overly cautious TRVs for the

specific key receptors, if the organisms on which the AWQC are based are far more

sensitive than any on-site receptors. In these cases, toxicity information used to

develop the original AWQC may be used in conjunction with other toxicity data and

literature benchmark values to develop a more site- and receptor-specific TRV, or it

may be decided to do site-specific bioassays or toxicity testing.

4.5.6.2. In terrestrial ecosystems, two types of TRVs are needed: concentration-

based TRVs for soil-dependent organisms and dose-based TRVs for wildlife. TRVs for

soil-dependent organisms (e.g., plants, earthworms) are similar to AWQC in that they

are concentration based. TRVs for wildlife are similar to the critical toxicity values

(reference doses) used in human health risk assessments. In order to appropriately

select and use TRVs and to identify assumptions and uncertainties associated with

TRVs, an understanding of the general practice currently followed in selecting TRVs is

needed. Site-specific TRVs for aquatic and terrestrial ERAs should be developed in

consultation with local wildlife and regulatory agencies.

4.5.6.3. Development of Aquatic TRVs. As stated above, aquatic TRVs can be

based on state or federal AWQC. However, especially in the case of metals, toxicity

can be significantly affected by site-specific factors. Factors that can affect site-specific

As assessment endpoints are typically phrased in terms of protecting populations, the TRVs focus on

measures of growth, survival and reproduction. Under some circumstances, it may be appropriate to protect

lower levels of biological order and employ biomarkers as benchmark values. Additionally, certain

biomarkers are indicative of conditions which have direct implications to assessment endpoints of growth,

survival, or reproduction and are not merely exposure markers.

EM 200-1-4

31 Dec 10

4-31

values include: ambient water chemistry, different patterns of toxicity for different

metals, metals fate and transport, use of standardized protocol for clean and ultraclean

metals analysis. Also, applicability of the chronic criterion or acute criterion to the

species of concern should be confirmed. Because AWQC have been calculated to

protect populations of the most sensitive aquatic species, these criteria may be over (or

under) protective of the aquatic ecological receptor(s) selected for the risk assessment.

Methods used to calculate AWQC are described in Appendix A of the "Gold Book"

(USEPA 1986a) and more recently in the USEPA's Water Quality Standards Handbook

(USEPA 1993f) and Interim Guidance on Interpretation and Implementation of Aquatic

Life Metals Criteria (USEPA 1992g, 1993b, 1995d). To determine the basis for a

particular chemical, the AWQC document for that metal or compound should be

consulted. In the case of metals, the basis (total, total recoverable, or dissolved

concentration) for the TRV or criterion and the chemical concentrations to which it is

compared should be verified and consistent.

4.5.6.4. Development of Terrestrial TRVs for Soil-Dependent Organisms.

Screening ecological benchmarks are used to identify chemical concentrations in

environmental media that are at or below thresholds for effects to ecological receptors.

The Environmental Sciences Division of ORNL developed and compiled a

comprehensive set of ecotoxicological screening benchmarks for surface water,

sediment, and surface soil applicable to a range of aquatic organisms, soil

invertebrates, and terrestrial plants. These benchmarks, or updates performed in

collaboration with the Center for Information Studies at the University of Tennessee and

the Bechtel Jacobs Corp., are provided as a searchable database. Links to supporting

technical reports from which the benchmarks were obtained are also provided.

4.5.6.4.1. The USEPA Eco-SSLs provide screening values for soil invertebrates

and plants. It is emphasized that the Eco-SSLs are soil screening numbers, and as

such are not appropriate for use as cleanup levels. Screening ecotoxicity values are

derived to avoid underestimating risk. Requiring a cleanup based solely on Eco-SSL

values would not be technically defensible.

4.5.6.4.2. Countries outside the U.S. (Canada, Netherlands) have developed

various cleanup criteria for soils. Most of these criteria are with respect to groundwater

protection although some countries (e.g., Canada) have developed a limited number of

soil criteria based on phytotoxicity and animal health (ASTM 1995).

4.5.6.5. Development of Terrestrial TRVs for Wildlife. In general, TRVs are needed

to represent levels of exposure that are associated with low risk for entire taxonomic

classes (e.g., mammals) or for selected foraging guilds (e.g., carnivorous mammals).

USACHPPM’s Technical Guide 254 (USACHPPM 2000) focuses upon the development

of chemical-specific TRVs for these receptor groups.

4.5.6.5.1. The methodology for generating defensible wildlife TRVs and for

preparing acceptable documentation to support such TRVs consists of two phases.

The three steps for Phase 1, Toxicity Profile, are performance of data collection and

literature search, identify relevant studies, and prepare a toxicity profile. The three

EM 200-1-4

31 Dec 10

4-32

steps for Phase 2, TRV Report, are derive TRVs and document selection rationale,

assign confidence levels to the TRV, and prepare the TRV report.

4.5.6.6. Use of an Acute to Chronic Conversion Ratio. In some cases, chronic

toxicity data are not available and an acute/chronic ratio must be applied to acute

toxicity data (typically mortality) to estimate chronic effects levels. Because wildlife

toxicity databases are fairly limited, use of a factor for extrapolating from acute data to

chronic data will likely be large and result in an overly conservative TRV.

4.5.6.7. Short Term Critical Toxicity Values. Certain exposures, such as during

construction or remediation activities, may occur only for a brief time. Likewise,

exposure of mobile wildlife to site contamination may be brief and intermittent. These

exposures require the use of short term, or acute toxicity values. In most cases, risk

assessments are concerned with longer exposures that are appropriately addressed by

subchronic or chronic TRVs. Applying these values, however, to very short term

exposures (less than two weeks) may not be valid. Results of primary toxicology

studies should be used in evaluating potential effects of short-term chemical exposures.

Direct comparisons should be made cautiously, however, because of the limitations of

single study results. The uncertainties and assumptions involved in the use of acute

TRVs should be clearly stated in the assessment.

4.5.6.8. Feeding and Drinking Rates. When drinking and feeding rates and body

weight are needed to express the NOAEL or LOAEL in mg/kg-bw/d, they should be

obtained from the literature benchmark study from which the NOAEL or LOAEL was

derived. As noted earlier, dietary chemical concentrations in mg/kg must be normalized

for body weight and food intake of the test organism and receptor of concern before

they can be used as a screening benchmark.

4.5.6.8.1. Depending on the organism and study, dry weight chemical

concentrations may also need to be converted on a wet-weight basis. Use of wet weight

versus dry weight in estimating dietary exposures can be problematic, particularly

where the moisture content of the diet is highly variable (e.g., in plants). Dietary

concentrations in most toxicological studies are reported on a wet-weight basis.

However, moisture content of laboratory diets is also typically less than 10 percent, so

this difference is sometimes ignored (Beyer and Stafford 1993). The risk assessor

should, at a minimum, strive to be consistent (or conservative) in reporting between wet

weight when comparing the TRV to the exposure intake value in the risk calculation.

The basic equation for converting tissue analyte concentration between dry and wet

weight samples is:

wet weight tissue concentration = dry weight tissue concentration x (% solids/100).

where: % solids = 100 - % moisture

Given a 230 mg/kg wet weight of lead in plants and 20% moisture content, the dry weight concentration

would be 287.5 mg/kg.

EM 200-1-4

31 Dec 10

4-33

4.5.6.8.2. If the literature benchmark study does not provide the needed values,

they should be determined from appropriate data tables for the particular study species.

For studies done with domestic laboratory animals, Registry of Toxic Effects of

Chemical Substances (RTECS) (National Institute of Occupational Safety and Health

[NIOSH], latest edition) can be consulted. When insufficient data exist for other

mammalian or avian species, the allometric equations from Calder and Braun (1983),

Nagy (1987), and USEPA (1988d, 1993d) can be used to calculate feeding and drinking

rates (see Section 4.3.3.3.1). Reference food and water intake values for a variety of

wildlife are also provided in ORNL (1998).

4.5.7. Additional Considerations in Developing TRVs. There are a number of

additional factors that should be considered when conducting the effects

characterization, reviewing the toxicological literature and determining TRVs. These

are discussed in the following sections.

4.5.7.1. Absorption Considerations. Most toxicity values are based on

administered, rather than absorbed, doses, and the absorption efficiency has not been

considered. However, whatever absorption has occurred during the toxicological study

is inherent in the toxicity value. Therefore, use of a toxicity value assumes that the

extent of absorption observed in the study is also appropriate for the exposure pathway

being assessed. Differences in absorption efficiencies between that applicable to the

TRV, and that being assessed may occur for a number of reasons. Two factors that will

influence absorption efficiencies are differences in chemical form and differences in the

exposure medium.

4.5.7.1.1. The form of the chemical used in the literature benchmark wildlife study

may not be the same as the chemical form present in the environmental medium being

assessed, and may be absorbed to a different degree. Therefore, use of the toxicity

value may over- or underestimate the actual absorption potentially occurring in

receptors. This is especially important for certain metals where inorganic forms (e.g.,

metallic lead) differ widely from organic forms (e.g., lead acetate) in their potential

toxicity. The basis of the chemical's TRV should be reported in the effects

characterization and compared with the form (if known) in the site media. Often the

form in site media is not known, but can sometimes be inferred based on site history or

by the medium in which the chemical is found (for example, a metal in soil is unlikely to

be present in its soluble form).

4.5.7.1.2. In toxicity studies, chemicals are often administered in drinking water,

mixed with food, or mixed in an administration vehicle such as olive oil to facilitate

absorption. In environmental settings, exposure to chemicals may occur in a medium

similar to that used in the study (e.g., in drinking water) or in a medium quite different

from that used in the study (e.g., the soil matrix). Certain media, particularly soil and

sediments, may bind chemicals, reducing the amount that is available for absorption

(i.e., bioavailability). In these instances, it may be appropriate to reduce the COPEC

intake value in the exposure calculation with a matrix effects or bioavailability factor to

account for this binding (see Section 4.3.3.3.5).

EM 200-1-4

31 Dec 10

4-34

4.5.7.1.3. Numerous studies show that not only metals but organic chemicals,

including pesticides, bind tightly to soil, reducing their bioavailability through both oral

and dermal exposure. Calderbank (1989) showed that clays and organic colloids have

a large surface area and cation exchange capacity, which permits significant adsorption

of virtually all classes of pesticides; furthermore, the adsorbed fraction (20% to 70%)

desorbs slowly and is effectively a bound fraction that increases over time as the soil-

pesticide bond "ages". Shu et al. (1988) reported a bioavailability range of 25 to 50%

for TCDD to rats from soils at Times Beach, Missouri. Goon et al. (1991) showed that

benzo(a)pyrene (BaP) that had aged 6 months in soil was only 34 and 51% orally

bioavailable for clayey and sandy soils, relative to BaP administered alone to rats. In

general, differences in absorption between lab media and site media should not be

assumed, unless there is adequate information to the contrary.

4.5.7.2. Assessment of Inhalation Exposure Route for Wildlife. Inhalation exposure

routes are generally not addressed in ERAs due to the lack of toxicity information for

wildlife species and the lesser significance of the inhalation exposure route to the oral

ingestion route.

4.5.7.2.1. In order to quantitatively evaluate this exposure route, the risk assessor

may need to consider factors such as the target species' airway size, branching pattern,

breathing rate (volume and frequency), and clearance mechanisms, whether the

contaminant is a gas or aerosol, whether the chemical's effects are systemic or

confined to the respiratory tract, as well as particle size distribution, temperature and

vapor pressure and pharmacokinetic data (USEPA 1993d). In addition, the dose

deposited, retained and absorbed in the respiratory tract is a function of species

anatomy and physiology as well as physicochemical properties of the contaminant.

Allometric equations are available from USEPA (1993d). A procedure for calculating

inhalation exposure is also published by USDOI (1991).

In general, VOC concentrations of 100 ppm or greater in air are

needed to induce toxic responses in laboratory rats and mice from inhalation (NIOSH

1987). Concentrations in soils would have to be many times greater than this to

produce these toxic levels in air, even near the soil surface.

4.5.7.2.2. Total petroleum hydrocarbon (TPH) contamination is one example where

the inhalation of volatiles for small, burrowing animals is of concern in the ERA. W.

Kappleman in Maughan (1993) provides a methodology for determining ecological

effects levels for muskrat and beaver via inhalation and dermal exposure pathways for

benzene, toluene, ethylbenzene, total xylenes (BTEX) and PAHs. These

methodologies may be applied where site-specific conditions require inhalation

exposure to be considered an important exposure route. The methodology for

calculating inhalation concentrations for humans as discussed in USEPA's (1990b)

Interim Methods for Development of Inhalation Reference Concentrations may be

followed to some extent.

A notable exception is the great number of studies conducted on response and uptake by birds and

mammals from aerial pesticide spraying on agricultural crops.

EM 200-1-4

31 Dec 10

4-35

4.5.7.3. Assessment of Dermal Exposure Route for Wildlife. Dermal exposure

routes are generally not addressed in ERAs due to limited toxicity information for

terrestrial wildlife species and the lesser significance of the dermal exposure route to

the oral ingestion pathway. The dermal pathway may be of importance where wildlife

are directly sprayed or frequent areas with surface-contaminated vegetation or where

the animals are burrowing in contaminated soils/sediments.

4.5.7.3.1. Wildlife are generally assumed to be protected by their fur, feathers or

scales, which prevent a chemical from reaching an animal's skin and may allow the

chemical to dry or to be rubbed off during movement. Dermal absorption of

contaminants is a function of chemical properties of the contaminated medium, the

permeability of the receptor's outer covering, area in contact with the contaminated

medium and the duration and pattern of contact. The methodology for calculating

dermal exposure concentrations for humans is discussed in USEPA's (2004) Risk

Assessment Guidance for Superfund, Volume I: Human Health Evaluation Manual (Part

E, Supplemental Guidance for Dermal Risk Assessment)(RAGS E), and may be

followed to some extent where dermal exposure concentrations for wildlife need to be

calculated.

4.5.7.3.2. Dermal exposures may be of concern for wildlife that swim or burrow.

Mammals and birds groom themselves regularly and may receive an oral ingestion

dose from dermal contamination of their fur or feathers. An oral ingestion dose for

animals which groom themselves may be calculated based on a methodology published

by USDOI (1991) for determining dermal exposure to representative western rangeland

wildlife species from herbicide sprays. W. Kappleman in Maughan (1993) provides a

methodology for determining ecological effects levels for muskrat and beaver via

dermal exposure pathways for BTEX and PAHs. Such a methodology may be applied

where site-specific conditions require dermal exposure to be considered an important

exposure route.

4.5.8. Special Chemicals. Some commonly detected chemicals require special

consideration in the generation of a TRV (e.g., their potential to biomagnify, need for a

surrogate component evaluation, difficulty in obtaining toxicity information) or have

specific chemical forms that greatly influence bioavailability and toxicity. The following

chemicals are discussed in this light:

a. Metals,

b. PAHs,

c. Organochlorine Pesticides (OCPs) and PCBs,

d. Chlorinated Dibenzo-p-dioxins and chlorinated Dibenzofurans (CDDs/CDFs),

e. TPH and other petroleum groupings, and

f. Military chemicals.

4.5.8.1. Metals. The toxicity of metals depends foremost on chemical form. For

example, chromium (+3) occurs naturally and is common in the environment and has a

EM 200-1-4

31 Dec 10

4-36

relatively low toxicity. Chromium (+6) is largely related to anthropogenic releases and is

very toxic, but is readily reduced in the environment to chromium (+3). Organometallic

forms (methylmercury, alkylead) are more toxic than the elemental forms. Much of the

literature does not specify the chemical form of an element when discussing its toxicity

to biota. It may be assumed in these instances that only the total concentration of the

metal was known.

4.5.8.1.1. To be toxic an element must be available to the receptors. In order for

this to occur, the chemical must exist in a form that can enter tissues of the organisms.

Total amounts of a chemical in the environment are not relevant to an adequate

estimation of toxicity hazard unless it can be shown that the element exists in or is likely

to assume, an available form under the environmental conditions in which it occurs and

animals or plants are likely to contact this form either directly of indirectly (Gough et al.

1979).

4.5.8.1.2. Aquatic Organisms. The site-specific toxicity of a metal to aquatic

organisms depends on the physical form of the metal, the effect of other metals and

organic compounds (anthropogenic and naturally occurring) in the water, as well as the

chemical or ionic form of the metal of interest. Metals results from surface water

analyses can be reported in terms of the total recoverable metals, total metals, acid

soluble metals, or dissolved metals. All four methods measure all of the dissolved

metal present but differ (because of varying field or laboratory procedures) in the

amount of particulate metal measured. While federal AWQC are reported as total

recoverable metals, many states have standards based on dissolved metals. The basis

and form (dissolved versus total) of the specific criteria should be verified before being

applied at a site. The risk assessor may also need to take into account transformation

of on-site metals to bioavailable forms with migration off-site.

4.5.8.1.2.1. In order to develop a better understanding of metals criteria,

bioavailability, and toxicity, USEPA has issued a series of guidance documents

(USEPA 1992g; 1993b; 1995d) to supplement the Water Quality Handbook (USEPA

1993f). These documents describe:

a. Relationships among the various physical forms reported in water quality results;

b. The importance of site-specific bioassays (if this level of effort is justifiable) to

create a WER to account for the fact that in situ metals toxicities are frequently less

than reported from laboratory bioassay tests; and

c. Observed ratios between dissolved metals and total recoverable metals in order

to facilitate interpretation of AWQC and the more bioavailable dissolved metals.

4.5.8.1.3. Plants. Plants are intermediate reservoirs through which trace metals

from primary sources move to other living things, as well as receptors directly exposed

to metals in soils. Plants may be passive receptors of trace metals, as in root

adsorption, or they may accumulate and store metals in nontoxic forms for later

distribution and use (Tiffin 1977). A mechanism of tolerance in some plants apparently

EM 200-1-4

31 Dec 10

4-37

involves binding of potentially toxic metals at the cell walls of roots and leaves, away

from sensitive sites within the cell. The metal forms which occur in plants appear to

have a decisive role in metal transfers to other organisms (Tiffin 1977).

4.5.8.1.3.1. There are a large number of processes that operate to regulate metal

cycling, including ion exchange, adsorption, formation of organic complexes, and

precipitation. All these have different, and often opposing effects; and all are very

dependent on pH, and other soil/sediment characteristics. Since site conditions vary so

much in these respects, both spatially and temporally, metal reactions and fates often

vary. In addition to environmental variability, there are differences due to plant

physiology and genotype (Outridge and Noller 1991). Therefore, it is very difficult to

extrapolate from one study location or plant to another.

4.5.8.1.3.2. As described in Dunbabin and Bowmer (1992), there are some general

trends that have been noted. Potential bioavailability generally increases with

increases in acidity, reducing power, salinity, and concentration of organic ligands.

However, if sulfur is present, a reducing environment will result in the production of

insoluble metal sulfides. Other specific factors that influence bioavailability include

sediment size (clay provides more surface area for adsorption and reactions), presence

of hydrous iron and manganese oxides (which adsorb metals), and the nutrient regime

(which, for example, affects the ability of microbes to transform elemental mercury to

methylmercury) (Stewart et al. 1992).

4.5.8.1.4. Terrestrial Fauna. Several metals, while potentially toxic, are also

essential micronutrients for plants and animals, e.g., zinc, selenium. All metals,

whether essential or nonessential, can adversely affect terrestrial organisms, if included

in the diet at excessively high levels. In general, tolerance levels vary from animal to

animal and even from day to day in a single animal (NAS 1980). Many factors, such as

age and physiological status of the animal (growth, lactation, etc.), nutritional status,

levels of various dietary components, duration and route of exposure, and biological

availability of the compound, influence the level at which a metal may cause an adverse

effect in the organism (NAS 1980). Exposure of animals to excessively high

concentrations of metals can result in acute signs of toxicosis, which may be quite

different from the chronic effects displayed after the metal has been ingested at higher

than normal levels over an extended period of time.

Metals that biomagnify (e.g., mercury, selenium) require the application of food

chain multipliers (BAFs or biomagnification factor (BMF)) to concentrations in prey

organisms for higher trophic level predators. Concentrations of inorganic metals in a

BAF or BCF study should be greater than normal background levels and greater than

levels required for normal nutrition of the test species if the substance is a micronutrient

(e.g., selenium), while still below levels which adversely affect the species (USEPA

1995b).

19.

Care should be taken in using partitioning models to estimate BCFs or BAFs for soil dependent organisms

such as earthworms and plants. Models based on diffusivity constants and anaerobic conditions can result in

unrealistically toxic concentrations (>1 percent) in the soil organism.

Bioaccumulation of inorganic metals may be inappropriately overestimated if

EM 200-1-4

31 Dec 10

4-38

concentrations are at or below normal background levels due to, for example, nutritional

requirements of the test organisms (USEPA 1995b).

4.5.8.2. PAHs. PAHs, also known as polynuclear aromatics (PNAs), are a class of

compounds containing hydrogen and carbon in multiple ring structures. There are

numerous possible PAH molecules, several of which are common analytes in a semi-

volatile compound analysis. PAHs are natural components of petroleum and are found

in heavier petroleum fractions, such as lube oil, naphtha, etc. PAHs are also produced

by the incomplete combustion of organic matter. For this reason, PAHs are ubiquitous

in the environment at low levels, particularly in soil and sediments, to which they readily

bind.

4.5.8.2.1. In general, PAHs are rapidly metabolized and considered unlikely to

biomagnify despite their high lipid solubility (Eisler 1987b). Inter- and intra-species

responses to individual PAHs are quite variable, however, and are significantly modified

by many inorganic and organic compounds (Eisler 1987b). Until these interactive

effects are clarified, extrapolation of laboratory test results to field situations where

there is suspected PAH contamination should proceed cautiously.

4.5.8.2.2. Amphibians, are reported as quite resistant to PAH carcinogenesis when

compared to mammals due the amphibian's inability to produce mutagenic metabolites

of BaP and perylene (Anderson et al. 1982). The ability to metabolize PAHs in

nonmammalian species, however, is extremely variable and cannot be predicted on the

basis of phylogenic associations. When PAHs are not metabolized, they have been

shown to bioaccumulate and therefore pose a significant dietary route of exposure to

predatory species. In species which can metabolize PAHs, one significant mode of

toxicity is impairment of reproductive cycles.

4.5.8.2.3. Small mammals which burrow and ingest soil are likely to be the

ecological receptors with the greatest potential exposure and risk from PAHs. Data are

generally lacking on the acute and chronic toxicity of PAHs on avian wildlife (Eisler

1987b). Eisler (1987b) reports PAHs show little tendency for bioconcentration or

biomagnification, particularly in terrestrial ecosystems, probably because most PAHs

are rapidly metabolized. Beyer and Stafford (1993) also found PAH concentrations in

earthworms to be well below soil levels. Gile et al. (1982), however, report fairly high

bioaccumulation factors for terrestrial species. In their 3-month mesocosm experiment

using creosote coal tar distillate (which contained 21% phenanthrene and 9%

acenaphthene), PAH concentrations in various animals were found to be elevated over

average PAH soil concentrations.

4.5.8.2.4. PAHs can accumulate to some extent in terrestrial plants. Atmospheric

deposition on leaves, however, is likely to be a more significant pathway than uptake

from soil by roots (Vaughn 1984). Uptake of PAHs by plant roots are dependent on

numerous factors including concentration, solubility, molecular weight of the PAH, and

on the plant species (Edwards 1983).

4.5.8.3. OCPs and PCBs. OCPs and PCBs are extremely stable compounds and

slow to degrade under environmental conditions. The toxicological properties of

EM 200-1-4

31 Dec 10

4-39

individual PCBs and pesticides are influenced primarily by two factors: the partition

coefficient, K

, based on solubility in n-octanol/water and stearic factors, resulting from

different patterns of chlorine substitution. The more highly chlorinated forms of PCBs

and pesticides tend to be more persistent, more strongly sorbed, less volatile, and less

bioavailable (O'Connor et al. 1990, Sawhney 1988, Strek et al. 1981).

4.5.8.3.1. PCBs and pesticides are strongly sorbed in soils, sediments, and

particulates in the environment, with levels usually highest in aquatic sediments

containing microparticulates (Eisler 1986b, USEPA 1980, Duinker et al. 1983). PCB

and pesticide uptake from contaminated soils and sediments is governed by processes

that include both direct incidental ingestion of contaminated soil/sediment particles and

indirect ingestion via food webs or from parents to the fetus or embryo. Toxicity reports

based on plant (terrestrial) uptake of pure PCBs and pesticides can be misleading

because these chemicals are often added to the exposure medium at unreasonably

high concentrations to facilitate analysis or they are added to coarse-textured soils

extremely low in organic matter (O'Connor 1989).

4.5.8.3.2. PCBs, dioxins, and pesticides are all highly lipophilic, with the greatest

concentrations occurring in fatty tissues. PCBs, dioxins, and pesticides are of greatest

concern to higher trophic level predators. In mammals, these chemicals are readily

absorbed through the gut, respiratory system, and skin, and can be transferred to

young mammals either transplacentally or in breast milk. In birds, particularly

endangered raptors, a reduction in eggshell thickness has been the endpoint of

greatest concern from pesticides. Evidence implicating PCBs as a major source of

eggshell thinning is inconclusive (Eisler 1986b, Wiemeyer et al. 1984, Henny et al.

1984, Norheim and Kjos-Hanssen 1984). Consideration of the potential

bioaccumulative effects of PCBs, dioxins, and pesticides is important in the selection of

appropriate assessment and measurement endpoints.

4.5.8.3.3. The 197 PCB congeners that are not dioxin-like PCBs (see Section

4.5.6.4) are currently referred to as “non-dioxin-like” congeners. These congeners are

also often referred to as the “non-coplanar” or “ortho-substituted” congeners. The

NCEA, Ecological Risk Assessment Support Center has issued guidance for

assessment of these PCBs in Non-Dioxin-Like PCBs: Effects and Consideration in

Ecological Risk Assessment (USEPA 2003).

4.5.8.4. CDDs and CDFs. CDDs and CDFs, often abbreviated "dioxins and furans"

are a group of chlorinated compounds based on the dibenzo-p-dioxin or dibenzofuran

molecule (the two of which are structurally similar). CDDs/CDFs are not compounds

used for commercial purposes in the past, and, outside of research, have no known

use. Rather, CDDs/CDFs are byproducts of high temperature combustion of

chlorinated compounds and impurities in other chemical products such as

pentachlorophenol (CDDs) or polychlorinated biphenyls (CDFs). Although not

considered a "natural" product, some forms of CDDs and CDFs (specifically octa-CDD

and octa-CDF) are ubiquitous in the environment at very low concentrations.

EM 200-1-4

31 Dec 10

4-40

4.5.8.4.1. There are 75 possible CDD congeners and 135 possible CDF congeners.

As with PCBs, the degree of toxicity varies with the degree and location of chlorination,

becoming greatest when the 2, 3, 7, and 8 positions of the molecule are substituted.

The 2,3,7,8-TCDD is considered the most potent CDD, and is the reference against

which all other CDDs and CDFs are compared.

4.5.8.4.2. Analysis of CDDs and CDFs is most commonly reported by congener

group (i.e., as either tri-, tetra-, penta-, hexa-, hepta-, or octachlorodibenzo-p-dioxin or -

dibenzofuran). Within these groups, the results are often further separated into

"2,3,7,8- substituted" or "other" categories. This form of reporting is needed to

appropriately assess CDDs and CDFs. Reporting as "total dioxins" or even just by

congener group may require the assumption that all CDDs/CDFs present are as toxic

as 2,3,7,8-TCDD, resulting in an overestimate of potential risk posed by the presence of

CDDs/CDFs.

4.5.8.4.3. Piscivorous fish and wildlife are thought to be particularly at risk from

these chemicals due to their large exposure through aquatic food chains. The limited

available toxicological data indicate that fish, especially salmonid sac fry, and mink

(Mustela vison) are among the most sensitive animals to TCDD and related

compounds. Assessment of the toxicity of these compounds along with environmental

concentrations associated with TCDD risk to aquatic life and associated wildlife has

been released by USEPA (1993g, 2006).

4.5.8.4.4. Two basic methods are recommended for evaluating the toxicity of

mixtures of dioxin-like PCBs, PCDFs, and PCDDS in environmental samples to

determine sample "toxic equivalents" relative to TCDD (USEPA 1993g). In the first

method (commonly used in screening ERAs), individual PCB, PCDF, and PCDD

congeners are determined and multiplied by toxic equivalent factors (TEFs) to express

potential toxicity in TCDD-equivalents (EQs). In the TEF approach for dioxin-like

PCBs/CDDs/CDFs, the toxicity of the TCDD compounds are expressed relative to the

toxicity of 2,3,7,8-TCDD for mammalian systems (Safe 1990, Ankley et al. 1992). Soil

or prey tissue doses of dioxin-like PCBs/CDDs/CDFs may be calculated by applying

congener-specific TEFs to the concentrations prior to conversion of concentrations to

doses. TEFs, however, are a species-specific construct and the TEF multipliers vary

widely among species, depending on their ability to metabolize specific congeners.

TEFs recommended by USEPA (1995b, 2006) and Safe (1990) are frequently used in

screening ERAs. Further discussion of TEFs for dioxin-like PCBs/CDDs/CDFs can be

found in Interim Report on Data and Methods for Assessment of 2,3,7,8-

Tetrachlorodibenzo-p-Dioxin Risks to Aquatic Life and Associated Wildlife (USEPA

1993g), USEPA's (1994e) dioxin wildlife workshop report, the Framework document

(USEPA 2006), and in the GLWQI (USEPA 1995b).

4.5.8.4.5. In the second method, the total PCB/PCDF/PCDD mixture is extracted

from the environmental samples and then tested for potency, relative to TCDD, using a

standard biological response (rat hepatoma cytochrome induction) as an endpoint

(USEPA 1993g). This latter approach bypasses the assumption of an additive model of

toxicity for complex mixtures. If the latter biological approach for measuring TCDD-EQ

EM 200-1-4

31 Dec 10

4-41

is to be used for quantitative risk assessment, it is important to calibrate the biological

system used with specific toxicological endpoints in the receptors of concern (USEPA

1993g).

4.5.8.5. TPH and Other Petroleum Groupings. TPH are common contaminants at

DoD sites. Petroleum hydrocarbons originate from a variety of petroleum-derived fuels

including jet fuel, fuel oils, and gasoline. Determination of the actual source material

(gasoline versus fuel oil) is not always possible, particularly where site history is

unknown. Composition of any given fuel will also vary depending on the source of the

crude oil, refinery processes, and product specifications. Also, due to differential

volatilization and biodegradation, the composition of the original fuel mixture in the

environment is altered over time. Therefore, the toxicity of the insoluble and nonvolatile

components remaining some time after a spill is often of more interest than volatile

compound toxicity.

4.5.8.5.1. Because of the originally unknown and potentially altered composition of

the spilled fuel, TPH toxicity is frequently assessed based on individually measured

constituent toxicity, rather then by assessing the measured TPH concentration as a

whole mixture. The primary constituents of petroleum components, such as paraffins

and naphthenes, are generally not considered to be highly toxic (Amdur et al. 1991;

Clayton and Clayton 1981) and are typically not included as COPECs in ERAs.

Aromatic constituents such as benzene and xylene and the carcinogenic PAH

compounds are the primary COPECs for risk assessments. Noncarcinogenic

compounds, such as toluene, ethylbenzene, xylenes, naphthalene, and other

noncarcinogenic PAH compounds, may be of concern for potentially acute toxic effects.

4.5.8.5.2. The impacts of TPH on terrestrial ecosystems are not as well

documented as the impacts on aquatic ecosystems.

4.5.8.6. Military Chemicals. Many DoD sites contain potentially toxic chemicals not

commonly found on non-military sites. Military-specific chemicals may include

explosives, rocket fuels, radioactive materials, chemical agents, or degradation

products of these compounds. Because of the unique status of many military

compounds, USEPA is often unable to supply toxicity information. Profiles containing

toxicological information relevant to an ERA can be obtained from USACHPPM WTAs

and the TTD. Technical reports that summarize environmental fate and behavior (plant

Some attempts have been made

in human health risk assessment to derive critical toxicity values for TPH. However,

since the composition of TPH varies from place to place (even within the same site) as

well as change in time (fresh versus aged product), it is unlikely that using critical

toxicity values for this group of chemicals provides valuable descriptors of the potential

toxicity of the components comprising the TPH detection.

The American Petroleum Institute (API) lists numerous reports regarding TPH toxicity in aquatic

ecosystems. Effects concentrations in water for various oil products (bunker, crude, diesel, gasoline, jet fuel,

lube oil), taxonomic group (invertebrates, fish, algae), and presence/absence of free product, can be found in

A Critical Review of Toxicity Values and An Evaluation of the Persistence of Petroleum Products for Use in

Natural Resource Damage Assessments, API, April 5, 1993.

EM 200-1-4

31 Dec 10

4-42

uptake, mammalian and aquatic toxicology) of munitions material are also available in

the open literature (Burrows et al. 1989, Cataldo et al. 1990, Layton et al. 1987).

Pertinent information can also be obtained from site-specific environmental studies at

installations such as Joliet AAP and Rocky Mountain Arsenal and by contacting the

regional USEPA or TSERAWG persons.

4.5.8.7. Toxicologic Uncertainties. Use of USEPA-derived aquatic and wildlife

toxicity values should be examined with regard to the degree of uncertainty associated

with their development. The uncertainties associated with the values should be stated

in the effects characterization, and the impact of applying the value estimated,

specifically (when the assessment is complete) for chemicals that are major

contributors to overall site risks and hazards. The following factors should be

addressed:

a. What are the cumulative uncertainties and modifying factors applied to derive the

TRV?

b. Is the form of the chemical used in derivation of the toxicity value the same or

similar to that in the environmental medium being assessed?

c. Is the duration of the toxicological benchmark study relevant to the exposure

conditions for the key receptors being assessed? Actual exposure durations for key

receptors may or may not exceed the test duration periods on which the TRVs are

based.

d. Was the medium applicable to the toxicological study used to derive the toxicity

value (e.g., the chemical was administered to the test animal in food, water) similar to

the medium being assessed? Could matrix effects or water effects be important in

bioavailability?

e. Has any route-to-route extrapolation been performed? Was it reasonable to do

so, and were assumptions used in the extrapolation appropriate?

f. Were surrogate toxicity values (toxicity values for other chemicals that are

structurally and/or chemically similar) used for chemicals that do not possess values?

Was this approach reasonable?

g. Were BCFs or BAFs applied in the development of the TRV? BAFs and BCFs

developed for one study may be quite different than bioaccumulation factors at other

areas.

h. If multiple contaminants are detected, is additivity assumed and is it a

reasonable assumption for the exposures expected? Note that some toxicity tests can

evaluate the toxicity of mixtures, but lack the ability to discriminate which contaminant is

causing the toxicity.

4.5.8.7.1. The potential exists for wildlife species to be more or less sensitive than

laboratory test species and the derived toxicological benchmarks. Toxicity benchmark

values for laboratory organisms may be substantially lower than those for wildlife due to

EM 200-1-4

31 Dec 10

4-43

the sensitive strains of laboratory animals used, the direct means by which they are

dosed, and the need to obtain a satisfactory toxic response. The LD

studies are

usually designed to promote maximum exposure (absorption) because less of the

chemical complexes with dietary material. The LD

dietary studies probably give a

better indication of the toxicity of the chemical tested, while no observed effects levels

from longer studies are the best (still imperfect) laboratory studies to be used as

predictors of field effects. On the other hand, laboratory species may be less sensitive

than their wild counterparts in that they must be hardy enough to be amenable to

culturing in a laboratory setting or endure animal husbandry and handling.

4.5.8.7.2. In contrast to laboratory tests of terrestrial organisms, laboratory tests of

aquatic invertebrates or fish show that the tested chemicals may be less toxic to the

same or similar animals under natural conditions. This is because the tested chemical

is not as bioavailable in natural waters due to the modifying effect of other water quality

characteristics (e.g., pH, hardness, suspended solids). In order to estimate the toxicity

of a chemical under natural conditions, a parallel series of toxicity tests are run using

site water and laboratory test water as dilution water and then calculating a water

effects ratio (site water LC

/lab water LC

EM 200-1-4

31 Dec 10

4-44

THIS PAGE INTENTIONALLY LEFT BLANK

EM 200-1-4

31 Dec 10

5-1

Per ERAGS (USEPA 1997a), risk characterization includes two major

components: risk estimation and risk description. Risk estimation integrates the

exposure profiles with the toxicity information and summarizes the associated

uncertainties. Then, the risk description provides the risk manager with information

useful for interpreting the assessment results.

USEPA has two requirements for the full characterization of risk (USEPA

1995a,c). First, the characterization should address qualitative and quantitative

features of the assessment. Second, it should identify the important strengths and

qualitative as well as quantitative uncertainties in the assessment as part of a

discussion of the confidence in the assessment. Risk characterization as the final

part of the ERA process provides:

Integration of the individual characterizations from the ecological effects

and exposure characterizations (see Chapter 4);

Evaluation of the overall quality of the assessment and the degree of

confidence in estimates of risk and conclusions drawn (see Chapter 6);

Description of risks in terms of extent, severity, and probable harm; and

Communication of risk assessment results to the risk manager.

CHAPTER 5

Risk Characterization

5.1. Risk Characterization – SLERA

5.1.1. Risk characterization is the summarizing step of the SLERA. The risk

characterization integrates information from the preceding components of the risk

assessment, performs a screening evaluation (or calculation) and synthesizes an overall

conclusion about risk that is complete, informative, and useful for decision-makers

(USEPA 1995c). The preliminary risk screen employs a conservative approach to ensure

that potential ecological threats are not overlooked.

5.1.2. For the risk estimation, the HQ approach, which compares point estimates

of TRVs and exposure values, is standard practice. The exposure value is either a

concentration (mg substance/kg media or mg substance/L water) or an estimated dose

(mg substance/kg body weight-day), and the TRV is either a concentration or an

estimated dose representing the threshold of a safe exposure. Thus, for each

contaminant and environmental medium, the HQ is expressed as the ratio of a potential

exposure level to the applicable toxicity-based benchmark. A complete discussion of

the HQ method is presented in Section 5.3.3. below. Decision rules are applied to the

results for interpretation of potential risks. For HQ values exceeding unity (1.0) the

potential for adverse effects to the receptor is concluded to be possible. In contrast, if

EM 200-1-4

31 Dec 10

5-2

the resulting HQ is equal to or less than unity, the potential for risks due to that

chemical can be considered negligible and therefore may be dropped from further

consideration of risk for that exposure pathway. The results of the SLERA can remove

COPECs, pathways, or even receptors from further consideration. This logic is

supported through the consistent application of conservative assumptions, biasing

towards overestimating potential risks. The other possibility is that the present

information available is insufficient to determine potential risks of exposure to the

chemical, and hence that chemical is retained for further review in the BERA.

5.2. Refinement of the SLERA. In circumstances where the results of the human

health assessment indicate lack of concern, it may be worthwhile to refine some

exposure parameters to evaluate less conservative exposures for the SLERA. This

decision would be based on the likelihood that more realistic exposure parameters

would bring the calculated HQs below one. This step (Step 3A) would occur prior to

beginning problem formulation for the BERA (Step 3 in the ERAGS process)(See

Chapter 3). For a complete discussion of the Step 3A process, see USA BTAG (2005a).

For this refinement, the following parameters can be reevaluated, as appropriate and

information is available, and HQs can be recalculated for those receptors and pathways

indicating the potential for ecological risk:

a. Area use percentage (home range).

b. Bioavailability < 100%.

c. Diet composition < 100% from the most contaminated media.

d. COPEC concentration in food items.

e. Detection frequency.

f. Comparison to background.

5.3. Risk Characterization – BERA. For the BERA, the risk characterization will

provide data on exposure and effects, analyzing the weight-of-evidence from various

studies employed for the risk assessment, and discussing the associated uncertainties.

The risk estimation consists of integrating the exposure and toxicity profiles, as well as

estimating and summarizing the associated uncertainties and assumptions (see

Chapter 6) to characterize current and potential adverse biological effects posed by the

COPECs. The potential impacts from all exposure routes and all media (water,

sediment, soil, and air) are included in this evaluation as appropriate. The risk

description provides information to the risk manager for interpreting the results and

identifies a threshold for adverse effects. The risk description can also include a

discussion of additional data or analyses that might reduce the uncertainty in the risk

estimates. These additional data collection efforts or analyses, if deemed necessary,

would be conducted in subsequent field efforts.

5.3.1. Risk Estimation. In a BERA, risk estimation can be either qualitative or

quantitative, depending on the data available, DQOs, and the stated level of effort.

EM 200-1-4

31 Dec 10

5-3

Quantitative risk estimation techniques can be fairly simple or more complex,

depending on the complexity of the food webs and exposure pathways that are to be

quantified. Other quantitative approaches that may be used include comparing

probabilistic distributions of effects, and exposure and simulation modeling.

5.3.1.1. Characterization of adverse effects on key receptor species at the

population, community, or ecosystem level is generally more qualitative in nature than

characterizing human risks. This is because the toxicological effects of most chemicals

are not well documented for most wildlife species. In the estimation and

characterization of risk, the adverse effects of chemicals on populations and habitats

should be considered rather than the effects on individual members of a species,

except in the case of threatened and endangered species, where individuals require

protection in order to preserve the population.

5.3.1.2. True risk estimation therefore also involves interpretation of results, with

professional judgment, to provide the ecological implication of the observations, made

at the level of the measurement endpoint. In some cases, this may involve a great deal

of professional judgment. In others, the ecological implications are either obvious or

inherent due to the level of the chosen measurement endpoint.

5.3.2. Terrestrial Ecosystem Methodologies. The following sections present

descriptions of two methodologies for performing quantitative risk characterization for

terrestrial and aquatic ecosystems. Methodologies for characterizing risk to receptors in

terrestrial and aquatic ecosystems are similar in some aspects, but are discussed

separately because of differences in the data forming the basis for the final risk

calculations.

5.3.3. HQ Method. The HQ method as applied to ERAs is similar to that for

calculating an HQ for human health risk characterization. The objective of a risk

characterization for a specific receptor is to compare the estimated chemical intake of

one chemical through one exposure route with the "threshold" concentration, that is, the

level of intake that is recognized as unlikely to result in adverse ecological effects (i.e.,

the TRV). The comparison (quotient) of estimated intake and acceptable exposure

level is called a HQ and is derived in the following manner:

HQ = Intake (mg/kg-bw/day)

TRV (mg/kg-bw/day)

where the intake is the chronic or sub-chronic daily intake (expressed as a dose in

mg/kg-bw/d) of the chemical (whichever is appropriate for the exposure being

assessed) and the TRV is the corresponding threshold value (sub-chronic or chronic,

oral) expressed as a dose. Short-term, sub-chronic, and chronic exposures should be

assessed separately.

5.3.3.1. There may be times when it is necessary and appropriate to examine the

potential for the occurrence of adverse ecological effects as a result of exposure to

multiple chemicals through multiple exposure pathways. In other words, even if

EM 200-1-4

31 Dec 10

5-4

exposure to each individual chemical is below its TRV (HQ less than 1), the sum of the

HQs for multiple chemicals may exceed unity, and adverse ecological effects could

occur. This is quantitatively derived in the following manner:

+ HQ

. . . . + HQ

= HI

where HQ

is the HQ for an individual chemical and HI

is the HI for a specific exposure

pathway. To derive an overall HI, considering multiple co-occurring exposure pathways

(and multiple chemicals), the following is performed:

+ HI

. . . . + HI

= Overall HI

5.3.3.2. Determination of when to calculate an HI requires an expert understanding

of toxicology and should be performed only by qualified individuals. Factors that need

to be considered include the critical toxicological effect upon which the TRV is based,

as well as other toxicological effects posed by the chemical at doses higher than the

critical effect. Major categories of toxic effects include neurotoxicity, developmental

toxicity, immunotoxicity, reproductive toxicity, and individual target organ effects

(hepatic, renal, respiratory, cardiovascular, gastrointestinal, hematological,

musculoskeletal, dermal, and ocular) (USEPA 1989d).

5.3.3.3. Deriving an overall HI using an additive approach assumes the following:

a. All chemicals will result in a similar adverse effect by the same mechanism of

action (or same target organ); and

b. Each chemical exerts its effect independently (i.e., there is no synergism or

antagonism).

5.3.3.4. Applying the assumption of additivity is a conservative approach that likely

overestimates the actual potential ecological risk presented by the exposure. However,

if the overall HI is greater than unity, consideration should be given to the known types

of adverse ecological effects posed by exposure to the chemicals. If the assumption of

additivity is not valid (i.e., if the chemicals most strongly contributing to the exceedance

of the HI display very different types of adverse effects), the HI should be segregated

according to toxicological endpoint. These segregated HIs may then be examined

independently.

5.3.3.5. HQs should be expressed to one significant figure only, because of the

uncertainties involved in deriving the TRVs. In addition, HQs should be reported in

decimal form (e.g., 0.001, not 0.0012 or 1x10

-3

5.3.3.6. It should be noted that an HQ is not a statistical value, nor a measurement

of risk. For example, an HQ of 0.01 does not indicate a one-in-one hundred probability

of the adverse effect occurring. Rather, it indicates that the intake is one hundred times

less than the TRV for the chemical. In addition, the HQ does not infer a linear

relationship, i.e., the hazards posed by exposure to the chemical do not increase

linearly as the HQ increases. This is so for several reasons, including the fact that

EM 200-1-4

31 Dec 10

5-5

TRVs are not precise descriptors of hazard, and the severity of potential ecological

effects varies with different chemicals (dose-response relationships differ). It is also

important to note that HQs are not directly population-based (although some measures,

such as reduced fercundity, can more clearly infer potential effects to populations). The

TRV is based on production of effects in individuals under laboratory conditions, which

may or may not extrapolate to the population or higher level of organization. Many field

studies have found levels of COPECs that have resulted in calculated HQs exceeding

several orders of magnitude. These unrealistically high values represent a major

limitation of the procedure, and are not a reflection of magnitude of the potential for

risks. For a more complete discussion of the limitations of the HQ method, see

Tannenbaum, et al, 2003.

5.3.4. Probabilistic Methodologies. A point estimate approach to risk assessment

is typically employed for work under CERCLA, where single values are used to

represent variables in the risk equations. The output, then, is a point estimate of risk,

which can be a CTE or RME, depending on the input values used. A probabilistic risk

assessment (PRA) uses probability distributions for one or more variables in the risk

equations to quantitatively characterize variability and/or uncertainty. The output of the

PRA is a probability distribution of risks, reflecting the combinations of the input

distributions. Monte Carlo Analysis is the most widely used probabilistic method in

PRA. It uses computer simulation to combine multiple probability distributions in a risk

equation.

5.3.4.1. A risk assessment performed using probabilistic methods is very similar in

concept and approach to the point estimate method, with the main difference being

incorporation of variability and uncertainty into the risk estimate. For some sites, PRA

can provide a more complete characterization of risks and uncertainties in the risk

estimates as opposed to the point estimate method. If a PRA is conducted, all

assumptions and inputs to the model need to be documented such that the results can

be independently evaluated.

5.3.4.2. In order for a PRA to be effective, the risk assessor must be familiar with

the distinction between variability and uncertainty. Variability reflects the true

heterogeneity or diversity of a population. Uncertainty results because of a lack of

knowledge. Uncertainty can be reduced by collecting additional data, whereas

variability can be better characterized, but cannot be reduced or eliminated. Efforts to

clearly distinguish between variability and uncertainty are important for both the risk

assessment and for risk communication.

5.3.4.3. For a PRA, inputs to the risk equation are described as random variables

and can be defined mathematically by a probability distribution. For continuous random

variables, the distribution may be described by a probability density function, and for

discrete random variables, the distribution may be described by a probability mass

function. The key feature of these functions is that they describe the range of values

that the variable may assume and indicate the likelihood of each value occurring within

that range.

EM 200-1-4

31 Dec 10

5-6

5.3.4.4. A work plan should be developed for review prior to beginning the PRA.

The work plan should document the decisions made during problem formulation, the

software to be used, the exposure routes, the models to be used and the probability

functions and their bases, including appropriate references. For human health risk

assessments, the USEPA allows probability distributions for exposure parameters only.

In ERAs, however, the USEPA allows distributions to reflect both exposure parameters

and toxicity information (USEPA 2001b).

5.3.5. Aquatic Ecosystem Methods. The HQ and probabilistic quantitative methods

can also be used for the estimation of risk to aquatic ecological receptors. The primary

difference between aquatic and terrestrial receptors is that contaminant concentrations

in surface water or sediments are used as input to the calculations instead of body

weight-based dose concentrations.

5.3.5.1. For calculation of an aquatic HQ, the comparison of a measured

concentration in water or sediment with an appropriate aquatic TRV is as follows:

HQ =

Aquatic TRV (mg/L)

Measured Concentration (mg/L)

where the measured concentration may be the overall RME concentration, maximum

concentration, or other appropriate measurement of exposure concentration and the

aquatic TRV is the AWQC, sediment criteria (units would be mg/kg), or a species-

specific TRV.

5.3.5.2. HIs for multiple chemicals and multiple exposure pathways are the sums of

individual HQs and pathway-specific HIs, respectively. It is only appropriate to sum the

HQs for contaminants with the same toxic effect mechanisms (e.g., PAHs). As was

noted for terrestrial systems, above, determination of when to calculate an HI requires

an expert understanding of toxicology and should be performed only by qualified

individuals.

5.3.5.3. Probabilistic methods can also be used to estimate aquatic risk. Instead of

using exposure concentrations in soils or forage, however, probability distributions of

chemical concentrations in surface water or sediments are used. Comparisons of

measured chemical concentrations can be made to probability distributions or point

estimates of aquatic TRVs.

5.3.5.4. A number of other potential quantitative methods are available for use with

aquatic receptors. In fact, nearly all of the ecological evaluation techniques previously

listed are applicable to aquatic receptors.

EM 200-1-4

31 Dec 10

5-7

Per USEPA (2001b), the following requirements must be met for the PRA,

ensuring that adequate supporting data and credible assumptions have been used

throughout:

The purpose and scope for the assessment should be clearly articulated in a

problem formulation section, including a full discussion of any highly exposed

or highly susceptible subpopulations evaluated. A ssessment endpoints must

be well defined.

The methods used for the analysis are to be doc umented and eas ily located

in the report. Sufficient information is to be provided to allow the results to be

independently reproduced.

The results of sensitivity analyses are to be pr esented and di scussed in the

report.

The presence or absence of moderate t o s trong c orrelations or dependen -

cies between the input variables is to be di scussed and ac counted for in the

analysis, along with the effects these have on the output distribution.

Information f or eac h i nput and out put di stribution i s t o be pr ovided in the

report. Selection of distributions i s t o be ex plained and j ustified. V ariability

and uncertainty are to be di fferentiated w here pos sible, f or bot h i nput and

output distributions.

The numerical stability of the central t endency and t he hi gher end of t he

output distributions are to be presented and discussed.

Calculations of ex posures and r isks us ing det erministic methods are to be

reported i f possible. T his w ill a llow c omparisons bet ween t he probabilistic

analysis and past or screening-level risk assessments.

Since f ixed ex posure as sumptions ar e s ometimes em bedded i n the toxicity

metrics, the exposure estimates f rom the probabilistic output distribution are

to be aligned with the toxicity metric.

Several computer-based proprietary simulation programs are available with

which to conduct this simulation. Performance of a Monte Carlo simulation should

only be performed by professionals with an understanding of the assumptions and

limitations of using it, including such factors as identifying the appropriate number

of runs and correlated input variables.

EM 200-1-4

31 Dec 10

5-8

5.3.6. Weight of Evidence. In the characterization of ecological risk, the information

collected concerning the identified hazards, the receptors, and the exposure

characterization are integrated through a comprehensive ecotoxicological evaluation of

source-receptor exposure pathways.

5.3.6.1. After identifying sensitive receptors and habitats, complete exposure

pathways, exposure points, and COPEC exposure point concentrations, the potential

for impacts is evaluated either quantitatively, qualitatively, or a combination of the two.

Results from a variety of measurement techniques, such as toxicity tests and HQs, may

be used in the weight-of-evidence characterization of potential and actual ecological

risk.

5.3.6.2. If actual or potential adverse impacts are found, those impacts are further

evaluated to determine to what extent they are site-related and to determine

appropriate remediation goals. The ERA also includes conclusions regarding impacts

from site chemicals, and a qualitative evaluation of limitations and uncertainties

associated with those conclusions.

5.4. Risk Description. A key to risk description is documentation of environmental

contamination levels that bound the threshold for adverse effects on the assessment

endpoints (USEPA 1997a). Of most importance, however, is the ability to convey to the

risk manager the likelihood of adverse effects at the site and the ecological significance

of those effects.

5.4.1 Factors Influencing Ecological Significance. The relative significance of

different effects may require further interpretation, especially when changes in several

assessment or measurement endpoints are observed or predicted. If the ERA is

concerned with adverse impacts on a variety of receptors and different ecosystems,

qualitative discussions should be presented as to the nature and magnitude of the

potential adverse effects associated with each receptor and ecosystem.

5.4.1.1. The spatial and temporal distributions of the effect provide another

perspective important to interpreting ecological significance (USEPA 1998a). Adverse

effects to a resource that is small in scale relative to the site and/or area of

contamination (e.g., a wetland or nesting grounds), may have a small spatial effect, but

may represent a significant degradation of the resource because of its overall scarcity.

Recovery potential is another factor influencing ecological significance that may need to

be considered depending on the assessment endpoints.

5.4.2 Interpreting Site-Wide Ecological Significance. It is often the case at large

federal facilities that individual chemicals and ecological receptors are not isolated in

the environment, and adverse effects are not necessarily related to a limited number of

chemicals confined to the immediate location of discharge. Organizing the ERA to

interpret the ecological significance of various chemicals to which a variety of ecological

receptors are exposed at sometimes distant locations is challenging.

5.4.2.1. Matrix ranking processes may be subjective or quantitative (depending on

data availability) based on site characterization, ecotoxicological information, and

EM 200-1-4

31 Dec 10

5-9

USEPA guidance. The ranking process may incorporate weighing factors to emphasize

specific factors (e.g., area use, toxicity, exposure area, bioavailability, and

biomagnification potential) which affect the ability of the chemicals considered to have a

deleterious impact on the ecological receptors.

5.4.2.2. Matrices can be updated or revised during the risk assessment process

should additional data regarding the COPECs, exposure pathways, or key receptors be

identified. The additional data will enhance risk decisions for smaller locations within

the facility (e.g., OUs/SWMUs) for which the risk assessment process has not been

completed.

EM 200-1-4

31 Dec 10

5-10

THIS PAGE INTENTIONALLY LEFT BLANK

EM 200-1-4

31 Dec 10

6-1

CHAPTER 6

Uncertainty in Ecological Risk Assessments

6.1. Introduction.

6.1.1. EPA has two requirements for the full characterization of risk (EPA 1995a,c).

First, the risk characterization should address qualitative and quantitative features of the

assessment (see Chapter 5). Second, it should identify the important strengths and

qualitative as well as quantitative uncertainties in the assessment as part of a discussion

of the confidence in the assessment. In the uncertainty discussion, the risk assessor

should also try to distinguish between variability and uncertainty. Variability arises from

true heterogeneity in characteristics such as dose-response differences between species

and individuals, or differences in contaminant levels in the environment. Uncertainty, on

the other hand, represents lack of knowledge, or data gaps, about factors such as adverse

effects of select contaminants on select species.

6.1.2. Particularly critical to full characterization of risk is a clear and open discussion

of the uncertainty in the overall assessment and in each of its components. The

discussion of uncertainty should highlight those uncertainties, which would tend to reduce

the degree of confidence in the conclusions drawn and therefore lessen confidence that

the site can pose no threat whatsoever. A discussion of uncertainty requires comment on

such issues as the quality and quantity of available data, gaps in the database for specific

chemicals, quality of the measured data, use of default assumptions, incomplete

understanding of general biological phenomena, and scientific judgments or science policy

positions that were employed to bridge information gaps (EPA 1995c).

6.2. Characterization of Uncertainty – SLERA. In the SLERA, conservatism and therefore

uncertainty is introduced to insure that the potential for risk is not overlooked. When the

potential for risk is shown, the extent of the exceedance of the screening criteria, and the

appropriateness of the screening value itself, help clarify uncertainty and should be

evaluated as part of the initial screen decision-making process. As a minimum

requirement, the potential effect of the following uncertainty factors should be discussed:

a. Uncertainties associated with the (limited) chemical database for the site

(availability of site-specific data for medium of concern);

b. Use of the maximum chemical concentration (instead of the 95% UCL) for

representing the exposure point concentration;

c. Use of surrogate or generic receptors and worst-case exposure scenarios; and

d. Use of screening criteria and the assumptions associated with those values.

As was noted above, uncertainties in the SLERA are essentially associated with the

conservative nature of the assessment itself. Step 3A (see Chapter 3) addresses these

uncertainties by allowing the risk assessment to adopt parameters more representative of

actual exposures. This refinement step also introduces uncertainty by minimizing the

conservatism of exposure parameters, allowing the possibility of underestimating site risks.

EM 200-1-4

31 Dec10

6-2

6.3. Characterization of Uncertainty – BERA. In a BERA, uncertainty is usually presented

as a qualitative discussion about the range of confidence in the risk estimation (i.e., low,

medium, or high) accompanied by the factors that may contribute to an overestimation or

underestimation of risk. Wherever possible, risk should be expressed in terms of

magnitude, direction (over or underestimation), and probability, using either a sensitivity

analysis (examining the appropriateness of the risk estimation by maximizing one or more

input variables) or a probabilistic analysis. By expressing risk in quantitative terms of

probability, plus magnitude and direction, the risk manager is better enabled to make

judgments on risks relative to other factors (such as costs), and not simply decide that

uncertainty levels in the risk assessment must be reduced by further study.

6.3.1. Objectives. This section discusses methods of identifying and describing

uncertainties in a BERA. Full disclosure and clear articulation of risk uncertainties are

guiding principles for this portion of the risk assessment (EPA 1992d, 1995a,c).

"EPA r isk as sessors and managers n eed t o b e c ompletely c andid a bout

confidence and un certainties in describing risks and in explaining regulatory

decisions. S pecifically, the Agency's risk assessment guidelines call for full

and o pen d iscussion o f u ncertainties i n the b ody o f each EPA r isk

assessment, i ncluding prominent display o f c ritical u ncertainties i n the r isk

characterization. N umerical r isk e stimates should always be a ccompanied

by de scriptive i nformation c arefully s elected t o e nsure a n objective an d

balanced c haracterization of r isk i n r isk a ssessment r eports and r egulatory

documents." (EPA 1992d)

6.3.1.1. Identification and discussion of uncertainty in an assessment is important for

several reasons (EPA 1992d):

a. Information from different sources carries different kinds of uncertainty, and

knowledge of these differences is important when uncertainties are combined for

characterizing risk.

b. Decisions must be made on expending resources to acquire additional

information to reduce uncertainties.

c. A clear and explicit statement of the implications and limitations of a risk

assessment requires a clear and explicit statement of related uncertainties.

d. Uncertainty analysis gives the decision-maker a better understanding of the

implications and limitations of the assessments.

6.3.1.2. The output from the uncertainty analysis is an evaluation of the impact of the

uncertaintires on the overall assessment and, when feasible, a description of the ways in

which uncertainty could be reduced.

6.3.2. Sources of Uncertainty in a Risk Assessment. Sources of uncertainty in a risk

assessment exist in almost every component of the assessment. Uncertainty generally

can arise from two main sources: variability and data gaps. Model error is an additional,

EM 200-1-4

31 Dec 10

6-3

potential main source of uncertainty that a risk assessor may encounter. Uncertainty from

variability can enter a risk assessment through random or systematic error in

measurements and inherent variability in the extent of exposure of receptors. Uncertainty

from data gaps is most prominently seen when numerous approximations are made

regarding exposures, chemical fate and transport, intakes, and toxicity.

6.3.2.1 Sampling Uncertainty. The identification of the types and numbers of

environmental samples, sampling procedures, and sample analysis all contain

components that contribute to uncertainties in the risk assessment. Decisions regarding

the scope of sampling and analysis are often made based on the ECSM developed at the

planning stages of the investigation. While appropriate planning may minimize the

uncertainty associated with these components, some uncertainty will always exist,

because the "real" state of the site is unknown prior to sampling and, in fact, may not be

fully elucidated even after sampling.

6.3.2.1.1. Some of the assumptions in this component that contribute to uncertainty

in the assessment include:

a. Media Sampled. Unless a decision has been made to sample all media, often a

subset of media is selected for sampling and analysis. This selection is usually based

upon the anticipated presence of a chemical in a medium from the site history and the

chemical's chemical and physical properties and may not include consideration of potential

transport through biological media. If all abiotic media in which a chemical is actually

present have not been sampled, appropriate risks may not be described.

b. Locations Sampled. The type of sampling strategy selected may impact the

uncertainty associated with the results. For example, purposive sampling (sampling at

locations assumed to contain the chemicals) will likely result in a higher frequency of

chemical detection and concentration than random sampling or systemized grid sampling.

Therefore, use of the results may skew the assessment towards greater assumed

exposures.

c. Number of Samples. Fewer samples result in a higher degree of uncertainty in

the results. This is demonstrated in the summary statistics, specifically the 95% UCL, in

which the statistical descriptor ("t" or "H" value), and hence the 95% UCL, increases with a

smaller number of samples. Planning for and success in obtaining a specific number of

samples to reach a specific degree of statistical confidence can limit the degree of

uncertainty.

d. Sampling Process. The sampling process itself can contribute to uncertainties in

the data from a number of factors, including sampling contamination (cross-contamination

from other sample locations, introduction of chemicals used in the field); poorly conducted

field procedures (poor filtering, incomplete compositing); inappropriate sample storage

In the following sections, specific sources of uncertainty in a risk assessment are

identified and discussed. Following this discussion, different approaches to

conducting an uncertainty evaluation are presented.

EM 200-1-4

31 Dec10

6-4

(head-space left in containers of volatile sample containers, inappropriate storage

temperatures); sample loss or breakage; and other factors. Some of these factors can be

controlled, however, planning does not prevent the occurrence of sampling errors.

e. Analytical Methodology. The analytical methodology can contribute to uncertainty

in a number of ways, including the scope of the chemicals analyzed (if analysis of all

important chemicals was not performed); the detection or quantitation limits applied (if not

sufficient); and limitations in the analysis due to matrix effects, chemical interferences,

poorly conducted analyses, or instrumentation problems. Some of these factors can be

addressed in up-front planning (such as selection of the analytical method); others cannot

(e.g., instrumentation problems).

f. Stochasticity. Natural variability is a basic characteristic of ecological systems, as

well as the factors which influence such systems (e.g., weather). Of all the contributions to

uncertainty, stochasticity is the only one that can be acknowledged and described but not

reduced (Suter in EPA 1992a).

6.3.2.2. Uncertainty in COPEC Selection. Evaluation of the data to select COPECs for

the ERA may result in uncertainties. Application of selection criteria may inadvertently

result in the inappropriate exclusion or inclusion of chemicals as COPECs. Improper

inclusion or exclusion of chemicals can result in an underestimation (if inappropriately

removed) or overestimation (if inappropriately retained) of potential ecological risks.

Uncertainties associated with the selection criteria include the following:

a. Background Comparison. If background measurements are not truly

representative of background conditions, chemicals may be inappropriately retained or

removed from the list of COPECs.

b. Sample Contamination. Uncertainty in the assessment can occur if chemicals are

not recognized as being present as a result of sampling or laboratory introduction and are

excluded as COPECs.

c. Frequency of Detection. Use of detection frequency (say, over 5%) as a selection

criterion may result in the inappropriate exclusion of chemicals as COPECs.

d. Toxicity/Concentration Screening. Removal of chemicals as COPECs as a result

of using a toxicity/concentration screen can result in uncertainty in the assessment, since

some chemical contributors to the risk (even if not significant) have been removed.

6.3.2.3. Uncertainty in Selecting Key Receptors. It is possible that the wildlife

selected as key receptors in an ERA are not those receptors that have the greatest

likelihood of being at risk or are sensitive to a particular chemical. Reptiles and

amphibians are typically not addressed in ERAs, as exposure and toxicity information on

which to base an assessment are generally lacking. Ecosystem and community level

assessment endpoints such as adverse impacts to nutrient cycling, predator-prey

relationships, community metabolism, and structural shifts are typically not addressed in

ERAs. Uncertainty is associated with the professional judgement used in the selection of

key receptors.

EM 200-1-4

31 Dec 10

6-5

6.3.2.4. Uncertainty in the ECSM. The ECSM is the product of the problem

formulation phase, which in turn, provides the foundation for the effects characterization

and risk estimation. If incorrect assumptions are made during development of the ECSM

regarding the potential toxic effects or the ecosystems and receptors potentially impacted,

then the final risk characterization may be seriously flawed.

6.3.2.5. Uncertainty in Exposure Assumptions. Numerous assumptions regarding

the amount of chemical intake by a receptor are commonly made as part of the exposure

characterization. Such exposure estimates are associated with a number of uncertainties

that relate to the inherent variability of the values for a given parameter (such as body

weight) and to uncertainty concerning the representativeness of the assumptions and

methods used. Uncertainties associated with chemical intake and exposure include:

a. Potential Exposure Pathways. Potential exposure pathways are identified by

examining the current and future land uses of the site and the fate and transport potential

of the COPECs. While current land use and potential exposure pathways are often easy

to identify, potential future uses can only be inferred from information available at the

current time. For many ERAs, potential future land use is assumed to be the same as

current land use. This and any assumption regarding future land use, any potential future

migration of contaminants off-site, and exposure pathways will add uncertainty to the

assessment.

b. Potentially Exposed Receptors. As discussed in the preceding bullet,

identification of potentially exposed receptors is based upon information currently

available. Assumed exposed receptors under future use scenarios can only be guessed

at, and this adds uncertainty to the assessment.

c. Exposure and Intake Factors. Point values (e.g., maximum or 95% UCL) for

exposure estimates are commonly used in risk assessments rather than a distribution of

exposure values that describe the distribution of exposures. These point values are

usually conservative, and their use results in introduction of conservatism into the risk

assessment that should be addressed. Use of average (i.e., central tendency), rather than

upper-end exposure and intake factors may underestimate potential health risks, since

only half the population is exposed to that degree or less; the other half is exposed to a

greater degree. Using average values, therefore, also contributes to uncertainty that

should be addressed in the assessment.

Food and soil/sediment intake values for most wildlife are either unknown or highly

variable and very site-specific. Food and sediment intake values for key receptors may be

derived from allometric equations. Determining chemical concentrations in food may

require the use of bioconcentration or bioaccumulation factors. Uncertainty exists in the

use of such equations and factors.

d. Exposure Point Concentrations. Exposure point concentrations may be derived

either from measured site media chemical concentrations alone or in combination with fate

and transport modeling. With regard to estimating exposure point concentrations from

sampling data alone, use of 95% UCL and mean concentrations is associated with some

degree of uncertainty. The 95% UCL concentration is used to limit the uncertainty of

EM 200-1-4

31 Dec10

6-6

estimating the true mean concentration from the sample mean concentration. This value

may overestimate the true mean concentration. Use of the sample mean concentration

may under - or overestimate the true mean concentration.

Application of fate and transport modeling adds an additional tier of potential

uncertainty to exposure point estimates. Models cannot predict "true" exposure point

concentrations at different times and places or in different media, but provide an estimate

of the potential concentration under certain assumptions. Often, the assumptions used in

the models are conservative to avoid underestimating potential concentrations. In

addition, not all applicable processes are or can be considered (e.g., degradation, removal

processes).

6.3.2.6. Uncertainty in Toxicity Values. TRVs are developed from literature

benchmark values by applying conservative assumptions, and are intended to protect

sensitive species or populations. Use of non site-specific, generic TRVs will usually result

in overestimates of potential risk. Factors that contribute to uncertainty include:

a. Use of uncertainty factors (UFs) in the TRV

. TRVs are primarily derived from

laboratory animal toxicity studies performed at high doses to which UFs of 10 or more are

applied.

b. Choice of Literature Benchmark Study to Derive an TRV. The inclusion or

exclusion of studies in the derivation of a TRV is usually made by professional judgement;

this affects the numerical TRV value.

c. The Assumption of the Most Sensitive Species. When deriving TRVs, the animal

study showing an adverse effect at the lowest exposure or intake level, is often the basis

for deriving the TRV. EPA assumes that wildlife receptors are at least as sensitive as the

most sensitive laboratory animal used (toxicological data on wildlife are still very limited).

The LD

dietary studies probably give a better indication of the toxicity of the chemical

tested than LD

studies, while NOAELs from longer studies are the best (still imperfect)

laboratory studies to use as predictors of field effects. The potential exists for wildlife

species to be more or less sensitive than test species (some biota can adapt) and the

toxicological benchmarks used. Various uncertainty factors may be used to account for

differences in taxonomic levels (i.e., species, genus, order, family) between the test

species for the TRV and the key receptor(s) under consideration.

d. Exposure Duration. Actual exposure durations for key receptors may or may not

exceed the test duration periods on which the toxic literature benchmark value and

resultant TRV are based. Because mobile receptors are likely to feed or visit several

locations, or avoid contaminated areas, their daily dose, if averaged over time, could be

less than that used for evaluating risk. Unless exposure modifying factors are used, risk is

likely to be overestimated.

6.3.2.7. Additional Uncertainties. Standardized algorithms to calculate chemical

intakes and associated risks are generally lacking for many wildlife receptors. There are

numerous assumptions inherent in use of such equations that add uncertainty to the

assessment. These include:

EM 200-1-4

31 Dec 10

6-7

Assumption of Additivity. Calculation of HIs assumes (at least as a first line

approach) additivity of toxic effects. This assumption adds uncertainty to the assessment,

and may result in an overestimate or underestimate of potential risks, depending on

whether synergistic or antagonistic conditions apply.

Omission of Certain Factors. Exposure modifying factors, such as absorption,

bioavailability, soil matrix effects, area use, and exposure frequency should be considered.

In cases where these processes are important, use of a standard algorithm without

modification may result in an overestimation of potential chemical intakes.

6.3.3. Level of Effort. Various approaches can be applied to describe the

uncertainties of the assessment, ranging from descriptive to quantitative. The method

selected should be consistent with the level of complexity of the assessment. It may be

appropriate to conduct an in-depth quantitative evaluation of uncertainty for a detailed,

complex assessment, but may not be appropriate or even needed for a screening level or

simplistic assessment. In the section below, qualitative and quantitative approaches to

expressing uncertainty are discussed.

6.3.3.1. Qualitative Evaluation. A qualitative evaluation of uncertainty is a descriptive

discussion of the sources of uncertainty in an assessment, an estimation of the degree of

uncertainty associated with each source (low, medium, high), and an estimate of the

direction of uncertainty contributed by that source (under- or overestimation). A qualitative

uncertainty assessment does not provide alternate risk values, but provides a framework

in which to place the risk estimates generated in the assessment.

6.3.3.2. Quantitative Evaluation. A quantitative uncertainty assessment is any type

of assessment in which the uncertainty is examined quantitatively, and can take several

forms. A sensitivity analysis is one form in which specific parameters are modified

individually and resultant alternate risk estimates are derived. Probabilistic approaches

are more complex forms of uncertainty analyses that simultaneously examine the

combined uncertainty contributed by a number of parameters.

6.3.3.2.1. A sensitivity analysis is the process of changing one variable while leaving

the others constant and determining the effect on the output. These results are used to

identify the variables that have the greatest effect on exposure. This analysis is performed

in three steps:

a. Define the numerical range over which each parameter varies;

b. Examine the relative impact each parameter value has on the risk and hazard

estimates; and

c. Calculate the approximate ratio of maximum and minimum exposures obtained

when range limits for a given parameter are applied to the risk algorithm. Exposure

parameters should not, however, be combined in ways that are not reasonable; for

example, combining maximum intake rates with minimum body weight.

6.3.3.2.2 A probabilistic uncertainty analysis, such as the Monte Carlo simulation,

examines the range of potential exposures associated with the distribution of values for

EM 200-1-4

31 Dec10

6-8

select or all input parameters of the risk algorithm. Probability density functions are

assigned to each parameter, then values from these distributions are randomly selected

and inserted into the exposure equation. After this process is completed many times, a

distribution of predicted values is generated that reflects the overall uncertainty of inputs

to the calculation. The results are presented graphically as the cumulative exposure

probability distribution curve. In this curve, the exposure associated with the 50th

percentile of the exposure may be viewed as the "average" exposure and those

associated with the 90th or 99.9th percentile may be viewed as "high end" exposure.

An example of this approach, Analysis of Extrapolation Error, is presented in Barnthouse

et al. (1986).

EM 200-1-4

31 Dec 10

7-1

CHAPTER 7

Alternative Evaluation

7.1. Introduction. Various types of ERAs may be applied to conduct an evaluation of

remedial alternatives or a more detailed analysis of a selected alternative. Generally, the

SLERA procedures will be sufficient in providing the risk inputs for selection of potential

remedial alternatives or corrective measures (including the no-further action alternative)

or the need for procedural changes or engineering controls to minimize short-term risks or

residual risks. Use of more involved studies may be necessary for sites requiring

implementation of remedial action for a large areal extent and/or multiple years of

remediation, and sites with complex ecosystems and many effected trophic levels. Again,

early project planning with involvement of expert ecological risk assessors, BTAG or

ecological technical assistance group (ETAG) persons, regulatory agencies, and

stakeholders will be the key to identifying the level of effort most appropriate for specific

site conditions.

7.1.1. As part of FS activities, different remedial alternatives are examined from a

number of perspectives as part of the selection process. The NCP specifies nine

selection criteria to be examined as part of remedial alternative evaluation: (1) protection

of human health and the environment, (2) compliance with ARARs, (3) long-term

effectiveness and permanence, (4) reduction of toxicity/mobility/volume through

treatment, (5) short-term effectiveness, (6) implementability, (7) cost, (8) state

acceptance, and (9) community acceptance.

7.1.2. Generally, there are two applications of the ERA methodology useful for the

FS: 1) development of comparative risk assessments between different remedial options,

and 2) the development of remediation goals to be applied to site cleanup. The first type

of ERA is not commonly performed, but it can be useful in distinguishing the advantages

and disadvantages between potential remedial options. The second type is sometimes

performed as a component of the RI, but is distinguished in this section because of its

use in the development of remedial options. Each type of ERA is discussed individually in

the following sections.

7.2. Comparative Risk Assessment of Remedial Alternatives. For a remedial alternative

to be acceptable, it must be protective of the environment as well as human health.

However, more than one alternative may meet this (and the remaining criteria). An ERA

can evaluate the potential for long-term residual risks as well as the short-term risks

associated with the remedial action (i.e., habitat destruction or alteration).

7.2.1. Evaluation of Residual Ecological Threats. The potential risks to be

addressed in this type of comparative risk assessment are those remaining after the

implementation of the remedial alternative (those potentially incurred during the

implementation are discussed in Paragraph 7.2.2.). The methodology for performing the

comparative risk assessment is the same as for a SLERA. The potential exposure

pathways and receptors should also be the same unless exposure pathways would be

EM 200-1-4

31 Dec 10

7-2

modified due to habitat removal, for example. The main factor that will change is the

chemical concentration to which the key receptors may be exposed.

7.2.1.1. An assessment of the potential for long-term residual risks associated with

multiple alternatives can be developed as a tool to assist in selecting an alternative. By

comparing the degree to which an alternative reduces potential threats with respect to

other factors such as cost, acceptability, and effectiveness, one alternative may be

identified preferable. For example, Alternative A may reduce potential risks to an HQ of

well below 1, but cost $5 million to implement; Alternative B may reduce potential risks to

an HQ of slightly below 1, but cost only $1 million to implement. Since both risk (hazard)

levels are acceptable in terms of the assessment endpoint, it may be preferable to select

Alternative B because of its cost/benefit advantage.

7.2.1.2. The reduction of risk offered by the alternative should also be examined with

respect to the nature of the assessment endpoint and the size of the population affected

by the contamination. Although protection of all key receptors is the primary goal, a

modest reduction of risk for large populations of key receptors may be preferable to a

large reduction of risk for a small group of key receptors.

7.2.1.3. When developing an estimate of potential exposure point concentrations

after remediation, careful consideration must be given to where remediation is to take

place and where no action is anticipated. It is not uncommon for remedial actions to

focus in some areas of a site, leaving others untouched. Therefore, estimating the

potential exposure point concentration is not as simple as assuming exposure to the

remedial level, but to a combination of attaining the remedial level in some locations,

being below the remedial level at others, and perhaps exceeding the remedial level in

some isolated areas where remediation is not anticipated. The potential risks associated

with different combinations of remedial alternatives can be addressed by examining each

medium separately, and then combining the associated risks.

7.2.2. Evaluation of Short-Term Ecological Threats. Ecosystems are dynamic and

capable of recovery from many different types of assaults. Implementing a remedial

action that destroys habitat will require time for the receptors to reestablish and the

ecosystem to recover. Obviously, the more habitat destruction necessary, the longer

recovery will take. Therefore, from a purely ecological point of view, in situ technologies

are preferable to ex situ technologies, as habitat impacts are minimized.

7.2.2.1. Sometimes habitat destruction or alteration during implementation of a

remedial action can cause more problems for the environment than performing no

remediation at all. A key aspect often overlooked is the evaluation of the short-term

impacts to the environment from remediation for human health concerns. The receptors

effected and the extent of those effects from various alternatives can be significantly

different. Displacement of large numbers of key receptors can have greater impacts than

displacement of only a few key receptors.

7.2.2.2. The ERA for the FS should compare the short-term impacts of the various

alternatives in terms of habitat destruction, displacement of receptors and the benefits

EM 200-1-4

31 Dec 10

7-3

potentially realized. Although an ERA to evaluate these threats will most likely not be

quantitative, it can give a clear indication of which alternative presents greater short-term

threats, and can help determine if remedial action will provide the necessary long-term

benefits to the site.

7.3. Development of PRGs. Preliminary remediation goals (PRGs) for aquatic systems

may be derived by sorting and screening site-specific data on chemical concentration and

co-occurring bioeffects in a manner analogous to the derivation of ER-Ls, threshold

effects levels (TELs), and AETs. Remedial levels may also be derived by performing the

HQ calculation in reverse by rearranging the terms in the terrestrial or aquatic HQ

equations.

7.3.1. For calculation of a medium-specific PRG, the HQ is set equal to an

acceptable level (e.g., HQ = 1), the exposure route-specific intake factors developed

during the ERA are applied, and the chemical concentrations associated with the

ingestion factors and HQs are calculated.

7.3.2. PRGs are receptor-and chemical-specific. They should be based upon all key

receptors and all significant exposure pathways assessed in the ERA for that medium.

However, since the pathways resulting in the highest degree of risk will most greatly

influence the PRG, exposure pathways that have minimal contribution to overall risks can

be excluded from the remedial level development with little or no impact.

7.4. Development of Remediation Levels. Remediation (remedial) levels, which are not

synonymous with PRGs, are media-specific chemical concentrations that are associated

with acceptable levels of chemical exposure for the site-specific ecological receptors.

Remedial levels, also referred to as target cleanup levels, are considered along with other

factors, such as ARARs, in identifying chemical concentrations to which impacted media

may need to be remediated in order to achieve acceptable risk. Remedial levels differ

from PRGs in that site-specific factors are considered.

7.4.1. PRGs may be developed as a screening level tool prior to the performance of

an RI. Conversely, remedial levels are developed from the site-specific BERA, conducted

during the RI. Remedial levels are just one element of the weight of evidence the risk

assessment can provide to the risk manager to assist in remedial decision-making. Some

regulatory agencies recommend including the development of remedial levels as part of

the BERA in order to assist the risk manager in the decision-making process.

7.4.2. Establishing remediation levels is not without problems. If the only line of

evidence available is comparison of site concentrations to conservative benchmarks or

TRVs (e.g., HQ method), establishing the need to remediate is confusing, as HQs are not

measures of risks (see Section 5.3.3.). The comparison may show that exposures are

greatly exceeding TRVs (e.g., HQ much greater than one), yet indications of risks may

not be present in the field. Some have attached descriptors to HQs indicating various

levels of concern (i.e., an HQ between 1 and 10 is identified as a small potential for risks),

however, HQs are not linear measures, and an HQ of 10 is no more indicative of actual

impacts than is an HQ of 1. Recent procedures developed by the USEPA have applied

EM 200-1-4

31 Dec 10

7-4

the “rule of five,” which balances remedial decisions to the concentrations between a

NOAEL-based benchmark and a LOAEL-based benchmark. This, too, is not without

controversy, as laboratory studies of toxicity may not clearly translate to actual adverse

effects the field. Generally, if the only line of evidence is an HQ, establishing remedial

goals is not advisable.

EM 200-1-4

31 Dec 10

8-1

Chapter 8

Risk Management

8.1. Introduction

8.1.1. Consistent with NAS, USACE has developed the HTRW risk management

decision-making (RMDM) process. This process identifies factors to consider when

making decisions, developing and recommending options, and documenting of risk

management decisions (Figure 8-1, 8.2). The process establishes a framework to

manage risk on a site-specific basis. It emphasizes that risk management must consider

the strengths, limitations and uncertainties inherent in the risk assessment; the

importance of public and other stakeholders' input; and other non-risk factors.

. The NAS defines risk management as "a process of weighing

policy alternatives and selecting the most appropriate regulatory action, integrating the

results of risk assessment with engineering data and with social, economic and political

concerns to reach a decision" (NAS 1983). NAS has identified four key components for

managing risk and resources: public participation, risk assessment, risk management,

and public policy decision-makers (NAS 1994). Risk characterization is considered the

"bridge" or "interface" between risk assessment and risk management. USEPA

recommends that risk characterization should be clearly presented and separated from

any risk management considerations. USEPA (1995c) policy indicates that risk

management options should be developed using risk input and should be based on

consideration of all relevant factors, both scientific and non-scientific.

8.1.2. Risk and uncertainty are important factors to be considered in RMDM

(USEPA 1991a, 1995c). Other factors, including the customer's and stakeholders'

concerns, cost, schedule, value of resources to be protected, political, and technical

feasibility are also to be considered before selecting the best option for a project decision.

The consideration of risk is critical, since site actions are driven by statutes and

regulations, which explicitly require the "protection of human health and the

environment".

8.1.3. The HTRW risk management decision-making process can be represented

by the following equation, with many variables contributing to the final decision:

Therefore, selecting the proper risk tool and collecting data to assess

environmental risk is a primary responsibility of the PM and the risk assessor.

RM = f (X

, X

....X

)

where,

RM = Risk Management Decision

f = Function of

= Input variables (e.g., risk and uncertainty)

Examples of these requirements are 40 CFR 300.430(e)(1) of the NCP for deciding if remedial action is

needed for a CERCLA site.

EM 200-1-4

31 Dec 10

8-2

8.1.4. In addition to risk and uncertainty, there are many non-risk variables

influencing the risk management decision. The major ones are cost, schedule, value of

resources to be protected, competing priorities among sites managed by the customer,

economic, compliance/regulatory, political, and technical feasibility. A relatively sensitive

political and/or economic factor to be considered is "Environmental Justice or Equity".

This phrase relates to the government's initiatives to cleanup sites located in "poor and

disadvantaged" areas.

Figure 8-1. Inputs for Risk Management Decision-Making HTRW Project Decision Diagram.

Need for Further Action; PA/SI

(Has a release occurred?)

Need for Removal Action; the EE/CA HRA and Throughout Site

Process

(Time Critical: Is there an imminent health threat; Non-time Critical: Is the

removal action consistent with the final action or remediation strategy?)

Need for Remedial Action; the RI

(Is the baseline risk acceptable? What are the uncertainties? Are the

PRGs reasonable for screening of remedial alternatives?)

Risk and Non-Risk Variables to be Considered

(Risk and Uncertainty, Budget, Schedule, Competing Risk Reduction

Priorities, Compliance, Political, Economic, Societal Values of Resources to

be protected, Environmental Justice, and other Stakeholders' Concern)

Need for Mitigation of Short-Term Risks

Associated with Construction; RD/RA

(What is the exposure pathway of the risk? What are the uncertainties?

Will operational and institutional control or engineering modifications

mitigate risks?)

EM 200-1-4

31 Dec 10

8-3

What is the project decision for the project phase?

(Regulatory/Statutory Decision Statement)

What are the inputs/study elements into the decision?

(Comparison with health-based PRGs, screening risk assessment, baseline risk

assessment, risk analysis of alternatives, development of remedial action

objectives)

What are the anticipated options?

(Interim measures, removal actions, ARARs)

What are the risk and uncertainty?

(Reasonable maximum/high-end; average; population; and probabilistic risks)

What are other relevant non-risk factors?

(Uncertainty, Budget, Schedule, Competing Priorities, Compliance, Political,

Economic, and Societal Values of Resources to be protected, Environmental

Justice, and other Stakeholders' concerns.)

What are the options?

(An array of potential options and their ramification on the site decision)

What is the recommended option?

(and the rational for the recommended option)

Decision by the Customer and Document Rationale for Decision

Figure 8-2. HTRW Risk Management Decision-Making Process Flow Diagram

8.1.5. The risk assessment, in conjunction with other important "non-risk" decision

criteria, provides information on the need for remedial or early actions. Therefore, a clear

understanding of the risk assessment results and their uncertainties is essential.

Informed risk management decision-making will lead to protection of human health and

EM 200-1-4

31 Dec 10

8-4

the environment; cost saving; meeting the agreed schedule; political harmony; better

management of resources; and other social and economic benefits. The HTRW RMDM

process is consistent with recent initiatives by various USEPA officials: Habicht (USEPA

1992d); Denit (USEPA 1993h) and; Browner (USEPA 1995a), USDOD (1994a), and

various proposed legislations by the 104th Congress (e.g., Dole-Johnston Bill (S-343) and

HR 1022) that suggest the need for risk reduction based on "real world" or realistic risk

assessment, cost benefit analysis, and prioritization of environmental issues. The HTRW

risk management decision-making paradigm (Figure 8-3) presents an overview of this

process.

8.1.6. Prior to gathering data and performing the ERA, the PM defines the site

decision for the project phase, the required study elements (types of ERA or risk tools to

be used), and the potential uncertainties associated with the outputs of the study element.

Based on risk information and other considerations, the customer can select from an

array of recommended risk management options. Options can include gathering

additional data, recommending no further action, interim measures, or removal and/or

remedial actions. To facilitate RMDM, the USACE PM should anticipate potential risk

management options early in the project planning phase. Examples of the use of risk

assessment in various project phases include:

a. PA/SI: A screening risk assessment, environmental mapping, and an exposure

pathways analysis may be performed to determine the need for further investigations.

b. RI (prior to FS): The baseline ERA determines the need for the remedial action.

c. FS: Results of the ERA are used to develop preliminary remedial goals (i.e.,

chemical concentrations which pose acceptable hazard or ecological effects).

d. FS: Qualitative or quantitative risk assessments to compare and evaluate

potential ecological impacts from the remedial alternatives. A qualitative or simple

quantitative risk assessment (like those used in the baseline ERAs) may be conducted to

screen alternatives for their potential short-term and residual risks.

e. RD (prior to conducting RA): Detailed risk analysis may be performed to

determine if protective measures should be taken to minimize the impact to health and

the environment during remediation. For example, a toxicity assessment may be

conducted to evaluate the short-term acute, subchronic and chronic ecotoxicities of

potential releases from the remediation process. A hazard-response assessment should

also be conducted to determine the design measures to reduce the impact of non-

chemical stressors, e.g., habitat alteration and destruction, siltation, or other physical or

chemical changes in the environment caused by construction of the remediation.

8.1.7. This section describes how the results of risk assessment procedures are to

be used in risk management decision-making. The decisions include the need for further

investigation, removal and remedial actions, selection of remedy, and provision of

measures for designing removal or remedial actions that are protective of the

environment (Figure 8-1). Information provided by the risk assessment is a key for

selecting risk management options. Further, potential removal or remedial alternatives

EM 200-1-4

31 Dec 10

8-5

should be evaluated and compared according to their effectiveness to reduce site risks,

and any associated short-term risks posed by implementation of the alternatives.

Figure 8-3. HTRW Paradigm for Risk Management Decision-Making.

This chapter does not address comparative analyses of other environmental risks, i.e., risks from radon

gas, cigarette smoking, exposure to ultra violet light due to stratospheric ozone depletion, ingestion of

pesticide contaminated food products, etc. These risks, although they may be significant in terms of the total

risk posed to human receptors at a Superfund or RCRA site, are not related to HTRW site response actions

and are considered background risks which are addressed by other environmental laws and policies. This

chapter, however, does address the importance of risk assessment inputs in setting priorities for resource

management with respect to environmental cleanup under RCRA and CERCLA. In making site risk

management decisions, the project manager should be familiar with the statutory language/limitations

regarding the application of funds under the defense environmental restoration account (DERA), BRAC, and

other HTRW response actions.

RISK ASSESSMENT – GOOD

SCIENTIFIC PRINCIPLES

AND POLICIES

RISK MANAGEMENT – GOOD

SOCIAL, ECONOMIC, AND

REGULATORY POLICIES

COMMUNICATION WITH STAKEHOLDERS – IDENTIFICATION

OF VALUES TO BE PROTECTED, CONCERNS AND

COMMUNICATION OF INTELLIGBLE RISK ASSESSMENT

RESULTS

EXPOSURE

RESPONSE

ASSESSMENT

RISK

REDUCTION

OPTIONS

HAZARD

IDENTIFICATION

RISK

CHARACTERIZATION

AND UNCERTAINTY

RISK MANAGEMENT

DECISION

EXPOSURE

ASSESSMENT

NON-RISK

ANALYSIS

EM 200-1-4

31 Dec 10

8-6

8.1.8. It is important to recognize that risk managers often make difficult decisions

with considerable uncertainties in both risk and non-risk information. Therefore, a

focused and balanced risk approach is recommended that recognizes the reasonable

limits of uncertainty for the protection of human health and the environment as the

primary consideration, along with the considerations for non-risk issues. The risk

manager should clearly communicate the decision and the associated assumptions, and

document the basis for the decision. This chapter is organized to present the following

information:

8.1.9. Paragraph 8.2 describes how risk information can be used to support project

decisions at various project phases (e.g., determining whether the project should proceed

to the next phase or to site closeout). The section highlights key non-risk considerations

and emphasizes the importance of integrating the ERA results and uncertainties into an

overall risk management decision.

8.1.10. Paragraph 8.3 discusses the design considerations for implementing an

overall site remediation strategy. Such a strategy considers issues such as off-site

source areas, current and future land uses, compliance with chemical and site-specific

ARARs (USEPA 1989l) and verification of cleanup.

8.2. Determining Requirements for Action

8.2.1. Preliminary Assessment/Site Inspection. The purpose of PA/SI under

CERCLA is to identify if chemical releases have occurred, or if the site can be eliminated

from further action. The PAs are typically performed by the State, USEPA, or the federal

agency, and are generally preliminary in nature. Under some circumstances, federal

agencies may perform these activities with greater depth and vigor under Executive Order

12580. Unless good evidence exists that a site is contaminated, it is a crucial for the PM

to methodically review each identified site, area of contamination, and AOC, and decide if

these units should be eliminated from the next project phase. In addition, it may be

important to determine if an environmental threat or a substantial site risk potentially

exists that would require an early response action (e.g., non-time critical removal actions

or interim remedial action).

. The fundamental requirement associated

with any HTRW response action is the "protection of human health and the environment".

This requirement focuses on the acceptability of site risks from the potential actions.

Section 300.430 (d) and (e) of the NCP (USEPA 1990a) require a baseline risk

assessment or environmental evaluation to be performed to assess threats to the

environment. Risk management options are exercised in key phases of the HTRW

project life cycle (see Table 8-1). Risk information required to support a decision is

presented below:

8.2.1.1. Actual or Potential Release/Exposure. Under the PA/SI phase, the risk

management decision will be based on documented past spills and releases, the

likelihood of such spills/releases, the presence of endangered or threatened species,

sensitive environments or resources to be protected, and the existence of transport

mechanisms that could bring the chemicals in contact with these receptors.

EM 200-1-4

31 Dec 10

8-7

8.2.1.2. Potential Natural Resource Damage Assessment Action. Under CERCLA

Section 104(b)(2) and 107(f)(2)(C), the lead agency for clean-up (e.g., USD0D, USEPA)

must notify appropriate Federal and State trustees of natural resources for any

discharges or releases that may have injured natural resources under their jurisdiction.

The project manager is responsible for coordinating all response activities with the natural

resource trustees. The project manager should also consult with the U.S. Department of

the Interior (USDOI) (i.e., USFWS), the Department of Energy (USDOE), or Department

of Commerce (USDOC) where a discharge or release may adversely affect an

endangered or threatened species or result in destruction or adverse modification of the

habitat of such species. The trustees are responsible for assessing damages (i.e.,

monetary compensation) and presenting a "demand in writing for a sum certain" to the

potentially responsible parties. Although the PA/SI is an early project phase and the

potential for a NRDA action may not be known, the project manager and the risk assessor

should be cognizant of the potential when reviewing site history and background

information. Any findings with potential implications for NRDA uncovered in this process

should be provided to the customer and its legal counsel. This is recommended because

the customer's goals for site closeout may be different upon further review of the potential

for NRDA. By coordinating and working with Federal co-trustees, an overall remedial

action (which might include restoration or mitigation) can be devised which will reduce an

installation's NRDA liability.

8.2.1.3. Risk Screening and Prioritization of Units of Concern. Initial risk screening is

an important tool for ranking or prioritizing units (OUs). This tool can result in substantial

savings of resources, allowing the implementation of a more focused site investigation.

The risk screening results are likely to provide significant inputs into the risk management

decision-making for this project phase.

8.2.1.3.1. It is not uncommon to have tens or hundreds of "sites" within a site or

facility boundary. Risk managers at these facilities are faced with potentially complex

investigations. Rather than taking a "piece meal" approach of investigation, the list of

sites should be pared down if possible. The risk manager may negotiate with the

agencies and enter in the Interagency Agreement (IAG) or Federal Facility Agreement

(FFA) to permit the use of an approach that "addresses the worst sites first," and at the

same time, group OUs within the same ecological receptor exposure units or

geographical locations, as appropriate. This prioritization should result in the greatest

USEPA's Deputy Administrator (1994a) is concerned with the need for ensuring consistency while

maintaining site-specific flexibility for making remedial decisions (from site screening through final risk

management decisions) across programs. USEPA stresses that priority setting is reiterative throughout the

decision-making process because limited resources do not permit all contamination to be addressed at once

or receive the same level of regulatory oversight. USEPA suggests that remediation should be prioritized to

limit serious risks to human health and the environment first, and then restore sites to current and reasonably

expected future uses, whenever such restorations are practicable, attainable, and cost effective. USEPA

further suggests that in setting cleanup goals for individual sites, we must balance our desire to achieve

permanent solutions and to preserve and restore media as a resource on the one hand, with growing

recognition of the magnitude of the universe of contaminated media and the ability of some cleanup

problems to interact with another.

EM 200-1-4

31 Dec 10

8-8

environmental benefit with limited available resources. Site prioritization should include

the following:

a. Eliminate sites administratively by record review (including ascertaining if

endangered or sensitive species/environment or valued resources are present on site),

and/or interviews with current and former workers.

b. Conduct a site reconnaissance and group sites with common exposure pathways

or exposure units, if appropriate.

c. Rank the remaining sites or groups of sites qualitatively or quantitatively based on

the ECSM or a screening risk analysis.

8.2.1.3.2. Generally, the above tools will serve well if they are objectively and

uniformly applied. The use of site prioritization:

a. Provides justification for No Further Action (NFA) for low priority sites.

b. Allows better resource allocation for investigation of the remaining sites.

c. Provides the opportunity to develop ECSMs to guide data collection

(see Chapter 3).

d. Helps identify potential boundaries where the ecological receptors of concern are

to be protected.

e. Identifies high priority sites for non-time critical response actions.

8.2.1.3.3. USDOD’s (1994) Relative Risk Site Evaluation Primer recommends

evaluation based on three criteria: (1) contaminant hazard factor; (2) migration pathway

factor; and (3) receptor factor (Figure 8-4). Information generated from the initial

ecological risk screening can be used as a decision-making basis using a similar site

ranking process. Sites may be ranked high, medium or low based on non-quantitative

exposure pathway considerations such as the following:

(1) High Relative Risk: Sites with complete pathways (contamination in the media is

moving away from the source) or potentially complete pathways in combination with

identified receptor or potential receptors;

Significant Contaminant Levels

(2) Low Relative Risk: Sites with confined pathways (i.e., contaminants not likely to

be release or transported) and limited potential for receptors to exist; and

(3) Medium Relative Risk: Sites with characteristics not indicated in the above.

b. Moderate Contaminant Levels

(1) High Relative Risk: Sites with complete pathways or potentially complete

pathways in combination with identified receptor; or sites with complete pathways in

combination with potential receptors;

EM 200-1-4

31 Dec 10

8-9

(2) Low Relative Risk: Sites with confined pathways and any receptor types (i.e.,

identified, potential, or limited potential), or sites with potentially complete pathways in

combination with limited potential for receptors to exist; and

(3) Medium Relative Risk: Sites with characteristics not indicated in the above.

c. Minimum Contaminant Levels

(1) High Relative Risk: Sites with complete pathways in combination with identified

receptor;

(2) Medium Relative Risk: Sites with potentially complete pathways in combination

with identified receptor or sites with evident pathway in combination with potential

receptors; and

(3) Low Relative Risk: Sites with characteristics not indicated in the above.

8.2.1.3.4. The relative risk site ranking process may also be modified to include

consideration of the degree of confidence in the relative risk rating. Sites with a low

degree of confidence and a low relative risk may then be given a higher rating than sites

with a high degree of confidence and a low degree of risk.

THIS SPACE INTENTIONALLY LEFT BLANK

EM 200-1-4

31 Dec 10

8-10

Table 8-1.

The Potential Use of Risk Assessment Concepts/Procedures as a Risk Management Tool.

Project

Phase

Objectives

Risk Management

Options

Product/Deliverable

PA/SI

Should the site be

eliminated from further

evaluation

Identify sites with no

release or insignificant

release

Site ranking/prioritization

Need for Removal action

Need for RI

NO FURTHER ACTION

(NFA);

LIMITED

SAMPLING/VER.;

STAB, REMOVAL,

RESP;

LIMIT SCOPE OF RI;

PHASED RI SAMPLING

chemical fate and transport properties;

toxicity assessment (chemicals not expected to

pose an ecological concern;

environmental mapping (sensitive receptors and

food source identification); and

exposure pathway analysis/food web and use of

ECSM;

land use assessment

Does the site pose an

ecological risk?

Need for FS

NFA;

MONITORING;

INTERIM REMEDIAL

ACTIONS;

CONDUCT FS

baseline risk assessment

- comparison with published criteria or

benchmark toxicity values

- toxicity-based ERA to assess stress-response

relationship

Preliminary Remediation

Goals

Select remedial

alternatives

REMEDIAL ACTION

OBJECTIVES;

ON-SITE/OFF-SITE

MANAGEMENT;

NFA; MONITORING

development of site-specific PRGs or benchmark

toxicity values

assessment of short-term risks from remedial

alternatives

RD/RA

Protective control

measures/remedy

EFFECTIVENESS AND

DESIGN

BASIS FOR CONTROLS

TO REDUCE SHORT-

TERM RISKS

comparison with short-term acute risk levels;

exposure pathway analysis

identification of impact areas, traffic patterns,

and discharges

Delisting/

site

closeout

Residual risks/5-year

review, permit review

NFA; MONITORING;

RA;

ADDITIONAL FS AND

land use/pathway analysis

comparison with PRGs or RAOs

provide justifications for meeting cleanup

objectives or

technical impracticability

Legend:

Technical Impracticability = Technology not practical, e.g., remediation of groundwater aquifer

contaminated by Dense Non-Aqueous Phase Liquids (DNAPL);

EM 200-1-4

31 Dec 10

8-11

Figure 8-4. Flow diagram of relative risk site evaluation framework.

Figure 8-4. Flow Diagram of Relative Risk Site Evaluation Framework.

8.2.1.4. Risk Management Decisions and Options. Risk management decisions, risk

information needs, risk assessment tools to satisfy the information needs, and risk

management options are presented in this section. ("Non-risk" factors to be considered

in the decision-making are presented in Paragraph 8.2.4.)

Should a site be eliminated from further investigation in the RI project phase?

Risk Management Decision

Further Evaluation Needed

Risk Management Options/Rationale

Rationale: If a site cannot be justified for NFA, further evaluation (Expanded SI; Extent of

Contamination Study; RI) will be needed.

No Further Action (NFA)

Rationale:

a. Environmental mapping, functional group characterization, database search

published lists from natural resources agencies indicate that endangered species are not

present, and there are no sensitive environments or valued resources on and nearby the

site.

b. No knowledge of documented releases or major spills/low likelihood of

spills/procedures existed to promptly cleanup all spills.

c. Transport mechanisms do not exist, e.g., presence of secondary containment.

d. The substances released are not expected to be present due to degradation and

attenuation under the forces of the nature.

EM 200-1-4

31 Dec 10

8-12

e. Spills or releases have been addressed by other regulatory programs (e.g., the

Underground Storage Tank program or RCRA closure under Subpart G of 40 CFR 264 or

265).

f. The unit does not meet the definition of a "SWMU."

g. The unit is part of another identified unit or site, which will be addressed

separately.

8.2.1.4.1. Although risk assessment is traditionally performed in the RI phase of

HTRW response actions, risk assessment can assist the risk managers in all project

phases. Results of risk assessment activities are used to answer three key questions: 1)

whether or not there is a need to go forward with the next project phase, 2) whether or not

early response actions (removal actions, interim measures, or interim remedial actions)

should be taken to mitigate potential risks, and 3) effectiveness of the potential response

action and the short-term risks associated with implementation of the removal actions.

Providing an understanding of the usefulness of risk assessment in the HTRW removal

phase is the focus of this section.

Should early response action be undertaken to mitigate risk?

Risk Management Decision

No Early Response Action

Risk Management Options/Rationale

Rationale:

a. No imminent endangerment to ecological receptors of concern; lack of food

sources to support or attract ecological species, lack of endangered species or sensitive

environment/valued resources, low likelihood of exposure by the receptors. (Uncertainty

for the determination is related to thoroughness by the record search, visual observation,

or purposive limited sampling.)

b. Transport mechanisms probably do not exist, e.g., presence of secondary

containment.

Removal actions must be flexible and tailored to specific needs of each site and applicability i.e.,

complexity and consistency should be used in evaluating whether non-time critical removal actions are

appropriate. Examples of removal actions are: (1) sampling drums, storage tanks, lagoons, surface water,

groundwater and the surrounding soil and air; (2) installing security fences and providing other security

measures; (3) removing and disposing of containers and contaminated debris; (4) excavating contaminated

soil and debris, and restoring the site, e.g., stabilization and providing a temporary landfill cap; (5) pumping

out contaminated liquids from overflowing lagoons; (6) collecting contaminants through drainage systems,

e.g., French drains or skimming devices; (7) providing alternate water supplies; (8) installing decontamination

devices, e.g., air strippers to remove VOCs in residential homes; (9) evacuating threatened individuals, and

providing temporary shelter/relocation for these individuals (Superfund Emergency Response Actions,

USEPA 1990c). Items (3) through (5) could be used to reduce exposure to ecological receptors of concern.

EM 200-1-4

31 Dec 10

8-13

c. Low concentration of site contaminants or the levels measured probably do not

pose an acute hazard, and it is questionable whether the levels pose unacceptable

chronic risk or hazard.

d. There is no anticipated risk of stress or physical hazards.

e. Site contaminants are not likely to be persistent or the contaminants are relatively

immobile.

Early Response Action

Rationale:

a. There is no current impact, but if uncontrolled, the site could pose a substantial

threat or endangerment to humans or the environment. (Examples are: physical hazard,

acute risk from direct contact of the unit or site, or effluents or contaminated media are

continuously being discharged to a sensitive environment, e.g., a spill that could impact

salmon spawning, egg hatching, or survival of fry.)

b. The principal threat has reasonably been identified because of the evidence of

adverse impacts. In this context, the contaminants of ecological concern (COECs) are

known and the exposure pathways are judged to be complete, e.g., the exposure point or

medium has been shown to contain the COECs.

c. Due to the slow rate of degradation, excretion, or depuration, the potential COECs

may pose a threat to the food web via bioconcentration and biomagnification.

d. The boundary of contamination is reasonably well defined, so that removal

action(s) can be readily implemented.

e. There is a potential risk to ecological receptors or valued resources and the

removal or early response actions have been demonstrated to be highly effective in

reducing exposure to ecological receptors of concern, although candidate removal actions

may differ in terms of cost and magnitude of risk reduction achieved.

f. The early actions are consistent with the preferred final remedy anticipated by the

customer, reducing risks to both human and ecological receptors.

g. The response action will be used to demonstrate cessation or cleanup of

releases, resulting in substantial environmental gain which is the basis for early site close-

out or further investigation.

h. If removal actions are justified (e.g., addressing hot spots or high concentration

plumes discharging to a receiving body of water with sensitive aquatic species, food chain

or valued resources), the removal actions will then be evaluated for their potential short-

term risks and hazards, based on ECSM developed for the specific removal actions.

i. A high likelihood of releases and transport of site contaminants to the ecological

receptors of concern, e.g., runoff from the site is expected to reach a receiving body of

water containing endangered species or valued resources.

EM 200-1-4

31 Dec 10

8-14

j. High concentration (acute hazard level) of site contaminant is found in the

exposure medium.

k. Highly toxic chemicals or highly persistent and bioaccumulative chemicals found

on-site which may be transported off-site.

l. Documented unacceptable sediment, soils, surface water or groundwater seep

contamination in media that could be contacted by endangered species.

m. Ecological impacts have been observed due to volume of the release and the

habitat destruction of valued resources.

n. A high risk of physical hazards or stress to the environment.

o. The exposure pathway(s) for ecological species was one of reasons for the basis

for NPL listing or ongoing enforcement actions on spills or releases.

p. Non-complex site (no cost recovery issue, limited exposure pathways, small area

sites, etc.)

8.2.1.4.2. Early response actions or removal actions, consistent with the final

remedial action, may be taken to prevent, limit or mitigate the impact of a release. To

encourage early site closeout or cleanup, USEPA has encouraged early response actions

at sites where such actions are justified. To the extent possible the selected removal

actions must contribute to the efficient performance of long-term remedial actions.

USEPA's RCRA Stabilization Strategy (USEPA 1992h) and Superfund Accelerated

Cleanup Model (SACM) (USEPA 1992i) emphasizes controlling exposure and preventing

further contaminant migration. While these concepts are intended to expedite site

actions, risk assessment provides important information for justifying cleanup actions.

The applicable risk assessment methods include:

a. Environmental mapping/functional group assessment,

b. Exposure pathway analysis; development of ECSM,

c. Identifying short-term (acute) benchmark toxicity values for screening site data,

d. Qualitative evaluation of removal actions for their effectiveness to reduce

exposure to ecological receptors, and

e. For complex sites (sites with multiple pathways, without ARARs, large geographic

areas, and with a need for cost recovery), activities to support a baseline ERA may be

appropriate.

8.2.1.4.3. In order to allow input for the removal actions, the risk assessment should

be conducted in a timely manner. As an initial and highly conservative screening tool,

comparison of worst-case exposure point concentrations can be compared with short-

term (acute or subchronic) ecological benchmark values. Such risk evaluation should be

qualitative, simple, and concise.

EM 200-1-4

31 Dec 10

8-15

8.2.1.4.4. Early actions or accelerated cleanup can often be justified as long as the

actions are consistent with the preferred site remedy. Since remedies are generally not

selected until late in the FS, the customer's concept of site closeout and anticipated

action is critical for deciding which types of early actions are appropriate. Based on

experience gained in the Superfund program, USEPA has identified certain site types

where final remedies are anticipated to be the same (presumptive remedies). The current

list of presumptive remedies includes:

a. Municipal Landfill – capping and groundwater monitoring,

b. Wood Treatment Facility – soil and groundwater remediation,

c. Groundwater contamination with VOCs – air stripping/capture wells, and

d. Soil contamination with VOCs – soil vapor extraction.

8.2.1.4.5. Additional presumptive remedies are being developed by USEPA Region

7 for PCB sites, manufactured gas plants and grain fumigation silos. USEPA is

continuing to identify site types for which early actions are likely to result in substantial

environmental benefits. However, it should be noted that certain sites are not conducive

to early actions based on ecological concerns. Examples can include where: current and

future land use is highly industrial; there is a lack of food sources on site or nearby the

site for the ecological receptors of concern; there is low or generally low level, wide-

spread contamination; spilled or released substances are not bioavailable; contaminants

have short halve-lives or are anticipated to degrade rapidly under natural conditions; there

is a lack of viable environmental transport media (highly arid regions).

8.2.1.5. Qualitative Evaluation of Response Actions for Their Effectiveness to

Reduce Risks. Removal of hot spots can sometimes provide substantial improvements in

the site environment. In some cases, actions can reduce exposure to receptors

drastically, and allow natural attenuation to further reduce exposure point concentrations.

If removal actions are needed, the risk manager should request two types of risk

information. First, if there is more than one removal option, what is the comparative

effectiveness of the options to reduce exposure and risks? Second, what is the risk or

environmental impact associated with the proposed removal action? To answer the first

question, the HTRW risk assessor or risk manager provides information on how the

removal option can reduce risk or reduce the level of exposure both on-site and off-site, if

contaminant migration has occurred to off-site exposure points. If substantial risk

reduction can be obtained by all options, the risk manager should consider other factors,

such as effectiveness, reliability, etc. To answer the second question, the project

engineer estimates the destruction or treatment efficiency of the medium to be treated or

disposed, and the type/quantity of wastes or contaminated debris to be generated for

each potential option. This information is important if an action is likely to generate waste

or damage sensitive environments in the course of the remediation.

8.2.1.5.1. It is important to communicate and obtain an early buy-in of the removal

action from the local community. If the proposed removal actions are likely to pose

unacceptable short-term risks to on-site or off-site ecological receptors, the removal

EM 200-1-4

31 Dec 10

8-16

action should either be discarded or monitoring/control measures be instituted. (As

discussed later, the risk assessor and HTRW technical project planning team members

provide options for making decisions when there are divergent interests between the

protection of humans and the protection of ecological receptors of concern.) The risk

assessor should work with other project team members to evaluate the potential for

chemical releases or habitat destruction potentially associated with a remedial option.

These evaluations should be qualitative and not extensive, and can be based on a

consensus of professional judgment/opinion. These individuals should recommend

alternatives or precautionary/protective measures to the risk manager to mitigate any

potential risks.

8.2.2. Remedial Investigation. The primary objective of the RI or other equivalent

HTRW project phases is to determine if site contamination could pose potentially

unacceptable human health or environmental risks. Determination of unacceptable risk,

according to the NCP, is identified through a baseline risk assessment under

"Reasonable Maximum Exposure".

The ERA associated with the RI project phase can assist the risk management

decision-making process in the following ways:

a. The ERA presents the degree of site risk posed to ecological receptors and the

associated uncertainties. Risks are generally assessed based on individual effects,

although effects on populations and communities may be studied, as required.

b. Results of the ERA can be used to answer questions relating to the site decisions

on: 1) whether sufficient information exists to confidently eliminate a site as posing no

significant risk or there is a need to proceed to the next project phase; and 2) whether or

not removal actions are still appropriate and should be implemented to mitigate potential

ecological risks.

c. If a site poses unacceptable acute or chronic hazard to ecological receptors,

remediation will be needed for the significant exposure pathways. Pathways which do not

pose an unacceptable risk may be eliminated from further concern. Algorithms

developed in the ERA can be used in reverse to develop site-specific environmental-

based preliminary remediation levels in the FS.

d. If removal actions are still appropriate and are to be implemented, the short-term

impact of such actions should be evaluated.

Should remedial action be required based on the baseline ecological risk?

Risk Management Decision

Further Evaluation Needed

Risk Management Options/Rationale

Rationale: The ERA indicates unacceptable risk or the risk cannot be confidently

established, and therefore the customer has weighed all options and determines the

uncertainty associated with the ERA should be reduced. Further evaluation and/or data

EM 200-1-4

31 Dec 10

8-17

evaluation is needed to reduce uncertainty and determine ecological risk. Since risk

assessment is an iterative process, data used to support the risk estimates should be

critically reviewed by the PDT. The review may lead to the need for additional data to

more fully characterize potential risk. Alternatively, the PM may ask for a more detailed

analysis of uncertainties so that the decision for remedial action can be made.

Undertake Interim Response Action

Rationale: Action is based on finding of unacceptable risk to ecological receptors, after

giving consideration to the uncertainties associated with the ERA. The selected interim

remedial action or interim measure should be part of, or is consistent with, the final

anticipated remedy.

No Further Action (NFA)

The rationale for no action based on the ERA could be any (or a combination) of the

following:

a. Documentation that endangered species or sensitive environments are not going

to be impacted by the site due to the lack of complete exposure pathways, or the impact

is judged to be insignificant or acceptable by the risk assessor and/or expert

ecologist(s)/advisory panel such as BTAG/ETAG.

b. Lack of habitat or food sources to support the ecological receptors of concern and

potential off-site migration of site-related COECs to any nearby habitats or food-webs of

concern is negligible, or site land use will remain industrial/commercial based on

stakeholder's inputs.

c. The HQ is below unity or ten, as appropriate, based on uncertainty of the toxicity

data (or the frequency of exceedance of this point of departure value is low), given the

uncertainty inherent in the ERA involving multiple surrogate or indicator species

(measurement endpoints).

d. An existing ERA has been revised, reflecting that removal actions or interim

measures taken have substantially reduced the exposure to the level that the estimated

risks are acceptable.

e. The potential environmental risk or injuries associated with any and all remediation

is greater than the baseline risk (i.e., further efforts should be expended to find a suitable

remedial action or viable alternatives, such as off-site mitigation, restoration, or

compensation).

f. With source control in place, given natural attenuation of the COECs (based on

fate and transport properties), risk is expected to be short-term, and remediation is judged

to be cost-prohibitive.

g. There could be marginal risks, however considering uncertainties, the potential

incremental gain does not justify the action.

EM 200-1-4

31 Dec 10

8-18

h. No practical remedial action objectives or target cleanup levels can be established

to sufficiently document risk or such levels would be highly uncertain and the

environmental gain cannot be readily measured.

i. Potential remedy will cause substantial economic or scenic damage and is not

consistent with the public and stakeholders' goals and objectives.

j. Interim remedial action has removed the migration/transport mechanisms to

impact ecological receptors.

k. Site contaminants are not likely to ever pose unacceptable risk as they are not

persistent or the contaminants are relatively immobile and not bioavailable.

Remediation/Removal Action Required

a. The requirements for removal action taken at the RI/FS project phase is the same

as that described under Paragraph 8.1. above. Upon completion of RI/FS (and before

signing of the Superfund Records of Decision), a decision will be made whether remedial

action should be required. If there are site ARARs, such as State water quality standards,

remediation will be required unless an ARAR waiver is successfully completed. From the

risk assessment standpoint, if the baseline ERA is valid and the uncertainty deemed to be

acceptable, requirements for remediation for part of, or the entire site, will be based on

the following considerations:

b. Endangered species or sensitive environments/valued resources such as viable

wetlands or wildlife refuge, could be impacted by the site, and the estimated risk is judged

to be significant or biologically relevant.

c. There is viable habitat and sufficient food sources to sustain the ecological

receptors of concern.

d. The COECs are persistent or bioaccumulative and will potentially impact ecological

receptors of concern.

e. The site poses an unacceptable risk.

f. The environmental risk associated with the remedial action is acceptable.

g Short-term impacts from remediation, although potentially severe, are not

permanent and outweigh the alternative of long-term, chronic exposure.

h. COECs are persistent and expected to pose a long-term threat to the ecological

receptors of concern.

i. The RAO or target cleanup level (TCL) is based on a reliable or adequately

characterized exposure-response relationship and is practical for use to verify cleanup

and the environmental gain is measurable.

j. There is a low potential for recovery without removal or remedial actions.

k. Remediation is consistent with the stakeholders' goals and objectives.

EM 200-1-4

31 Dec 10

8-19

8.2.2.1. Risk Characterization/Uncertainty Information for Risk Management Decision

Making

8.2.2.1.1. From the risk manager's perspective, the baseline ERA should adequately

present risk estimates in an objective and unbiased manner. The PM understands that

although the risk assessment is a scientific tool, the results cannot be easily used to

determine specifications. Moreover, it is a tool for risk management decision-making, and

is rarely a tool for the prediction of actual occurrence of environmental effects. Therefore,

as long as the uncertainties are presented and understood by the customer and other

decision-makers, the results can be accepted or rejected for use in site decisions.

. The sources of uncertainty in a baseline ERA were presented in Chapter 6. The

objective of the risk characterization and uncertainty analysis is to make the ERA

transparent to the risk managers and the stakeholders so that informed risk management

decisions can be made. Given proper early project planning, it is expected that

uncertainties will be acceptable to the risk managers and other stakeholders, including

the BTAG members and other independent expert ecologists. The risk manager can

balance his or her selection of options with the findings of the risk assessment and the

degree of uncertainty in mind.

8.2.2.1.2. When making site decisions, the risk manager or PM can substantially

benefit from by consultation with responsible technical experts (risk assessors, expert

ecologist[s]/advisory panel [BTAG/ETAG]). It is the responsibility of these experts to

document and present uncertainties so the RM makes an informed decision. In the final

baseline ERA, the risk assessment summary presents risks and the associated

uncertainty information in a weight-of-evidence discussion which focuses on strengths

and weaknesses of the risk estimates, providing information to assist in determining the

overall objectives and decisions to be made in this project phase.

8.2.2.1.3. In order to make informed risk management decisions, the risk manager

should have a clear understanding of the following:

a. What are the receptors or resources to be protected?

b. Does the ecological risk involve individual organisms, communities, populations, or

different trophic levels?

c. What is the aggregate hazard (HI)? (this assumes that calculation of an HI is

justified for the contamination and the exposed receptors)

d. How do effects or ecosystem characteristics between the site and the reference

locations compare?

e. What is the likelihood of recovery based on consideration of the contaminants' fate

and transport properties, the substrate or media characteristics, natural attenuation, and

lessons learned from similar sites?

f. How do hazards under RME and CTE compare? What are the "order of

magnitude" differences?

EM 200-1-4

31 Dec 10

8-20

g. What is the key and overall uncertainty of the baseline ERA in terms of chemical

data, COEC selection, exposure assessment and modeling, toxicity information, and

characterization method? Is uncertainty quantifiable to the extent that the TCLs could be

substantially altered?

h. If the risk estimates are unacceptable, will quantitative analysis of uncertainty be

able to demonstrate that the risk estimate is based on overly conservative assumptions,

i.e., in the theoretical upperbound range?

i. What are the COEC(s) and exposure pathways that constitute the principal threat?

j. How are the exposure units defined in the baseline ERA?

k. Are there any "hot spots" which would require further characterization, or removal

action?

l. Are there any acute hazards or risks which will require emergency response or

removal action? Is there a risk of further spills, releases, or physical hazards that could

further degrade the environment or adversely impact the ecological receptors of concern?

m. If removal or early response actions are desirable, how effective are the proposed

removal actions to reduce site risk?

n. Which are the anticipated or preferred options for actions?

8.2.3. Feasibility Study/Remedial Design/Remedial Action. The FS is triggered when

the baseline risk is unacceptable and remediation is needed to mitigate risks and prevent

further contaminant migration. In some instances, the FS could be driven by a legal

requirement to meet ARARs, although ARARs are not necessarily risk-based. The FS

evaluates potential remedial alternatives according to established criteria in order to

identify the appropriate remedial alternative(s). The FS can be performed for the entire

site or any portion of the site that poses unacceptable risks. The results of the FS include

recommendations for the risk managers or site decision-makers, including an array of

remedies for selection, RAOs, or TCLs for verification of cleanup.

The selected

remedies or revisions thereof will be entered into the record of decision (ROD).

What are the Remedial Action Objectives?

Risk Management Decision

The risk management decision for selection of final remedies depends substantially

on the RAOs. Uses of RAOs are summarized below:

Risk Management Options/Rationale

For the purpose of protecting the environment, the TCLs sometimes known as RAOs, may be the same

as the environmental-based preliminary remediation levels, or they may be a different. TCLs or RAOs are

negotiated levels for verification of cleanup and take into consideration performance of the proposed cleanup

technology, practical quantitation limits, and uncertainties associated with the preliminary remediation levels

to protect ecological resources of concern.

EM 200-1-4

31 Dec 10

8-21

a. Developed or agreed upon by the agencies prior to the FS or signing of the ROD,

RAOs are used to evaluate the feasibility of candidate remediation technology in the FS;

b. Initial estimation and costing of remediation (e.g., excavation and stabilization);

c. Delineation of cutlines for remediation;

d. For use in negotiation or final determination of specific areas, or site-wide cleanup

goals, by considering uncertainties, technology, and cost.

Before embarking on an FS, RAOs should be developed using site-specific risk

information consistent with site conditions. Factors to be considered when RAOs are

used as the basis for designing and implementing remediation are presented below:

8.2.3.1. Remedial Action Objectives Must be Based on Ecological Conceptual Site

Model. The ECSM provides the framework for the baseline ERA and identifies the

specific pathways of concern; RAOs must be able to address these pathways and the

associated risks. A refined ECSM, based on the results of the ERA is paramount to the

establishment of focused RAOs. The RAOs are based on preliminary remediation levels

developed as the project strategy goals in Phase I of the HTRW project planning under

RI/FS.

8.2.3.2. Remediation Goals Must be Protective and Practical. Remedial goals are

performance and numerical objectives developed in the FS to ensure that the remedial

alternative will contribute to site remediation, restoration, and closeout/delisting. As such,

they must be protective and workable. To ensure protectiveness, risk-based preliminary

remediation goals should be first derived using the screening or baseline ERA procedures

in reverse (see procedures described in Chapter 7). The uncertainty associated with

development of the remediation goals should be discussed and quantified. Preliminary

remediation levels can be derived early in the site investigation process or at the end of

the RI, when it is determined that remediation may be needed because of unacceptable

risks. Site decision-makers carefully consider technology, practical quantitation limits,

ARARs or to-be-considered criteria, reference location concentrations, acceptable

hazards, field or laboratory analytical uncertainties, etc., before setting the RAOs.

8.2.3.3. Action Must Be Consistent with Other Project Phases. Understanding of the

nature and extent of contamination, as well as the media and exposure pathways of

concern, is a critical requirement for successful completion of the FS and remedy

selection. Therefore, data used in the FS must interface with the RI and other previously

collected site data. Inadequate data or data of poor quality misrepresent site

contamination and may lead to an inadequate baseline risk assessment and FS. For

each exposure pathway that presents an unacceptable ecological risk, the risk assessor

and the appropriate project team members (e.g., chemist, geologist or hydrogeologist)

should review the RI data before conducting the FS. This is particularly important when

the FS is performed simultaneously with the RI, based on assumptions and PA/SI data.

8.2.3.3.1. RAOs may be selected based on one of the following:

EM 200-1-4

31 Dec 10

8-22

a. Background.

Rationale: The environmental concentrations at the reference area or upgradient area

will be used as RAOs since the ecological receptors or the valued resources to be

protected are also located at the background locations. The reference area has the same

current land use as the site and the levels are reasonable and attainable.

b. RAOs are performance-based.

Rationale: No reasonable chemical-specific cleanup level can be derived due to high

uncertainty in the hazard-response relationship. For the purpose of remedy selection, the

best available or best demonstrated remedial technology will be utilized to achieve certain

risk reduction objectives according to the ECSM.

c. Risk-based Remediation Goals (Cleanup Goals).

Rationale: In lieu of performance based RAO or cleanup to the levels at the reference

area, risk-based RAO can be developed using dose-response information for the

ecological receptor of concern or its surrogate species. The risk-based RAOs may be

adjusted upward or downward according to other risk management factors or

considerations.

8.2.3.3.2. Minimal information or guidance has been developed by USEPA regarding

the development of RAOs for Superfund sites. RCRA has issued the Alternative

Concentration Limit (ACL) Guidance based on 264.94(b) criteria and case studies

(USEPA 1988c) which may be applied to developing ACLs at the source if the acceptable

groundwater/surface water mixing zone concentrations and the dilution/attenuation

factors are defined. Generally, ACLs are allowed under CERCLA, as well. For the

protection of aquatic receptors, cleanup levels can be set to chemical-specific water

quality criteria. Nonetheless, the key risk management issue concerning the above is that

the cleanup goals must be practical and verifiable. When cleanup goals are developed to

protect both humans and ecological receptors, according to Section 300.340 of the NCP,

the goals must be so adjusted that both receptor types are protected.

8.2.3.3.3. Environmental and human health-based RAOs should be developed

together and proposed to the risk manager and agencies for use in the FS for the

evaluation of remedial alternatives. It should be noted that the RAOs may have to be

revised or refined based on other considerations, e.g., technology, matrix effects, target

risks, uncertainties, and costs (associated with the extent of the remediation,

management of remediation wastes, cost of cleanup verification analyses).

What are the Remedial Alternatives?

Risk Management Decision

What are the Preferred or Optimal Remedial Alternatives?

Risk Management Options/Rationale

EM 200-1-4

31 Dec 10

8-23

a. In addition to a cost and engineering evaluation of the potential remedial

alternatives, each alternative must be evaluated for its ability to reduce site risk. Among

the nine criteria identified by the NCP for remedy selection, protection of human health

and the environment and satisfying ARARs are considered to be the threshold

(fundamental) criteria which must be met by any selected remedy. More recently, USEPA

has placed increased emphasis on short- and long- term reliability, cost, and

stakeholders' acceptance in the overall goal to select remedies. Therefore, the

assessment of residual risk (a measure of the extent of site risk reduction) is a critical

task.

b. Screening and detailed analyses of remedial alternatives will be conducted in the

FS project phase. The preferred remedial alternative will be proposed. As warranted,

analysis of short-term risks to assess the need for control measures will be conducted in

the RD project phase, and the control measure(s), if appropriate, will also be proposed.

c. In the FS, potential risk reductions associated with remedial alternatives are

assessed. The relative success of one alternative over another is simply the ratio of the

residual COEC concentrations in the exposure medium of concern. This screening

evaluation does not take into account short-term risks posed by the alternative or

technology due to acute hazards, releases or spills.

8.2.3.4. Screening Evaluation of Alternatives. This evaluation focuses on

determination of short-term risks posed by the removal or remedial alternatives. The

findings of this evaluation is compared among the alternatives to determine preferred

remedies based on the effectiveness of the remedies to satisfy remedial action goals with

the least environmental impact. This screening evaluation should focus primarily on

effectiveness, risk reduction and cost.

8.2.3.4.1. Risk screening of alternatives should generally be qualitative or semi-

quantitative. If a remedy has already been selected or is highly desirable for selection, a

detailed risk analysis may not be needed. Instead, the evaluation should focus on the

risk reduction of the preferred remedy, and identify any concerns or data gaps which

need to be addressed. The data needed to perform this screening evaluation may come

from many sources: RI data, bench scale or pilot scale treatability studies conducted for

the site or from comparable sites, compatibility test, test of hazardous characteristics, field

monitoring measurements, vendor's or manufacturer's information, literature values, and

professional judgment.

a. Identity and quantity of emissions, effluent, byproducts, treatment residues, which

may be released to the environment (during normal, start-up and shut-down operations);

Key information needed prior to conducting the screening

evaluation of remedial alternatives include:

The bench scale or pilot scale treatability studies may provide valuable information for the estimation of

remediation action or residual risks. Treatability studies provide data or information on the degree of removal

and/or destruction of the COECs, quantity and identity of chemicals in the emissions or effluent discharges,

and potential treatment standards to be applied to satisfy remedial action goals. This information is important

to quantify the magnitude of risk reduction and will be useful in the comparative analysis of potential remedial

alternatives.

EM 200-1-4

31 Dec 10

8-24

b. Toxicity of chemical substances or COECs in the above discharges;

c. Potential for dilution and attenuation;

d. Existence of exposure pathways and likelihood of the pathways to be significant

and complete;

e. Potential for spill or releases during remediation, material handling, storage and

transportation of remediation wastes;

f. Potential for the causation of non-chemical stressors such as destruction of critical

habitat for threatened and endangered species, wetlands, or other sensitive

environments, increased siltation and reduction of food sources for the ecological

receptors of concern or other receptors/valued resources;

g. Temporal attributes associated with a remedial action which could be altered to

reduce the action's impact; and

h. Potential release of additional COECs to the environment (e.g., re-suspension of

toxic sediments during dredging, and changes of pH, redox potential, oxygen, and

chemical state that may increase solubility and bioavailability).

8.2.3.4.2. The following are lists of qualitative evaluation criteria:

a. Risk Reduction Attributes (environmental protection, permanence, and toxicity

reduction).

(1) Able to remove, contain or effectively treat site COECs.

(2) Able to address the exposure pathways and media of concern.

(3) Able to meet the remedial action and overall project strategy goals.

b. Assessment of Residual Risk Potential.

(1) Reasonable anticipated future land use (for example, if the site remains

industrial/commercial in a foreseeable future, residual risk assessment should not be

performed for the potential return of and exposure to terrestrial receptors).

(2) Quantity of residues or discharges to remain on site.

(3) Toxicological properties of the residues.

(4) Release potential of residues based on their fate/transport properties (e.g., log

octanol/water partition coefficient, water solubilities, vapor pressure, density, etc.).

(5) Properties or characteristics of the environmental medium which facilitate

transport (e.g., hydraulic conductivity, organic carbon contents, wind speed and direction,

etc.).

(6) Potential for dilution and attenuation for residues released into the environment.

EM 200-1-4

31 Dec 10

8-25

(7) The extent of and permanence of remediation habitat destruction and alteration,

for example, the construction of an access road through wetlands would be considered

permanent.

8.2.3.5. Detailed Analysis of Alternatives. Detailed analysis is usually conducted for

the preferred remedial alternatives (or removal actions), identified in the screening

evaluation described above. This detailed analysis has three objectives: (a) detailed

assessment of potential short-term risk during remedial action, and residual risks if

appropriate; (b) assess the potential for the risks to be magnified due to simultaneous

implementation of this and other preferred alternatives; and (c) identify potential risk

mitigation measures for the preferred remedies. The findings of these tasks are

presented for final selection of remedies prior to ROD sign-off. All preferred remedies or

options should satisfy remedial action goals and should pose minimum health and

environmental impact.

8.2.3.5.1. This evaluation may be qualitative, semi-quantitative, or quantitative. If the

analysis is quantitative, procedures and approaches similar to the baseline risk

assessment may be followed. USEPA's (1995e) Air/Superfund National Technical

Guidance Study Series includes documents providing guidance for rapid assessment of

exposure and risk. For example, guidance on determining the volume of soil particulates

generated during excavation is provided in Estimation of Air Impacts for the Excavation of

Contaminated Soil (USEPA 1992j). The data sources used to perform this risk analysis

task should be similar to those identified for the screening evaluation of remedial

alternatives. Although it is conceivable that the level of effort required for this analysis

may be high (particularly, if the same analysis has to be performed for a number of

preferred remedies), it is anticipated that the documentation and report writing will be

focused and streamlined.

8.2.3.5.2. The report should focus on the risk analysis approaches, sources of data,

findings/recommendations for risk mitigation measures, and appendixes. Key factors or

criteria to be considered in the screening evaluation of remedial alternatives are:

a. The criteria or considerations in the assessment of short-term and residual risks

are substantially similar to those identified for the screening evaluation of remedial

alternatives. The key difference may be additional use of quantitative data input into the

risk calculations, e.g., sediment transport modeling to evaluate the potential for migration

of toxic sediment, amount of discharges or emissions, dilution/attenuation or atmospheric

dispersion factors, exposure frequency, duration, and other activity patterns which could

impact existing flora and fauna in time and space, and any indirect effects such as food

source reduction and the extent of habitat destruction/alteration.

b. Time required and extent of recovery from exposure to the above COECs and

non-chemical stressors.

c. The potential for fire, explosion, spill, and release of COECs from management

practice of excavated or dredged materials should remain qualitative or semi-quantitative.

Fault-tree (engineering) analysis for accidental events may be attempted under special

EM 200-1-4

31 Dec 10

8-26

circumstances (e.g., to address public comments or if demanded by citizens during public

hearing of the proposed remedies).

8.2.3.6. Risks from Simultaneous Implementation of Preferred Remedies.

a. Common exposure pathways for effluent or discharges from remedies.

b. Period of exposure to the ecological receptors of concern via the common

locations, time and pathways.

c. Sensitive environments and other threatened or sensitive wildlife or aquatic

populations.

d. Risk estimates or characterization results.

e. Toxicological evaluation for the validity of biomagnification and additivity of risk

(e.g., under the Quotient Method), based on literature review, mode of action, and

common target organs, etc.

f. Qualitative or quantitative assessment of potential short-term or residual risks.

8.2.3.6.1. Short-Term Risks Associated with Construction; the Design Risk Analysis.

All removal or remedial alternatives have a potential to pose short-term risks to on-site

mitigation workers, ecological receptors, and off-site humans. Risks may be associated

with vapors, airborne particles, treatment effluent, resuspension of sediment resulting in

an increase in the total suspended solids or siltation of substrate for macroin-vertebrates,

and residues generated during operation of the remedial alternative. Therefore, all the

alternatives should be reviewed for their short-term risks in conjunction with data from

their bench scale or pilot scale treatability studies or data from implementa-tion of the

remedy at comparable sites. The risk assessor should estimate the period of recovery

from these short-term insults and determine if biological or chemical monitoring of the

effects of remediation activities should be implemented. For all practical purposes, risk

may remain upon completion of the remedial action (residual risk).

8.2.3.6.2. Long-Term Risks Associated with Alternatives; the Residual Risks. Unless

all sources of contamination are removed or isolated, there will be residual risks at the site

upon completion of the remedial action. The COEC residuals could either remain or be

quickly degraded, depending on the COEC's physical and chemical properties. The level

of residual risk will depend on the effectiveness of the remedy in containing, treating or

removing site contaminants, and the quantity, and physical, chemical, and toxicological

characteristics of residues or byproducts remaining at the site. Site COECs which remain

on-site after the remedial action should be assessed for their potential risks.

8.2.3.6.2.1. This evaluation step focuses on a risk reduction assessment to determine

if a potential remedial alternative is able to meet the remedial action goals; and an

assessment of residual risk potential. The findings of these tasks are compared among

the alternatives to determine an array of preferred remedies based on the effectiveness of

EM 200-1-4

31 Dec 10

8-27

the remedies to satisfy remedial action goals with the least long-term health and

environmental impact.

8.2.3.6.3. Remedial Action/Residual Risks vs. Baseline Risk. There are notable

differences between remedial action/residual risks and the baseline risk. The key

difference is that baseline ecological risk refers to the potential risk to the receptors of

concern under the "no remedial action" alternative, and remedial action and residual risks

refer to short-term risks during remedial action and long-term risks which may remain

after completion of the remedial action, respectively. Residual risk may be considered

comparable to baseline ecological risk after remediation; since in both cases, the risks are

chronic or subchronic in nature. Remedial action risks are generally short-term (acute or

subchronic) risks.

8.2.4. Non-Risk Issues or Criteria as Determining Factors for Actions. The NCP

recognizes that it is not possible to achieve zero risk in environmental cleanup; therefore,

the approach taken by Superfund is to accept non-zero risk and return the site to its best

current use (not to conditions of a pre-industrialization era). This section presents and

discusses the non-risk factors, and recommends a balanced approach for resolution of

issues to enable quality risk management decision-making. These factors can be

categorized into scientific and non-scientific factors, as explained below.

8.2.4.1. Scientific Factors. The scientific factors, including engineering design and

feasibility, should be considered in risk management decision-making. These factors

focus on technology transfer (realistic performance of the technology), duration of

protection, and feasibility study data uncertainties. These factors will influence the

decision whether or not to proceed with selection of a particular remedy. They are

detailed below:

8.2.4.1.1. Technology Transfer. This factor concerns the treatability of the

contaminated debris or media by a preferred technology or early action. Although the

recommended technology may appear attractive, a number of problems must be

overcome before actual selection or implementation of the action. The following are a

few examples:

a. Scale up,

b. Downtime and maintenance (including supplies),

c. Ownership/control,

d. Throughput to meet the required completion schedule,

One exception (i.e., remedial action risk which is long-term) is a pump-and-treat remedy of groundwater to

meet maximum contaminant levels for organics which pose a threat to human health but not ecological

receptors. If the effluent is discharged to a surface water body and happens to contain trace elements at

high levels (or other COECs not reduced by the treatment process), then an exposure route to environment

receptors may remain which is not addressed by the baseline ERA, and which will exist for the operational

life-span of the remedy.

EM 200-1-4

31 Dec 10

8-28

e. Skills required or training requirements,

f. Quantitation and detection limits, and

g. Space requirements for the remediation process and management of remediation

wastes.

8.2.4.1.2. Duration of Protection. This factor concerns the duration of the removal or

remedial technology designed to treat or address site risk. This factor is particularly

important for site radionuclides or non-aqueous phase liquid (NAPL) compounds in the

aquifer. The maintenance or replacement of barriers or equipment is also a primary

concern for this factor. Although a technology or alternative is effective, its effectiveness

may not last long if there is no source control or contamination from off-site sources is not

controlled

8.2.4.1.3. Data Uncertainty. This factor considers reliability and uncertainty of certain

site or feasibility study data for use in selecting a remedy, or for determining whether no

further action is appropriate. Uncertainty in the following data may also impact the risk

analyses or baseline risk assessment results:

a. Adequacy of bench-scale or pilot-scale treatability data,

b. Data uncertainties (volume, matrices, site geology/hydrogeology),

c. Field data and modeling data, and

d. Overall uncertainty of the source of site contamination.

8.2.4.2. Non-Scientific Factors. Non-scientific factors should also be considered in

risk management decision-making because some of these factors are key to a successful

site remediation. Most of these factors are internal, but can also be external. Examples

of these factors are enforcement, compliance, schedule, budget, competing risk reduction

priorities, community inputs, and societal/economic value of the resources to be

protected. These factors will influence the decision on whether or not certain removal or

remedial actions should be taken, or on which remedies are to be selected. These

factors are detailed below.

8.2.4.2.1. Enforcement and Compliance. Certain courses of action (including risk

management decisions) have been agreed upon early in the process and have been

incorporated in the IAG or FFA. This is particularly germane to some earlier HTRW sites.

Nonetheless, the requirements specified in the enforcement documents or administrative

order of consent, IAG, FFA should be followed by the risk manager or PM with few

exceptions. When risk-related factors or other non-risk factors are over-arching, the risk

manager should then raise this issue to higher echelon or to the legal department for

further action or negotiation.

8.2.4.2.2. Competing Risk Reduction Priorities. Although related to risk, this factor

represents the competing interest among programs or within the project for a limited

source of funding to perform risk reduction activities. Since it is likely that not all sites will

EM 200-1-4

31 Dec 10

8-29

be cleaned up at an equal pace, the planning and execution of environmental restoration

among these units should follow a prioritization scheme. However, the scheme

developed according to risk may not be the same according to the customer, the base

commander, or the agencies. The risk manager or PM must seek common ground to

resolve this issue so that resources can be expended to produce incremental

environmental benefits.

8.2.4.2.3. Schedule and Budget. These factors usually go together because the

more protracted the project life, the more resources the project will demand. While each

PM would like to comply with risk-based considerations with little margin of error, the PM

may have no choice but to make risk management decisions with larger uncertainties

than he or she would prefer, due to schedule and budget constraints.

8.2.4.2.4. Community Input. Opportunity for the stakeholders or community to

provide input into the permit modification is provided when primary documents are

prepared, i.e., RI Work Plan, RI/FS reports, the proposed remedies, and the RA Work

Plan. Superfund also provides similar opportunities for public participation. To be

successful in site remediation and closeout, the risk managers must be able to

communicate risks effectively in plain and clear language without bias. Early planning

and solicitation of community input is essential to democratization of risk management

decision-making. Some of the following issues may be of concern to the communities:

a. Ineffective communication of risks and uncertainties.

b. Lack of action (some action is preferred to no action).

c. Not in my backyard (off-site transportation of contaminated soil, debris or sediment

should avoid residential neighborhoods).

d. Any treatment effluent or discharge is unacceptable (on-site incineration is seldom

a preferred option except for mobile incinerators, in certain instances).

e. The remedy should not impede economic growth or diminish current economic

and recreational value of resources to be protected.

f. Cleanup will improve the quality of life and increase property values or restoration

of recreational/economic resources.

8.2.4.2.5. Societal/Economic Value of the Resources to be Protected. This non-risk

factor concerns the community sentiment on how fast, or in what manner, the resources

impacted by site contaminants should be restored. These resources may include surface

water bodies, wildlife, and game animals. Most communities would like to see impacted

resources restored to original use, however, this can be difficult to achieve. Some

communities may be willing to accept natural attenuation or no action options for

impacted surface water bodies, given the opportunity to examine the pros and cons of all

options. Therefore, it is recommended that the risk manager execute a community

relations plan in earnest in order to solicit the citizens' input on the risk reduction

EM 200-1-4

31 Dec 10

8-30

approach and issues of concern. Key community spokespersons may also be appointed

to the site action committee to facilitate such dialogue and communication.

8.2.4.3. A Balanced Approach. In conclusion, the risk manager should consider all

risk and non-risk criteria before making risk management site decisions. Due to

uncertainties associated with ERA or analysis, the decision-maker must review risk

findings and the underlying uncertainties, and consider other non-risk factors in the

overall risk management equation. When making risk management decisions, the risk

manager should keep an open mind regarding the approaches to meet the project

objective. In order to make informed site decisions, the risk assessor must present risk

estimates in an unbiased manner. With an understanding of the volume of contaminants

of concern, significance and biological relevance of the ecological effects and potentially

impacted receptors, fate/transport properties of the COECs, and completeness of the

exposure pathways and the food web, the risk manager, PM, and stakeholders will be

better equipped to make informed decisions. These decisions should be consistent with

the overall site strategy, which is developed early in the project planning phase (see

Chapters 2 and 3), and which may evolve throughout the project.

8.3. Design Considerations

This section addresses the above issues, i.e., risk management considerations in

remedial design, compliance with ARARs, including the clean air act (CAA), CWA, ESA,

and other major environmental statutes, and control measures required to mitigate risks.

. Risk assessment methodology can be an important tool in

the design phase of CERCLA remedial actions. During the early phase of RD/RA, risk

assessment results can help determine: 1) whether the selected remedy can be

implemented without posing an unacceptable short-term risk or residual risk; and

2) control measures (operational or engineering) to mitigate site risks and to ensure

compliance with ARARs, and to-be-considered requirements, and permit conditions. The

risk and safety hazard information will be evaluated by the site decision-makers, along

with information concerning design criteria, performance goals, monitoring/compliance

requirements prior to making risk management decisions regarding the above questions.

Further, the decision-makers consider potential requirements such as ARARs and to-be-

considered (TBCs) in determining design changes or control measures.

8.3.1. Potential Risk Mitigation Measures.

8.3.1.1. Engineering Control. Where appropriate (when short-term risks are

determined to be unacceptable), engineering controls should be recommended by the

design engineer with inputs from the risk assessor, aquatic ecologist, compliance

specialist, and the air modeler. Examples of these control measures include:

a. VOC and semi-volatile organic compound (SVOC) emissions - activated carbon

canisters, after burners, or flaring, prior to venting.

b. Metals and SVOC airborne particles - wetting of work areas; particulate filter/bag

house, wet scrubber, or electrostatic precipitator (for thermal treatment devices or

incinerators).

EM 200-1-4

31 Dec 10

8-31

c Fugitive emissions - monitoring of valves, pipe joints, and vessel openings; and

barrier/enclosure of work areas (e.g., a can or shield over the augering stem).

d. Neutralization or chemical deactivation of effluent (continuous process or batch).

e. Use of remote control vehicle for handling, opening or cutting of drums containing

explosive or highly reactive or toxic substances.

8.3.1.2. Operational Control. Where appropriate, administrative control measures

(procedural and operational) safeguards should be recommended by the PM, design

engineer, field supervisor during RA, with inputs from the risk assessor and other relevant

technical and compliance specialists. Examples of these control measures include:

a. Establish short-term trigger levels which will require work stoppage or upgrade of

the remediation procedures (e.g., dredging of toxic sediments). Either biological or

chemical indicators, or their combination could be used as the trigger levels. These levels

should be developed in the RD/RA project phase by the risk assessor and other technical

specialists, including the modeler.

b. Consistent with the above trigger or acute concern levels, evaluate on-site

performance with field equipment to ensure adequate remediation.

c. Afford the proper protection of sensitive environments by careful planning and

positioning of staging area, storage or management of remediation wastes, selection of

equipment with low load bearing, and season or time period when the remediation should

be completed.

d. Establish a zone of decontamination and proper management of effluent or waste

generated from this zone.

e. Secure and control access to areas where remedial actions are being

implemented at all time.

8.3.1.3. Institutional Control. Although institutional control may not be relevant for

ecological receptors, it can be relevant in the sense that institutional control measures

may be needed to reduce human intrusion, thus allowing the sensitive environments to

recover or the ecological receptors to re-establish. Institutional controls are particularly

pertinent for remedies which involve containment, on-site disposal of wastes, or wetlands

remediation. Institutional controls should be recommended by the customer, PM, and

other site decision-makers. Examples of these control measures include:

a. Recording land use restrictions in the deeds (deed restrictions) for future use of

certain parcels or areas where hazardous substances or wastes are contained.

b. Erection of placards, labels, and markers which communicate areas where human

exposure may pose short-term or residual risks.

c. Security fences and barriers.

EM 200-1-4

31 Dec 10

8-32

8.3.2. Risk Management; Degree of Protectiveness. Not only should a selected

remedial action be able to meet balancing criteria, the remedial action must be protective,

i.e., in terms of reducing site risks. In designing a selected remedy, the site decision-

makers may face operational or engineering issues which are likely to require risk

management decisions. For example, if a detailed analysis of a selected remedy reveals

potential short-term or residual risks, the decision-makers must decide to what extent and

with what control measures are necessary to abate the risk. Inputs from the risk assessor

will be needed to help make informed risk management decisions. The following are

examples of key risk management considerations for designing an effective remediation

strategy:

a. Acceptability of control measures. There are potential operational (procedural) or

engineering control measures to address the short-term risks. The risk assessor, in

coordination with the design engineer, expert ecologist(s)/advisory panel, and other

project team members, assesses the effectiveness of any proposed control measures.

b. Removal of control measures. Before a control measure is implemented, the

decision on the minimum performance and when to stop requiring the control measure

has to be addressed. This is particularly important if control measures are costly to

implement and maintain.

c. Effectiveness of the remediation. Remediation should effectively address on-site

contamination if there is a continuing off-site (regional) source. This consideration is

particularly important for groundwater and sediment contamination remediation. This

regional source control strategy should not be confused with the identification of PRPs

since some of the discharges could be a permitted activity. Nonetheless, this issue has to

be resolved if the RAOs are risk-based and do not consider off-site influences or

contribution to the contaminants requiring remediation. Off-site source control and

containment, waste minimization, and closure issues should be raised by the risk

manager to the agencies, USACE customers, and higher echelon.

d. BRAC. With BRAC, the land use of closed defense facilities may not be

indefinitely controlled and the legislation governing BRAC holds the U.S. government

responsible for future cleanup of contamination caused by government activities.

Cleanup criteria and long-term remedies should take land use into consideration for

implementation of an effective site closeout strategy (see Chapter 2). For example,

conversion of military bases into a state park or refuge area will require different cleanup

objectives than cleanup to the level acceptable for industrial/commercial usage. This

issue should be addressed early in the site strategy development phase with input from

customers, local re-development commissions, state, and other stakeholders.

e. Verification of cleanup. The risk management decision concerning verification of

cleanup, i.e., the numerical value of the RAO should be based on a combination of

factors: risk, uncertainty, statistics, analytical detection limits/matrices, and costs.

Although RAOs have been negotiated or determined in the ROD, the sampling method

and statistical requirements must be clearly articulated before design and implementation

of the corrective measures or remedial alternatives.

EM 200-1-4

31 Dec 10

8-33

f. Risk management decisions during the design phase of a CERCLA remediation

should be flexible, considering the uncertainty in the risk assessment results, acceptable

risk range, confidence level of toxicity data or criteria to support the assessment,

engineering feasibility, reliability of the measures (operational changes vs. pollution

control equipment), state and community acceptance, and cost. It is recommended that

risk managers and site decision-makers request input from all members of the project

team for pros and cons of proposed control measures to address the short-term risks.

EM 200-1-4

31 Dec 10

8-34

THIS PAGE INTENTIONALLY LEFT BLANK

EM 200-1-4

31 Dec 10

A-1

APPENDIX A

References

A.1. Required Publications

References listed below have been cited in the text.

Executive Order (EO) 12580 (Presidential Document). Superfund Implementation.

52 FR 2923. January 29, 1987. This was amended in 91, 96, 03twice

U.S. Department of Defense (USDOD). 1994. Relative Risk Site Evaluation Primer.

Interim Edition. Washington, D.C. This was revised in 1997

U.S. Army (USA). 2005 (September). United States Army Interim Natural Resource

Injury Policy Guidance. Prepared by the U.S. Army Environmental Center.

U.S. Army Biological Technical Assistance Group (USA BTAG). 2002a. Technical

Document for Ecological Risk Assessment: Planning for Data Collection.

SDIM-AEC-ER-TR-2002017. January.

USA BTAG. 2002b. Selection of Assessment and Measurement Endpoints for Ecological

Risk Assessment. SFIM-AEC-ER-TR-2002018. February.

USA BTAG. 2005a. Technical Document for Ecological Risk Assessment: A Guide to

Screening Level Ecological Risk Assessment. April.

USA BTAG. 2005b. Technical Document for Ecological Risk Assessment: Process for

Developing Management Goals. August.

USA BTAG. 2006. Technical Document for Ecological Risk Assessments: Integrating

Multi-Site Ecological Risk Assessments for Wide-ranging Receptors. August.

U.S. Army Center for Health Promotion and Preventive Medicine (USACHPPM), 2000.

Standard Practice for Risk Assessment: Development and Use of Toxicity Reference

Values. Tech Guide No. 254. October.

USACHPPM. 2004. Development of Terrestrial Exposure and Bioaccumulation

Information for the Army Risk Assessment Modeling System (ARAMS). April.

U.S. Army Corps of Engineers (USACE). 1987. Wetlands Evaluation Technique (WET).

Volume II: Methodology.

USACE. 1992. Environmental Compliance Assessment System (ECAS) Assessment

Protocols. Construction Engineering Research Laboratories (CERL). Per AR 200-1.

USACE. 1998. Technical Project Planning (TPP) Process. EM 200-1-2.

USACE. 1999. Risk Assessment Handbook: Volume I - Human Health Evaluation.

EM 200-1-4.

EM 200-1-4

31 Dec 10

A-2

USACE. 2001. Standard Scopes of Work for HTRW Risk Assessments. EP 200-1-15.

USACE, 2003a. Conceptual Site Models for Ordnance and Explosives (OE) and

Hazardous, Toxic, and Radioactive Waste (HTRW) Projects. EM 111 0-1-1200.

USACE. 2003b. FUDS Program Guidance to Implement Army Interim Policy for

Integrating Natural Resource Injury Responsibilities and Environmental Response

Activities

USACE. 2005. Guidance for Evaluating Performance-Based Chemical Data.

EM 200-1-10.

USACE. 2008. Environmental Statistics. EM 111 0-1-4014.

U.S. Environmental Protection Agency (USEPA). 1980. Ambient Water Quality Criteria

for Polychlorinated Biphenyls. USEPA Report 440/5-80-068.

USEPA. 1982. Water Quality Assessment: A Screening Procedure for Toxic and

Conventional Pollutants - Part 1. EPA-600/6-82/004a.

USEPA. 1983. Hazardous Waste Land Treatment. Office of Solid Waste and

Emergency Response. Washington, D.C. SW-874. April.

USEPA. 1985. Rapid Assessment of Exposure to Particulate Emissions from Surface

Contamination Sites. EPA/600/8-85/002. February.

USEPA. 1986a. Quality Criteria for Water

. Office of Water Regulations and Standards.

EPA 440/5-86/001. May. (see at:

National Recommended Water Quality Criteria | US

EPA)

USEPA. 1986b. Hazard Evaluation Division, Standard Evaluation Procedure, Ecological

Risk Assessment. Office of Pesticide Programs. EPA/540/9-85/001. June.

USEPA. 1987. Selection Criteria for Mathematical Models Used in Exposure

Assessments: Surface Water Models. EPA/600/8-87/042. July.

USEPA. 1988a. Superfund Exposure Assessment Manual. OSWER Dir. 9285.5-1.

EPA/540/1-88/001. April.

USEPA. 1988b. A Workbook of Screening Techniques for Assessing Impacts of Toxic Air

Pollutants. EPA-450/4-88-009. September.

USEPA. 1988c. Selection Criteria for Mathematical Models Used in Exposure

Assessments: Ground-Water Models. EPA/600/8-88/075. May.

USEPA. 1988d. Recommendations for and Documentation of Biological Values for Use

in Risk Assessment. Environmental Criteria and Assessment, Office of Research and

Development. EPA/600/6-87/008. February.

EM 200-1-4

31 Dec 10

A-3

USEPA. 1989a. Risk Assessment Guidance for Superfund. Volume II (RAGS II):

Environmental Evaluation Manual (RAGS II). Interim Final. Office of Emergency and

Remedial Response. EPA/540/1-89/001. March. (this document was superseded by

USEPA 1997a)/

USEPA. 1989b. Ecological Assessments of Hazardous Waste Sites: A Field and

Laboratory Reference Document. Office of Research and Development.

EPA/600/3-89/013. March.

USEPA. 1989c. Policy and Program Requirements to Implement the Quality Assurance

Program. Washington, D.C. Office of the Administrator. EPA Order 5360.1. April 17.

USEPA. 1989d. Risk Assessment Guidance for Superfund. Volume I (RAGS I): Human

Health Evaluation Manual (RAGS I). EPA/540/1-89/002. December.

USEPA. 1989e. Air/Superfund National Technical Guidance Study Series. Volume II -

Estimation of Baseline Air Emissions from Cleanup Activities at Superfund Sites.

EPA-450/1-89-002.

USEPA. 1989f. Air/Superfund National Technical Guidance Study Series. Volume III -

Estimation of Baseline Air Emissions from Cleanup Activities at Superfund Sites.

EPA-450/1-89-003.

USEPA. 1989g. Rapid Bioassessment Protocols for Use in Streams and Rivers:

Benthic Macroinvertebrates and Fish. Office of Water (WH-553). EPA/444/4-89/001.

USEPA. 1989h. CERCLA Compliance with Other Laws Manual Parts I and II. Office of

Emergency and Remedial Response; OSWER Directive 9234.1-01.

USEPA. 1990a (March 8). National Oil and Hazardous Substances Pollution

Contingency Plan. Final Rule. 55 FR 8660.

USEPA. 1990b. Interim Methods for Development of Inhalation Reference

Concentrations. Review Draft. Office of Research and Development.

EPA/600/8-90/066A.

USEPA. 1990c. Superfund Emergency Response Actions. A summary of Federally

funded removals, fourth annual report, fiscal year 1989. Office of Research and

Development. Washington, D.C. EPA/540/8-89-014.

USEPA. 1991a. Role of the Baseline Risk Assessment in Superfund Remedy Selection

Decisions. Memorandum from Don R. Clay, the Assistant Administrator to Regional

Division Directors. OSWER Directive 9355.0-30. Washington, D.C. April 22.

USEPA. 1991b. Assessment and Control of Bioconcentratable Contaminants in Surface

Water. Office of Research and Development, Office of Water, Washington, D.C.

EM 200-1-4

31 Dec 10

A-4

USEPA. 1991c. Evaluation of Dredged Material Proposed for Ocean Dumping. Office of

Marine and Estuarine Protection, EPA and Dept. of the Army, USACE. EPA/503/B-

92/001. February.

USEPA. 1992a. Framework for Ecological Risk Assessment. Risk Assessment Forum.

EPA/630/R-92/001. (this document was superseded by USEPA 1998a)

USEPA. 1992b. Guidance for Data Useability in Risk Assessment. Part A. Publication

No. 9285.7-09A. PB92-963356. April.

USEPA. 1992c. Guidance for Data Useability in Risk Assessment. Part B. Publication

No. 9285.7-09B. PB92-963362. May.

USEPA. 1992d. Guidance on Risk Characterization for Risk Managers and Risk

Assessors. Memorandum from F. Henry Habicht, Deputy Administrator. February.

USEPA. 1992e. Supplemental Guidance to RAGS: Calculating the Concentration Term.

Office of Solid Waste and Emergency Response. OSWER Directive 9285.7-081. May.

USEPA. 1992f. Air/Superfund National Technical Guidance Study Series.

Volume I - Overview of Air Pathway Assessments for Superfund Sites (Revised).

EPA-450/1-89-001a.

USEPA. 1992g. Interim Guidance on Interpretation and Implementation of Aquatic Life

Criteria for Metals. Office of Science and Technology, Health and Ecological Criteria

Division. May.

USEPA. 1992h. RCRA Corrective Action Stabilization Technologies - Proceedings.

Office of Research and Development, Washington, D.C. EPA/625/R-92/014.

USEPA. 1992i. Superfund Accelerated Cleanup Model (SACM). Office of Solid Waste

and Emergency Response, OSWER Directive 9203.1-01, April 7. (Also see OSWER

Directive 9203-1-03, July 7, 1992, Guidance on the Implementation of the SACM under

CERCLA).

USEPA. 1992j. Estimation of Air Impacts for the Excavation of Contaminated Soil.

EPA-450/1-92-004.

USEPA. 1992k. May. ECO UPDATE, Developing a Work Scope for Ecological

Assessments, Publication 9345.0-05I, Vol. 1, #4.

USEPA. 1992l. May. Guidelines for Exposure Assessment. EPA/600/Z-92/00

USEPA. 1993a. Guideline for Deriving Site-Specific Sediment Quality Criteria for the

Protection of Benthic Organisms. EPA/822/R/93/017.

USEPA. 1993b. Guidance Document on Dissolved Criteria, Expression of Aquatic Life

Criteria for Metals. Proceedings to Water Mtg. Div. Dir. Washington, D.C.

EM 200-1-4

31 Dec 10

A-5

USEPA. 1993c. Air/Superfund National Technical Guidance Study Series.

Volume IV - Guidance for Ambient Air Monitoring at Superfund Sites (Revised).

EPA-451/R-93-007.

USEPA. 1993d. Wildlife Exposures Factors Handbook. Office of Research and

Development. 2 Volumes. EPA/600/R-93/187a&b. December.

USEPA. 1993e. A Review of Case Studies from a Risk Assessment Perspective. Risk

Assessment Forum, Washington, DC. EPA/630/R92/005.

USEPA. 1993f. Water Quality Standards Handbook. Office of Water.

EPA-823-B-93-002 (Second Edition: August 1994).

USEPA. 1993g. Interim Report on Data and Methods for Assessment of 2, 3, 7, 8-

Tetrachlorodibenzo-p-dioxin Risks for Aquatic Life and Associated Wildlife. Office of

Research and Development, Washington, D.C. EPA/600/R-93/005.

USEPA. 1993h. Initiative by EPA Official (Denit).

USEPA. 1994a. "Role of the Ecological Risk Assessment in the Baseline Risk

Assessment." Memorandum from Elliott Laws, Assistant Administrator. OSWER

Directive No. 9285.7-17.

USEPA. 1994b. Laboratory Data Validation, Functional Guidelines for Evaluating

Inorganics Analyses. EPA Hazardous Site Evaluation Division, Data Review Work Group.

July 1.

USEPA. 1994c. Laboratory Data Validation, Functional Guidelines for Evaluating

Organics Analyses. EPA Hazardous Site Evaluation Division, Data Review Work Group.

USEPA. 1994d. Sediment Quality Criteria

. Notice of Availability and Request for

Comment on Sediment Quality Criteria and Support Documents. 59 FR 2652, Vol. 59,

No. 11.

USEPA. 1994e. Workshop on the Use of Available Data and Methods for Assessing

Ecological Risks of 2,3,7,8-Tetrachlorodibenzo-p-dioxin to Aquatic Life and Associated

Wildlife. Risk Assessment Forum. Office of Research and Development, Washington,

D.C. EPA/630/R-94/002.

USEPA. 1994f. Eco Update, Vol. 2; No. 1. Using Toxicity Tests in Ecological Risk

Assessment. OSWER Publication 9345.0-051. EPA 540-F-94-012. September.

USEPA. 1994g. Eco Update, Vol. 2; No. 2. Catalog of Standard Toxicity Tests for

Ecological Risk Assessment. OSWER Publication 9345.0-051. EPA 540-F-94-013.

September.

USEPA. 1994h. Eco Update, Vol. 2; No. 3. Field Studies for Ecological Risk

Assessment. OSWER Publication 9345.0-051. EPA 540-F-94-014. September.

EM 200-1-4

31 Dec 10

A-6

USEPA. 1994i. Eco Update, Vol. 2; No. 4. Catalog of Standard Toxicity Tests for

Ecological Risk Assessment. OSWER Publication 9345.0-101. EPA 540-F-94-050.

September.

USEPA. 1995a. "EPA Risk Characterization Program" and "Policy for Risk

Characterization at USEPA." Memo from C. Browner. March 21.

USEPA. 1995b. Final Water Quality Guidance for the Great Lakes System. Final Rule.

Federal Register. Vol. 60. No. 56. March 23. p. 15366. (See supporting documentation:

Great Lakes Water Quality Criteria Documents for the Protection of Wildlife.)

USEPA. 1995c. Guidance for Risk Characterization. (Attachment to C. Browner

Memorandum, 1995). Science Policy Council. February.

USEPA. 1995d. Water Quality Standards: States Compliance - Revision of Metals

Criteria Interim Final Rule. 60 FR 22229.

USEPA. 1995e. Air/Superfund National Technical Guidance Study Series. Volume V -

Procedures for Air Dispersion Modeling at Superfund Sites. EPA-454/R-95-003.

USEPA, 1996. January. Eco Update, Ecotox Thresholds, EPA 540/F-95/038, PB95-

96334, Publication 9345.0-12FSI.

USEPA. 1997a. June. Ecological Risk Assessment Guidance for Superfund: Process

for Designing and Conducting Ecological Risk Assessments. Environmental Response

Team. EPA 540-R-97-006, OSWER Directive # 9285.7-25.

USEPA, 1997b. May. Policy for use of Probabilistic Analysis in Risk Assessment at the

U.S. Environmental Protection Agency.

USEPA. 1998a. Guidelines for Ecological Risk Assessments. Final.

USEPA, 1998b. Methodology for Assessing Health Risks Associated with Multiple

Pathways of Exposure to Combustor Emissions. NCEA. EPA 600/R-98/137.

USEPA, 1998c. Evaluation of Dredged Material Proposed for Discharge in Waters of the

U.S.-Testing Manual. USEPA and USACE. EPA 823-B-98-004. Washington, DC.

USEPA, 2000a. February. Bioaccumulation Testing and Interpretation for the Purpose of

Sediment Quality Assessment: Status and Needs. Bioaccumulation Analysis Workgroup,

Washington, D.C. EPA/823/R-00/001.

USEPA. 2000b. August. Guidance for the Data Quality Objectives Process. EPA QA/G-

4, EPA/600/R-96/055.

USEPA, 2001a. June. ECO Update. The Role of Screening-Level Risk Assessments

and Refining Contaminants of Concern in Baseline Ecological Risk Assessments. EPA

540/F-01/014, publication 9345.0-14.

EM 200-1-4

31 Dec 10

A-7

USEPA 2001b. December. Risk Assessment Guidance For Superfund: Volume III – Part

A, Process for Conducting Probabilistic Risk Assessment. EPA 540-R-02-002. OSWER

9285.7-45.

USEPA, 2002a. December. Calculating Upper Confidence Limits for Exposure Point

Concentrations at Hazardous Waste Sites. OSWER 9285.6-10.

USEPA. 2002b. December. Guidance on Choosing a Sampling Design for

Environmental Data Collection – for Use in Developing a Quality Assurance Project

Plan. EPA QA/G-5S. EPA/240/R-02/005.

USEPA, 2003. June. Non-Dioxin-Like PCBs: Effects and Consideration in Ecological

Risk Assessment. NCEA-C-1340, ERASC-003.

USEPA, 2004. July. Risk Assessment Guidance for Superfund, Volume I: Human

Health Evaluation Manual (Part E, Supplemental Guidance for Dermal Risk

Assessment)(RAGS E). Final. EPA/540/R/99/005. OSWER 9285.7-02EP.

USEPA 2005a. January. Procedures for the Derivation of Equilibrium Partitioning

Sediment Benchmarks (ESBs) for the Protection of Benthic Organisms: Metal Mixtures

(Cadmium, Copper, Lead, Nickel, Silver and Zinc). EPA/600/R-02/011.

USEPA, 2005b. September. Human Health Risk Assessment Protocol for Hazardous

Waste Combustion Facilities. EPA530-R-05-006.

USEPA, 2006. October. Framework for Application of the Toxicity Equivalence

Methodology for Polychlorinated Dioxins, Furans, and Biphenyls in Ecological Risk

Assessment. Final Draft. EPA/630/P-03/002A .

U.S. Department of Agriculture (USDA). 1983. Land Treatment of Hazardous Wastes.

J.F. Parr, P.M. Marsh, and J.M. Kla. (Eds.). Agricultural Environmental Quality, Agricultural

Research Service. Beltsville, Maryland.

U.S. Department of Energy (USDOE). 1984. A Review and Analysis of Parameters for

Assessing Transport of Environmentally Released Radionuclides Through Agriculture.

C.F. Baes III, R.D. Sharp, A.J. Sjoreen, and R.W. Shor (Eds.). Prepared by Oak Ridge

National Laboratory. ORNL-5786. September.

USDOE. Oak Ridge National Laboratory (ORNL). May 1998. A Guide to the ORNL

Ecotoxicological Screening Benchmarks: Background, Development, and Application.

Sample, Bradley E., Glenn W. Suter II, Rebecca A. Efroymson, and Daniel S. Jones.

May 1998. Revision 1.0. Oak Ridge National Laboratory, Oak Ridge, TN.

ORNL/TM-13615. Available at: http://www.esd.ornl.gov/programs/ecorisk/tools.html

This web site also provides access to the Screening Benchmarks Reports

on aquatic biota, wildlife, terrestrial plants, sediments and soil

invertebrates and microbial processes, and includes the following six

reports:

EM 200-1-4

31 Dec 10

A-8

Environmental Management and Enrichment Facilities. 1998. Radiological

Benchmarks for Screening Contaminants of Potential Concern for Effects

on Aquatic Biota at Oak Ridge National Laboratory, Oak Ridge,

Tennessee. Bechtel Jacobs Company LLC, Oak Ridge, TN. BJC/OR-80.

Suter, G.W. II, and C.L. Tsao. 1996. Toxicological Benchmarks for

Screening of Potential Contaminants of Concern for Effects on Aquatic

Biota on Oak Ridge Reservation: 1996 Revision. Oak Ridge National

Laboratory, Oak Ridge, TN. 104 pp. ES/ER/TM-96/R2.

Sample, B.E., D.M. Opresko, and G.W Suter II. 1996. Toxicological

Benchmarks for Wildlife: 1996 Revision. Oak Ridge National Laboratory,

Oak Ridge, TN. 227 pp. ES/ER/TM-86/R3.

Efroymson, R.A., M.E. Will, G.W. Suter II, and A.C. Wooten. 1997.

Toxicological Benchmarks for Screening Contaminants of Potential

Concern for Effects on Terrestrial Plants: 1997 Revision. Oak Ridge

National Laboratory, Oak Ridge, TN. 128 pp. ES/ER/TM-85/R3.

Jones, D.S., G.W. Suter II, and R.N. Hull. 1997. Toxicological

Benchmarks for Screening Contaminants of Potential Concern for Effects

on Sediment-Associated Biota: 1997 Revision. Oak Ridge National

Laboratory, Oak Ridge TN. 34 pp. ES/ER/TM-95/R4.

Efroymson, R.A., M.E Will, and G.W. Suter II. 1997. Toxicological

Benchmarks for Contaminants of Potential Concern for Effects on Soil

and Litter Invertebrates and Heterotrophic Processes: 1997 Revision. Oak

Ridge National Laboratory, Oak Ridge TN. ES/ER/TM-126/R2.

U.S. Department of Health and Human Services

Agency for Toxic Substances and Disease Registry (ATSDR). 1993. Toxicological Profile

for 1,1,1-Trichloroethane. U.S. Public Health Service. Atlanta, GA.

U.S. Department of the Interior (USDOI). 1991. Final Environmental Statement:

Vegetation Treatment on BLM Lands in Thirteen Western States. Bureau of Land

Management (BLM).

U.S. Department of the Navy. Naval Facilities Engineering Command (NAVFAC).

2000. Guide for Incorporating Bioavailability Adjustments into Human Health and

Ecological Risk Assessments at U.S. Navy and Marine Corps Facilities. Part 2:

Technical Background Document for Assessing Metals Bioavailability. NFESC User’s

Guide UG-2041-ENV. NAVFAC, Washington D.C.

U.S. Fish and Wildlife Service (USFWS). 1987. Type B Technical Information Document:

Guidance on Use of Habitat Evaluation Procedures, and Suitability Index Models for

CERCLA Application. National Technical Information Service. PB88-100151.

EM 200-1-4

31 Dec 10

A-9

USFWS. 1990. Evaluating Soil Contamination. Biological Report 90(2). Author: W.N.

Beyer. USDOI, Washington, D.C. 25 p.

Alloway, B.J. 1990. Heavy Metals in Soils. John Wiley and Sons, Inc. New York, New

York. 339p.

Amdur, M.O., J. Doull, and C.D. Klaasen (Eds.). 1991. Casarett and Doull's Toxicology:

The Basic Science of Poisons. 4th Ed. Pergamon Press. New York.

American Petroleum Institute (API). 1993. A Critical Review of Toxicity Values and an

Evaluation of the Persistence of Petroleum Products for Use in Natural Resource

Damage Assessments. Washington, D.C.

American Society for Testing and Materials (ASTM). 1995. Cleanup Criteria for

Contaminated Soils and Groundwater. Editor: Anthony J. Buonicore. ASTM Data Series

DS64.

ASTM. 2005. Standard Guide for Conducting Acute, Chronic, and Life-Cycle Aquatic

Toxicity Tests with Polychaetous Annelids. E1562-00.

Anderson, R.S., J.E. Doos, and F.L. Rose. 1982. "Differential Ability of Ambystoma

tigrinum Hepatic Microsomes to Produce Mutagenic Metabolites from Polycyclic Aromatic

Hydrocarbons and Aromatic Amines." Cancer Lett. Vo l. 16. pp. 33-41. In Eisler 1987.

Ankley, G.T., K. Lodge, D.J. Call, M.D. Balcer, L.T. Brooke, P.M. Cook, et al. 1992.

"Integrated assessment of contaminated sediments in the Lower Fox River and Green

Bay, Wisconsin." Ecotoxicol. Environ. Safety 23: 46-63.

Barbour, M.T., J. Gerritsen, B.D. Snyder, and J.B. Stribling. 1999. Rapid

Bioassessment Protocols for Use in Streams and Wadeable Rivers: Periphyton,

Benthic Macroinvertebrates and Fish, Second Edition. EPA 841-B-99-002. U.S.

Environmental Protection Agency; Office of Water; Washington, D.C.

Barnthouse, L.W., G.W. Suter, S.M. Bartell, J.J. Beauchamp, R.H. Gardener, E. Linder,

R.V. O'Neill, and A.E. Rosen. 1986. User's Manual for Ecological Risk Assessment.

Environmental Division. Prepared for Office of Research and Development, EPA.

Washington, D.C. Publication No. 2679. ORNL-6251.

Barnthouse, L.W., G.W. Suter, and A.E. Rosen. 1990. "Risks of Toxic Contaminants to

Exploited Fish Populations: Influence of Life History, Data Uncertainty, and Exploitation

Intensity." Environmental Toxicology and Chemistry. Vol. 9. pp. 297-311. In Hull and

Suter 1993.

Beyer, W.N. 1990. Evaluating Soil Contamination. Biological Report. Vol. 90. No. 2.

U.S. Department of the Interior, Fish and Wildlife Service. Washington, D.C.

Beyer, W.N., and C. Stafford. 1993. "Survey and Evaluation of Contaminants in

Earthworms and in Soils Derived from Dredged Material at Confined Disposal Facilities in

EM 200-1-4

31 Dec 10

A-10

the Great Lakes Region." Environmental Monitoring and Assessment. Vol. 24. pp. 151-

165.

Beyer, W.N., C. Stafford, and D. Best. 1993. "Survey and Evaluation of Contaminants in

Earthworms from Confined Disposal Facilities for Dredged Material in the Great Lakes."

Environmental Monitoring Assessment. Vo l. 24. pp. 151-165. In EPA 1993g.

Beyer, W.N., E.E. Connor, and S. Gerould. 1994. "Estimates of Soil Ingestion by

Wildlife." Journal of Wildlife Management. Vol. 58. No. 2. pp. 375-382.

Briggs, G.G., R.H. Bromilow, and A.A. Evans. 1982. "Relationship Between Lipophilicity

and Root Uptake and Translocation of Nonionized Chemicals by Barley." Pesticide

Science. Vol. 13. pp. 495-504. In Hull and Suter 1993.

Burrows, E.P., D.H. Rosenblatt, W. Mitchell, and D. Parmer. 1989. Organic Explosives

and Related Compounds: Environmental and Health Considerations. Technical Report

8901. U.S. Army Medical Bioengineering Research and Development Lab, Fort Detrick.

Frederick, MD.

Calder III, W.A., and E.J. Braun. 1983. "Scaling of Osmotic Regulation in Mammals and

Birds." American Journal of Physiology. Vol. 244. pp. 601-606.

Calderbank, A. 1989. "The Occurrence and Significance of Bound Pesticide Residues in

Soil." Reviews of Environmental Contamination and Toxicology, Vol. 108. Springer-

Verlag, Inc. New York, NY. pp. 71-103.

Callahan, C.A., M.A. Shirazi, and E.F. Neuhauser. 1994. "Comparative Toxicity of

Chemicals to Earthworms." Environmental Toxicology and Chemistry. Vo l. 13. No. 2.

pp. 291-298.

Canada Council of Ministers of the Environment (CCME). 1995. Protocol for the

Derivation of Canada Sediment Quality Guidelines for the Protection of Aquatic Life.

Guidelines Division, Evaluation and Interpretation Branch, Environment Canada, Ottawa.

Report CCME-98E. March.

Cataldo, D.A., S.D. Harvey, and R.J. Fellows. 1990. An Evaluation of the Environmental

Fates and Behavior of Munitions Material (TNT, RDX) in Soil and Plant Systems. Project

No. 88PP8853. Pacific Northwest Laboratory. Richland, Washington.

Chapman, P.M. 1989. "Current Approaches to Developing Sediment Quality Criteria."

Environmental Toxicology and Chemistry. Vol. 8. pp. 589-599.

Chapman, P.M. 1997. “Acid volatile sulfides, equilibrium partitioning, and hazardous

waste site sediments.” Environmental Management. Vol. 21; Issue 2; PBD: Mar-Apr

1997.

Clairain, E.J. and R.D. Smith. June 1, 1988. Environmental Effects of Dredging. The

Wetland Evaluation Technique (WET): A Technique for Assessing Wetland Functions and

Values. Technical Notes. USACE, ERDC, Vicksburg, MS. WES-EEDP-03-4.

EM 200-1-4

31 Dec 10

A-11

Clayton, G.D., and F.E. Clayton (Eds.). 1981. Patty's Industrial Hygiene and Toxicology:

Volume 2B - Toxicology. 3rd Ed. John Wiley and Sons. New York.

Cohen, J.E. 1978. Food Webs and Niche Space. Princeton University Press. Princeton,

New Jersey.

Connell, D.W. 1989. Bioaccumulation of Xenobiotic Compounds. CRC Press. Boca

Raton, Florida. In Suter 1993.

Debinski, D.M., and P.F. Brussard. 1992. "Biological Diversity Assessment in Glacier

National Park, Montana: I. Sampling Design." In McKenzie et al. 1992.

Drexler, J.W., W eis, C.P., W . Brattin. 2003. Lead B ioavailability: A validated in vitro

method. Fund. Appl. Toxicol.

Duinker, J.C., M.T.J. Hillebrand, and J.P. Boon. 1983. "Organochlorines in Benthic

Invertebrates and Sediments from the Dutch Wadden Sea; Identification of Individual

PCB Components." Neth. J. Sea Res. 17:19-38 (in Eisler 1986).

Dunbabin, J.S., and K.H. Bowmer. 1992. Potential Use of Constructed Wetlands for

Treatment of Industrial Wastewaters Containing Metals: The Science of the Total

Environment. Vol. III.

Edwards, N.T. 1983. "Polycyclic Aromatic Hydrocarbons (PAHs) in the Terrestrial

Environment - A Review." J. Environ Qual. Vol. 12 pp. 427-441. In ATSDR 1989.

Eisler, R. 1985a. Mirex Hazards to Fish, Wildlife, and Invertebrates: A Synoptic Review.

USFWS. Biological Report 85(1.1). Contaminant Hazard Reviews. Report No. 1.

March.

Eisler, R. 1985b. Cadmium Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 85(1.2). Contaminant Hazard Reviews. Report No.

2. July.

Eisler, R. 1985c. Carbofuran Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 85(1.3). Contaminant Hazard Reviews. Report No.

3. August.

Eisler, R. 1985d. Toxaphene Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 85(1.4). Contaminant Hazard Reviews. Report No.

4. August.

Eisler, R. 1985e. Selenium Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 85(1.5). Contaminant Hazard Reviews. Report No.

5. October.

Eisler, R. 1986a. Chromium Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 85(1.6). Contaminant Hazard Reviews. Report No.

6. January.

EM 200-1-4

31 Dec 10

A-12

Eisler, R. 1986b. Polychlorinated Bophenyl Hazards to Fish, Wildlife, and Invertebrates:

A Synoptic Review. USFWS. Biological Report 85(1.7). Contaminant Hazard Reviews.

Report No. 7. April.

Eisler, R. 1986c. Toxaphene Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 85(1.8). Contaminant Hazard Reviews. Report No.

8. May.

Eisler, R. 1986d. Mirex Hazards to Fish, Wildlife, and Invertebrates: A Synoptic Review.

USFWS. Biological Report 85(1.9). Contaminant Hazard Reviews. Report No. 9.

August.

Eisler, R. 1987a. Mercury Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 85(1.10). Contaminant Hazard Reviews. Report

No. 10. April.

Eisler, R. 1987b. Polycyclic Aromatic Hydrocarbon Hazards to Fish, Wildlife, and

Invertebrates: A Synoptic Review. USFWS. Biological Report 85(1.11). Contaminant

Hazard Reviews. Report No. 11. May.

Eisler, R. 1988. Arsenic Hazards to Fish, Wildlife, and Invertebrates: A Synoptic Review.

USFWS. Biological Report 85(1.12). Contaminant Hazard Reviews. Report No. 12.

January.

Eisler, R. 1988a. Chlorpyrifos Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report. Contaminant Hazard Reviews. Report No. 13.

March.

Eisler, R. 1988b. Lead Hazards to Fish, Wildlife, and Invertebrates: A Synoptic Review.

USFWS. Biological Report 85(1.14). Contaminant Hazard Reviews. Report No. 14.

April.

Eisler, R. 1989a. Mercury Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 85(1.15). Contaminant Hazard Reviews. Report

No. 15. January.

Eisler, R. 1989b. Index to Common and Scientific Names of Species Listed in

Contaminant Hazrd Reviews 1 through 15. USFWS. Biological Report 85(1.16).

Contaminant Hazard Reviews. Report No. 16. February.

Eisler, R. 1989c. Pentachlorophenol Hazards to Fish, Wildlife, and Invertebrates: A

Synoptic Review. USFWS. Biological Report 85(1.17). Contaminant Hazard Reviews.

Report No. 17. April.

Eisler, R. 1989d. Atrazine Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 85(1.8). Contaminant Hazard Reviews. Report No.

18. May.

EM 200-1-4

31 Dec 10

A-13

Eisler, R. 1989e. Molybdenum Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 85(1.19). Contaminant Hazard Reviews. Report

No. 19. August.

Eisler, R. 1990a. Boron Hazards to Fish, Wildlife, and Invertebrates: A Synoptic Review.

USFWS. Biological Report 85(1.20). Contaminant Hazard Reviews. Report No. 20.

April.

Eisler, R. 1990b. Chlordane Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 85(1.21). Contaminant Hazard Reviews. Report

No. 21. July.

Eisler, R. 1990c. Paraquat Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 85(1.22). Contaminant Hazard Reviews. Report

No. 22. August.

Eisler, R. 1991. Cyanide Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 85(1.23). Contaminant Hazard Reviews. Report

No. 23. December.

Eisler, R. 1992a. Fenvalerate Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 24. Contaminant Hazard Reviews. Report No. 24.

May.

Eisler, R. 1992b. Deflubenzuron Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USFWS. Biological Report 4. Contaminant Hazard Reviews. Report No. 25.

June.

Eisler, R. 1993. Zinc Hazards to Fish, Wildlife, and Invertebrates: A Synoptic Review.

USFWS. Biological Report 10. Contaminant Hazard Reviews. Report No. 26. April.

Eisler, R. 1994a. Famphur Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. National Biological Survey. Biological Report 20. Contaminant Hazard Reviews.

Report No. 27. February.

Eisler, R. 1994b. Acrolein Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. National Biological Survey. Biological Report 23. Contaminant Hazard Reviews.

Report No. 28. June.

Eisler, R. 1994c. Radiation Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. U.S. National Biological Service. Biological Report 26. Contaminant Hazard

Reviews. Report No. 29. December.

Eisler, R. 1995. Sodium Monofluoroacetate (1080) Hazards to Fish, Wildlife, and

Invertebrates: A Synoptic Review. U.S. National Biological Service. Biological Report 27.

Contaminant Hazard Reviews. Report No. 30. February.

EM 200-1-4

31 Dec 10

A-14

Eisler, R. 1996a. Planar PCB Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. U.S. National Biological Service. Biological Report 31. Contaminant Hazard

Reviews. Report No. 31. August.

Eisler, R. 1996b. Silver Hazards to Fish, Wildlife, and Invertebrates: A Synoptic Review.

U.S. National Biological Service. Biological Report 32. Contaminant Hazard Reviews.

Report No. 32. September.

Eisler, R. 1998a. Copper Hazards to Fish, Wildlife, and Invertebrates: A Synoptic

Review. USGS. Biological Science Report USGS/BRD/BSR—1997-0002. Contaminant

Hazard Reviews. Report No. 33. January.

Eisler, R. 1998b. Nickel Hazards to Fish, Wildlife, and Invertebrates: A Synoptic Review.

USGS. Biological Science Report USGS/BRD/BSR-1998-0001. Contaminant Hazard

Reviews. Report No. 34. April.

Eisler, R. 1999. Cumulative Index to Chemicals and to Common and Scientific Names of

Species Listed in Contaminant Hazard Reviews 1 through 34. USGS. Biological Science

Report USGS/BRD/BSR--1999-0002. Contaminant Hazard Reviews. Report No. 35.

March.

Federal Focus Inc. 1994. A Blueprint for Constructing a Credible Environmental Risk

Assessment Policy in the 104th Congress. Washington, D.C.

Fordham, C.L., and D.P. Reagan. 1991. "Pathways Analysis Method for Estimating

Water and Sediment Criteria at Hazardous Waste Sites." Environ. Toxicol. and

Chemistry. Vol. 10. No. 7. pp. 949-960.

Gilbert, R.O. 1987. Statistical Methods for Environmental Pollution Monitoring. Van

Nostrand Reinhold. New York.

Gile, J.D., J.C. Collins, and J.W. Gillet. 1982. "Fate and Impact of Wood Preservatives in

a Terrestrial Microcosm." J. Agric. Food Chem

. Vo l. 30. pp. 295-301. In ATSDR 1989.

Goon, D., N.S. Hatou, M.J. Klan, J.D. Jernigan, and R.G. Farmer. 1991. "Oral

Bioavailability of "Aged" Soil-adsorbed Benzo(a)pyrene (BaP) in Rats." Society of

Toxicology Annual Meeting Abstracts #1356.

Gough, L.P., H.T. Shacklette, and A.A. Case. 1979. Element Concentrations Toxic to

Plants, Animals and Man. U.S. Geologic Survey Bulletin No. 14. United States

Government Printing Office. Washington, D.C.

Halvorson, Curtis H. 1984. Long-term Modeling of Small Vertebrates: A Review with

Suggestions. In Morrison 1988.

Henningsen, G. 2003. In Vivo and In Vitro Assessment of Bioavailability of Soil Lead.

Presentation at the USEPA Bioavailability Workshop. April.

EM 200-1-4

31 Dec 10

A-15

Henny, C.J., L.J. Blus, A.J. Krynitsky, and C.M. Bunck. 1984. "Current impact of DDE on

Black-Crowned Night-Herons in the Intermountain West." J. Wildl. Manage. 48:1-13. In

Eisler 1986.

Hill, A.B. 1965. The Environment and Disease: Association or Causation? Proceedings

of the Royal Society of Medicine. 58:295-300.

Hull, R.N., and G.W. Suter. 1993. Decreasing the Risk of an Unacceptable Ecological

Risk Assessment. Oak Ridge National Laboratory, Environmental Sciences Division, Oak

Ridge, Tennessee; and David A. Belluck, Minnesota Pollution Control Agency, St. Paul,

Minnesota.

Hulzebros, E.M., D.M.M. Adema, and E.M. Dirven-Van Breemen. 1993. "Phytotoxicity

Studies with Latuca Sativa in Soil and Nutrient Solution." Environmental Toxicology and

Chemistry. Vol. 12. pp. 1709-1094.

Johnson, M.L., D.G. Huggins, and F. DeNoyelles, Jr. 1991. "Ecosystem Modeling with

LISREL: A New Approach for Measuring Direct and Indirect Effects." Ecological

Applications. Vol. 1. No. 4. pp. 383-398.

Kabata-Pendias, A., and H. Pendias. 1984. Trace Elements in Soils and Plants. CRC

Press. Boca Raton, Florida.

Karr, J.R., K.D. Fausch, P.L. Angermeier, P.R. Yant, and I.J. Schlosser. 1986. "Assessing

Biological Integrity in Running Waters: A Method and Its Rational." Illinois Natural History

Survey. Special Publication No. 5.

Keith, Lloyd B. 1983. "Role of Food in Hare Populations Cycles." Oikos 40: 385-395. In

Morrison 1988.

Kimball, K.D., and S.A. Levin. 1985. "Limitations of Laboratory Bioassays: The Need for

Ecosystem-Level Testing." BioScience. Vol. 35. No. 3.

Layton, D., B. Mallon, W. Mitchell, L. Hall, et al. 1987. Conventional Weapons

Demilitarization: A Health and Environmental Effects Data Base Assessment. Produced

by Lawrence Livermore National Laboratory for U.S. Army Medical Research and

Development Command, Fort Detrick, Federal, MD. AD/UCRL-21109.

Long, E.R., and L.G. Morgan. 1991. The Potential for Biological Effects of Sediment-

Sorbed Contaminants Tested in the National Status and Trends Program. NOAA

Technical Memorandum NOS OMA 52. Seattle, Washington.

Long, E.R., D.D. MacDonald, S.L. Smith, and F.D. Calder. 1995. Incidence of Adverse

Biological Effects within Ranges of Chemical Concentrations in Marine and Estuarine

Sediments. Environmental Management, Vol. 19, No.1, pp.81-97.

Ludwig, J.A. and J.F. Reynolds. 1988. Statistical Ecology. John Wiley and Sons. New

York, N Y

EM 200-1-4

31 Dec 10

A-16

MacDonald, D.D. 1994. Approach to the Assessment of Sediment Quality in Florida

Coastal Waters. Vol. I: Developmental Evaluation of Sediment Quality Assessment

Guidelines. Vol. II: Application of the Sediment Quality Assessment Guidelines. Vol. III:

Supporting Documentation: Biological Effects Database for Sediments. Vol. IV:

Supporting Documentation: Regional Biological Effects Database for Sediments.

Prepared for Florida Dept. of Environmental Protection. Tallahassee, FL. November.

MacIntosh, L., G.W. Suter II, and F.O. Hoffman. 1994. "Uses of Probabilistic Exposure

Models in Ecological Risk Assessments of Contaminated Sites." Risk Analysis. Vol. 14,

No. 4, pp. 405-418.

Maughan, James T. 1993. Ecological Assessment of Hazardous Waste Sites. Van

Nostrand Reinhold, New York, NY. 352 p.

McDowell, L.R. 1992. Minerals in Animal and Human Nutrition. Academic Press, San

Diego, CA. 524 pp.

Morrison, Michael. 1988. "The Design and Importance of Long-Term Ecological Studies:

Analysis of vertebrates in the Inyo-White Mountains, California." In: Management of

Amphibians, Reptiles, and Small Mammals in North America. Proceedings of

Symposium, Flagstaff, Arizona, July 19-21 (1988). U.S. Forest Service, Rocky Mountain

Forest and Range Experiment Station, CO.

Mueller-Dombois, D., and H. Ellenberg. 1974. Aims and Methods of Vegetation Ecology.

John Wiley and Sons. New York.

Nagy, K.A. 1987. "Field Metabolic Rate and Food Requirement Scaling in Mammals and

Birds." Ecological Society of America. Ecological Monographs. Vol. 57. No. 2. pp. 111-

128. Arizona State University, Tempe, Arizona.

Nagy, K.A. 2001. Food requirements of wild animals: predictive equations for free-living

mammals, reptiles, and birds. Nutrition Abstracts and Reviews, Series B: Livestock

Feeds and Feeding. 71:1-12

National Academy of Sciences (NAS). 1980. Mineral Tolerance of Domestic Animals.

Subcommittee on Mineral Toxicity in Animals, Committee on Animal Nutrition, National

Research Council.

NAS. 1983. National Research Council (NRC). Risk Assessment in the Government:

Managing the Process. National Academy Press/National Academy of Science.

Washington, D.C.

NAS. 1993. National Research Council (NRC). Science and Judgment in Risk

Assessment. National Academy Press/National Academy of Science. Washington, D.C.

NAS. 1994. National Research Council (NRC). Issues in Risk Assessment.

Washington, D.C. National Academy Press/National Academy of Science.

EM 200-1-4

31 Dec 10

A-17

National Institute for Occupational Safety and Health (NIOSH). 1987. Registry of Toxic

Effects of Chemical Substances (RTECS). R.J. Lewis and R.L. Tolkin (Eds.). U.S.

Department of Health and Human Services, Center for Disease Control. Publ. No. 79-

100.

National Oceanic and Atmospheric Administration (NOAA). 1992. The Coastal Resource

Coordinator's Bioassessment Manual. Authors: MacDonald, D.A., M.B. Matta, L.J. Field,

C. Cairncross and M. Munn (EVS Consultants). NOAA/Hazard Materials Response and

Assessment Division (HMRAD), Report No. HAZMAT 93-1. Seattle, WA.

Newsted, J.L., J.P. Giesy, G.T. Ankley, D.E. Tillitt, et al. 1995. Development of Toxic

Equivalency Factors for PCB Congeners and the Assessment of TCDD and PCB

Mixtures in Rainbow Trout. Environmental Toxicology and Chemistry, Vol.14, No.5,

pp. 861-872.

Norheim, G., and B. Kjos-Hanssen. 1984. "Persistent Chlorinated Hydrocarbons and

Mercury in Birds Caught off the West Coast of Spitzbergen." Environ. Pollut. 33A:143-

152 (In Eisler 1986).

Noss, R.F. 1990. "Indicators for Monitoring Biodiversity: A Hierarchical Approach." EPA.

In Mouat et al. 1992.

O'Conner, G.A. 1989. Degradation, Crop Uptake, and Risk of Micropollutants in Sewage

Sludge. Proceedings of Sewage and Sludge Conference (Qual. Aspects and Risk in

Connection with Land Application), Swedish Water and Wastewater Association. April.

O'Conner, G.A., R.L. Chaney, J.A. Ryan. 1990. "Bioavailability to Plants of Sludge-Borne

Toxic Organics." In Ware 1991.

Outridge, P.M., and B.N. Noller. 1991. "Accumulation of Toxic Trace Elements by

Freshwater Vascular Plants." In Ware 1991.

Overcash, M.R. and D. Pal. 1979. Design of Land Treatment Systems for Industrial

Wastes - Theory and Practice. Ann Arbor Science Publishers, Inc. Ann Arbor, Mich.

Persaud, D., R. Jaugumagi, and A. Hayton. 1992. Guidelines for the Protection and

Management of Aquatic Sediment Quality in Ontario. Water Resources Branch. Ontario

Ministry of the Environment, 1 St. Clair Ave. W., Toronto, Ontario. June. ISBN 0-7729-

9248-7

Pielou, E.C. 1975. Ecological Diversity. John Wiley and Sons. New York.

Porebski, L.M., K.G. Doe, B.A. Zajdik, D. Lee, P. Pockington, J.M. Osborne. 1999.

“Evaluating the Techniques for a Tiered Testing Approach to Dredged Sediment

Assessment – A Study over a Metal Concentration Gradient” Environmental Toxicology

and Chemistry; Vol 18; Issue: 11; PBD: Nov 1999.

Rand, G.M., and S.R. Petrocelli. 1985. Fundamentals of Aquatic Toxicology, Methods

and Applications. Hemisphere Publishing Corp., New York, NY, 666 p.

EM 200-1-4

31 Dec 10

A-18

Ross, D.A. 1989. Amphibians and Reptiles in the Diets of North American Raptors.

Bureau of Endangered Resources, Wisconsin Department of Natural Resources.

Madison, Wisconsin.

Ruby, M.V. 2003. Application of In Vitro Extraction Tests to Estimate Lead

Bioavailability from Soil. Presentation at the USEPA Bioavailability Workshop. April.

Safe, S. 1990. "Polychlorinated Biphenyls (PCBs), Dibenzo-p-dioxins (PCDDs),

Dibenzofurans (PCDFs), and Related Compounds: Environmental and Mechanistic

Considerations Which Support the Development of Toxic Equivalency Factors (TEFs)."

Critical Reviews in Toxicology. Vo l. 21. No. 1. pp. 51-71.

Sawhney, B.L. 1988. "Chemistry and Properties of PCBs in Relation to Environmental

Effects." In Waid. 1990.

Shu, H., D. Paustenbach, F.J. Murray, L. Marple, B. Brunck, D. Dei Rossi, and P.

Teitelbaum. 1988. "Bioavailability of Soil-Bound TCDD: Oral Bioavailability in the Rat."

Fundam. Appl. Toxicol. 10:648-654.

Schulz, T.W. and S. Griffin (1999). Estimating risk assessment exposure point

concentrations when the data are not normal or lognormal. Risk Analysis 19:577-584.

Singh, A.K., A. Singh, and M. Engelhardt (1997). The lognormal distribution in

environmental applications. EPA/600/R-97/006

Simini, M., R. Checkai, M.E. Maly. 2000. February. Tri-Service Remedial Project

Manager’s Technical Handbook for Ecological Risk Assessment. Final. SDIM-AEC-

ER-CR-200015.

Sims, R.C., and R. Overcash. 1983. "Fate of Polynuclear Aromatic Compounds (PNAs)

in Soil-Plant Systems." Residue Rev

. Vo l. 88. pp. 1-68. In Eisler 1987.

Stewart, A.J., G.J. Haynes, and M.I. Martinez. 1992. "Fate and Biological Effects of

Contaminated Vegetation in a Tennessee Stream." Environmental Toxicology and

Chemistry. Vol. 2.

Strek, H.J., J.B. Weber, P.J. Shea, E.J. Mrozek, and R. Overcash. 1981. "Reduction of

Polychlorinated Biphenyl Toxicity and Uptake of Carbon-14 Activity by Plants Through the

Use of Activated Carbon." Journal of Agriculture and Food. Vo l. 29.

Suter, G.W. 1993. Ecological Risk Assessment. Lewis Publishers. Chelsea, Michigan.

Tannenbaum, L.V., Johnson, M.S., and Bazar, M., 2003. Application of the Hazard

Quotient Method in Remedial Decisions: A Comparison of Human and Ecological Risk

Assessments. Human and Ecological Risk Assessment, Volume 9 (1): 387-401.

Texas Natural Resource Conservation Commission (TNRCC; now Texas Commission on

Environmental Quality, TCEQ), 2001. December. Guidance for Conducting Ecological

Risk Assessments at Remediation Sites in Texas. RG-263 (Revised).

EM 200-1-4

31 Dec 10

A-19

Tiffin, L.O. 1977. "The Form and Distribution of Metals in Plants: An Overview."

Proceedings of the Fifteenth Annual Hanford Life Sciences Symposium - Biological

Implications of Metals in the Environment. Richland, Washington. Sept. 29 to Oct. 1,

1975. Published by Technical Information Center, Energy Research and Development

Administration.

Topp, E., I. Schenert, A. Attar, A. Korte, and F. Korte. 1986. "Factors Affecting the Uptake

of C

-Labelled Organic Chemicals by Plants from Soil." Ecotoxicol. Environ. Saf. Vol. 11.

P. 219. In Suter 1993.

Travis, C.C., and A.D. Arms. 1988. "Bioconcentration of Organics in Beef, Milk, and

Vegetation." Environ. Sci. Technol. Vol. 22. pp. 271-274. In Suter 1993.

Van Gestel, C.A.M., and W.C. Ma. 1988. "Toxicity and Bioaccumulation of

Chlorophenols in Earthworms in Relation to Bioavailability in Soil." Ecotoxicol. Environ.

Saf. Vol. 15. pp. 289-297. In Suter 1993.

Van Leeuwen, C.J., P.T. Van Der Zandt, T. Aldenberg, H.J. Verhaar, and J.L. Hermens.

1992. "Application of QSARs, Extrapolation and Equilibrium Partitioning in Aquatic

Effects. Assessment I - Narcotic Industrial Pollutants." Environ. Tox. and Chem. Vol. 2.

pp. 267-282.

Vaughn, B.E. 1984. "State of Research: Environmental Pathways and Food Chain

Transfer." Environ. Health Perspect. pp. 353-371. In ATSDR 1989.

Washington State Department of Ecology. 1991. Sediment Management Standards,

Chapter 173-204 WAC (April) and Summary of Criteria and Guidelines for Contaminated

Freshwater Sediments. Sediment Management Unit (September). Olympia, WA.

Water Environmental Research Foundation (WERF). 1994. Methodology for Aquatic

Ecological Risk Assessment. Draft Final Report. April 25. Prepared by B.R. Parkhurst,

W. Warren-Hicks and R.D. Cardwell. WERF, Alexandria, Virginia.

Wentsel, R.S., T.W. LaPoint, M. Simini, R. Checkai, D. Ludwig and L. Brewer. 1994.

Procedural Guidelines for Ecological Risk Assessments at U.S. Army Sites. U.S. Army

Environmental Center, Aberdeen Proving Ground, MD, 21010.

Wentsel, R.S., T.W. LaPoint, M. Simini, R. Checkai, D. Ludwig and L. Brewer. 1996.

May. Tri-Service Procedural Guidelines for Ecological Risk Assessments. Air Force

Center for Environmental Excellence (AFCEE), U.S. Army Environmental Center

(USAEC), Naval Facilities Engineering Service Center (NFESC).

Wheatley, G.O., and J.O. Hardman. 1968. "Organochlorine Insecticide Residue in

Earthworms from Arable Soils." J. Sci. Food Agric. Vol. 19. pp. 219-229. In Suter 1993.

Wiemeyer, S.N., T.G. Lamont, C.M. Bunck et al. 1984. Organochlorine pesticide,

polychlorobiphenyl, and mercury residues in bald eagles - 1966-1979 - and their

EM 200-1-4

31 Dec 10

A-20

relationships to shell thinning and reproduction. Arch. Environ. Contam. Toxicol. 13:529-

549 (In Eisler 1986).

Winter, S., and B. Streit. 1992. "Organochlorine Compounds in a Three-Step Terrestrial

Food Chain." Chemosphere. Vol. 24. No. 12. pp. 1765-1774.

Zar, J.H. 1984. Statistical Analysis. Prentice-Hall Inc. Englewood Cliffs, NJ. 718 p.

A.2. Related Publications

The following list represents additional sources of information related to this

manual.

EO 11990 (Presidential Document). May 24, 1977. Protection of Wetlands. 42 FR

26961.

EO 11988 (Presidential Document). May 24, 1977. Floodplain Management. 42 FR

26951.

EO 12088 (Presidential Document). October 13, 1978. Federal Compliance with

Pollution Control Standards. 43 FR 47707.

EO 12498 (Presidential Document). January 8, 1985. Regulatory Planning Process.

50 FR 1036.

EO 12898 (Presidential Document). February 16, 1994. Federal Actions to Address

Environmental Justice in Minority Populations and Low-Income Populations. 59 FR No.

32.

USDOD. 1978. Directive 6230.1, Safe Drinking Water. April.

USDOD. 1979. Directive 6050.1, Environment Effects in the United States of

Department of Defense Actions. July 30.

USDOD. 1979. Directive 6060.7, Environment Effects Abroad of Major Department of

Defense Actions. March 31.

USDOD. 1993. Memorandum from the Deputy Secretary of Defense to Addressees;

Subj: Fast Track Cleanup at Closing Installations. Washington, D.C. September 9.

USDOD. 1994. Memorandum from the Deputy Secretary of Defense to Addressees;

Subj: Fast Track Cleanup - Finding of Suitability to Transfer (FOST) for BRAC Property.

Washington, D.C.

USA. 2007. Army Regulation 200-1, Environmental Quality, Environmental Protection,

and Enhancement.

USA. 1991 (February 25). Health Risk Assessment Guidance for the Installation

Restoration Program and Formerly Used Defense Sites. Pamphlet 40-578. Washington,

D.C.

EM 200-1-4

31 Dec 10

A-21

USACE. 1992. USACE HTRW Management Plan. Draft. Washington, D.C.

USACE. 1995. Engineering and Design, Chemical Data Quality Management for

Hazardous, Toxic, and Radioactive Waste Remedial Activities. ER 1110-1-263. October.

USEPA. 1986. RCRA Facility Assessment Guidance. EPA/530-SW-86-053.

USEPA. 1986. Superfund Remedial Design and Remedial Action Guidance. Office of

Emergency Response and Remedial Design, Washington, D.C. OSWER Directive

9355.0-4A. June.

USEPA. 1987. Superfund Selection of Remedy. Office of Solid Waste and Emergency

Response. August.

USEPA. 1988. Guidance for Conducting Remedial Investigations and Feasibility Studies

under CERCLA. Interim Final. OSWER Directive 9355.3-01.

USEPA. 1988. RCRA Corrective Action Interim Measures Guidance. EPA/530-SW-88-

029.

USEPA. 1988. RCRA Corrective Action Plan. EPA/530-SW-88-028.

USEPA. 1988. Federal Facilities Compliance Strategy. Office of Federal Activities, Office

of External Affairs, Washington, D.C. 20460. November.

USEPA. 1988. Enforcement Actions under RCRA and CERCLA at Federal Facilities.

USEPA. 1988. Evaluation Process for Achieving Federal Facility Compliance.

USEPA. 1988. Estimating Toxicity of Industrial Chemicals to Aquatic Organisms Using

Structure Activity Relationships. Office of Substances. EPA/560/6-88/001.

USEPA. 1988. Guidance on Remedial Actions for Contaminated Groundwater at

Superfund Sites. Office of Emergency and Remedial Response, Washington, D.C.

EPA/540/G-88/003.

USEPA. 1988. Interim Procedures for Estimating Risks Associated with Exposures to

Mixtures of Chlorinated Dibenzo-p-dioxins and Dibenzofurans (CDDs and CDFs).

Update. Risk Assessment Forum. EPA/625/3-89/016. March.

USEPA. 1989. RCRA Facility Investigation (RFI) Guidance. Development of an RFI

Work Plan and General Considerations for RCRA Facility Investigations. Waste

Management Division, Office of Solid Waste. Interim Final G-89-00075.

USEPA. 1989. Federal Facilities Negotiations Policy. Office of Emergency and

Remedial Response, Washington, D.C. OSWER Directive 9992.3.

USEPA. 1989. Statistical Analysis of Groundwater Monitoring at RCRA Facilities. Interim

Final Guidance. Office of Solid Waste, EPA/530-SW-89-026.

EM 200-1-4

31 Dec 10

A-22

USEPA. 1989. Ecological Risk Assessment Methods: A Review and Evaluation of Past

Practices in the Superfund and RCRA Programs, EPA/600/8-89/043. Office of Policy

Planning and Evaluation. June.

USEPA. 1989. Sediment Classification Methods Compendium. Draft Final. Watershed

Protection Division, Washington, D.C.

USEPA. 1990. The Revised Hazard Ranking System: Background Information.

OSWER Directive 9320.7-03FS. 55 FR 51532-51667. December 14.

USEPA. 1990. Federal Facilities Hazardous Waste Compliance Manual. Office of

Waste Programs Enforcement, Washington, D.C. OSWER Directive 9992.4.

USEPA. 1991. Guidance for Performing Preliminary Assessments Under CERCLA.

Office of Emergency and Remedial Response, EPA/600/6-91/010. OSWER Directive

9345.0-01A.

USEPA. 1991. Regional Guidance for Conducting Ecological Assessments. Region VI.

USEPA. 1991. Streamlining Superfund Studies: Using the Data Quality Objective

Process for Soil Scoping. Quality Assurance and Management Staff, Office of Research

and Development.

USEPA. 1991. Technical Guidance on Chemical Concentration Data Near the Detection

Limit. Region III Technical Guidance Manual. November.

USEPA. 1992. Guidance for Performing Site Inspections Under CERCLA. Interim Final.

Office of Emergency and Remedial Response, Washington, D.C. OSWER Directive

9345.0-05.

USEPA. 1992. Hazard Ranking System Guidance. Interim Final.

USEPA. 1992. Regional Guidelines for Conducting Ecological Assessments. Region V.

April.

USEPA. 1992. Health Effects Assessment Summary Tables, Annual FY 92. Prepared by

Environmental Criteria and Assessment Office for Office of Emergency and Remedial

Response. OERR 9200.6-303. Cincinnati, Ohio. March.

USEPA. 1992. Dermal Exposure Assessment: Principles and Applications. Interim

Report. Office of Research and Development. EPA/600/8-91/001B.

USEPA. 1992. Guidance on Setting Priorities for NPL Candidate Sites. OSWER

Directive 9203.1-06. October 28.

USEPA. 1993. Superfund Program Draft Checklist for Ecological Assessment/Sampling.

Emergency Response Branch, OSWER. January.

EM 200-1-4

31 Dec 10

A-23

USEPA. 1993. Technical Basis for Deriving Sediment Quality Criteria for Nonionic

Organic Contaminants for the Protection of Benthic Organisms by Using Equilibrium

Partitioning. Washington, D.C., Office of Water. EPA/822/R-93/001.

USEPA. 1994. Interim Ecological Risk Assessment Guidelines. Region III. July.

USEPA. 1994. Evaluating and Identifying Contaminants of Concern for Human Health.

EPA Region VIII Superfund Tech. Guidance RA-03.

USEPA. 1995. Land Use in CERCLA Remedy Selection Process. OSWER Directive

No. 9355.7-04.

U.S. Air Force (USAF). 1989. The Air Force Installation Restoration Program

Management Guidance. Washington, D.C.

USAF. 1990. General Guidance for Ecological Risk Assessment at Air Force

Installations. Prepared for Human Systems Division/IRP Program Office. Brooks Air

Force Base, TX. Authors: DeSesso and Price.

USAF. 1991. The Air Force Hazardous Waste Management Policy HQ USAF.

Washington, D.C.

USAF. 1992. FY 93/94/95 DERA Eligibility and Programming Guidance. HQ USAF.

Washington, D.C.

U.S. Department of the Navy. 1992. The Navy and Marine Corps Installation Restoration

(IR) Program Manual. Chief of Naval Operations (CNO). Washington, D.C.

U.S. Department of the Navy. 1994 (November 1). Environmental and Natural

Resources Program Manual. OPNAVINST 5090.1B. Washington, D.C.

ATSDR. 1989. Toxicological Profile for Polycyclic Aromatic Hydrocarbons. Draft. U.S.

Department of Health and Human Services, Public Health Service. Prepared by Clement

Associates, Inc. Contract No. 205-88-0608. October.

Adams, W.J., R.A. Kimberle, and J.W. Barnett, Jr. 1992. "Sediment Quality and Aquatic

Life Assessment." Environ. Sci. Technol. Vol. 26. No. 10. pp. 1864-1875.

Bartholomew, G.A., and T.J. Cade. 1963. "The Water Economy of Land Birds." Auk.

Vol. 80. pp. 504-539. In EPA 1993e.

Burmaster, E. and P.D. Anderson. 1994. "Principles of Good Practice for the Use of

Monte Carlo Techniques in Human Health and Ecological Risk Assessments." Risk

Analysis. Vol. 14, No. 4, pp. 4055-419.

California Department of Toxic Substances Control (CAL EPA). 1994. Guidance for

Ecological Risk Assessment at Hazardous Waste Sites and Permitted Facilities. Part A:

Overview; Part B: Scoping Assessment; State of California, California Environmental

Protection Agency.

EM 200-1-4

31 Dec 10

A-24

Conover, S., K.W. Strong, T.E. Hickey, and F. Sander. 1985. "An Evolving Framework for

Environmental Impact Analysis. I-Methods." Journal of Environmental Management.

Vol. 21. pp. 343-358.

Dietz, D.D. et al. 1991. "Toxicity Studies of Acetone Administered in the Drinking Water

of Rodents." Fundam. Appl. Toxicol. 17:347-360. In ATSDR 1993.

Environment Canada. 1994. A Framework for Ecological Risk Assessment at

Contaminated Sites in Canada: Review and Recommendations. Authors: C. Gaudot

(EVS) and ESSA Environmental and Social Systems Analysts. Environmental Quality

Division, Evaluation and Interpretation Branch. Ecosystem Conservation Directorate

ISBN 0-662-22156-7.

Franson, J.C., L. Sileo, O.H. Pattee, and J.F. Moore. 1983. "Effects of Chronic Dietary

Lead in American Kestrels (Falco sparverius)." J. Wildlife Dig. 19:100-113. In Eisler

1988.

Jorschgen, L.J. 1970. "Soil-Food Chain-Pesticide Wildlife Relationships in Aldrin-Treated

Fields." Journal of Wildlife Management

. Vol. 34. No. 1.

Lemly, A.D. 1993. "Guidelines for Evaluating Selenium Data from Aquatic Monitoring and

Assessment Studies." Environmental Monitoring and Assessment. Vo l. 28:. pp. 83-100.

MacDonald, D.A., M.B. Matta, L.J. Field, C. Cairncross and M. Munn (EVS Consultants).

December 1992. The Coastal Resource Coordinator's Bioassessment Manual.

NOAA/Hazard Materials Response and Assessment Division (HMRAD), Report No.

HAZMAT 9 3-1. Seattle, WA.

Massachusetts Department of Environmental Protection (MDEP). 1994. Environmental

Risk Characterization Guidance. January.

McKenzie, L.D., D. Hyatt, and V. McDonald (Eds.). 1992. Ecological Indicators. Vols. I

and II. Elsevier Applied Science. New York.

Mouat, D.A., C.A. Fox, and M.R. Rose. 1992. "Ecological Indicator Strategy for

Monitoring Arid Ecosystems." In McKenzie et al. 1992.

National Toxicology Program (NTP). 1991. National Toxicology Program - Technical

Report No. 3: Toxicity Studies of Acetone in F344/N Rats and B6C3F1 Mice (Drinking

Water Studies). NIH PB91-3122. Research Triangle Park, N.C. U.S. Department of

Health and Human Services, Public Health Service, National Institute of Health. In

ATSDR 1993.

Neptune, Dean (w/EPA). no date. Streamlining Superfund Soil Studies: Using the Data

Quality Objectives Process for Scoping.

New York Department of Environmental Conservation (NYDEC). 1991. Fish and Wildlife

Impact Analysis for Inactive Hazardous Waste Sites. June.

EM 200-1-4

31 Dec 10

A-25

Newell, A.J., D.W. Johnson, and L.K. Allen. 1987. Niagara River Biota Contamination

Project: Fish Flesh Criteria for Piscivorous Wildlife. New York State, Dept. of

Environmental Conservation Tech. Report 87-3.

Norton, S., M.McVey, J. Colt, J. Durda and R. Hegner. 1988. Review of Ecological Risk

Assessment Methods. Prepared for USEPA, Office of Planning and Evaluation, ICF, Inc.,

Fairfax, Virginia.

Parkhurst, B.R., H.L. Bergmann, M.D. Marcus, C.S. Creger, W. Warren-Hicks, H. Olem,

A. Boelter and J.P. Baker. 1990. Evaluation of Protocols for Aquatic Ecological Risk

Assessment and Risk Management. Prepared for WPCF Research Foundation,

Technology Assessment Dept., Alexandria, Virginia.

Pastorok, R.J. and J.R. Sampson. 1990. Review of Ecological Risk Assessment

Methods to Develop Numerical Criteria for Cleanup of Hazardous Waste Sites. Prepared

for Washington Dept. of Ecology.

Pennsylvania Department of Environmental Resources (DER). 1991. Risk Assessment

Guidelines for Facilities Burning Hazardous Waste.

Pye, L.H. 1993. "Decision Analysis/Risk Analysis (DARA) Methodology for Evaluating

Ecological Risk at Federal Facilities." Presentation at 14th Annual Meeting of Society of

Environmental Toxicology and Chemistry. Houston, Texas.

Royce, C.L., J.S. Fletcher, and P.G. Risser. 1984. "Phytotox: A Database Dealing with

the Effect of Organic Chemicals on Terrestrial Vascular Plants." Journal of Chemical

Information and Computer Sciences. Vol. 24. pp. 7-10.

Stakhiv, E.Z. 1988. "An Evaluation Paradigm for Cumulative Impact Analysis."

Environmental Management

. Vol. 12. No. 5. pp. 725-748.

Tannenbaum, L.V., 2005. Two Simple Algorithms for Refining Mammalian Receptor

Selection in Ecological Risk Assessments. Integrated Environmental Assessment and

Management, Volume 1(3): 290-298.

Tannenbaum, L.V., 2005. A Critical Assessment of the Ecological Risk Assessment

Process: A Review of Misapplied Concepts. Integrated Environmental Assessment and

Management, Volume 1(1): 66-72

Waid, J.J. (Ed.). 1990. PCBs and the Environment. Vol. 1. CRC. Boca Raton, Florida.

Ware, G.W. (Ed.). 1991. Reviews of Environmental Contamination and Toxicology. Vol.

121.

EM 200-1-4

31 Dec 10

A-26

THIS PAGE INTENTIONALLY LEFT BLANK

Glossary-1

EM 200-1-4

31 Dec 10

GLOSSARY

Acronyms

ACL

alternative concentration limit

AET

apparent effects threshold

AOC

Area of Concern

API

American Petroleum Institute

AQUIRE

Aquatic Information Retrieval Database

ARAR

ARAR applicable or relevant and appropriate requirement

ASTER

Assessment Tools for the Evaluation of Risk

ASTM

American Society for Testing and Materials

ATSDR

Agency for Toxic Substances and Disease Registry

AVS

acid volatile sulfide

AWQC

ambient water quality criteria

BAF

bioaccumulation factor

BAF

PLANT

Plant-soil bioaccumulation factor

BaP

benzo(a)pyrene

BCF

bioconcentration factor

BERA

Baseline Ecological Risk Assessment

BLM

Bureau of Land Management

BMF

biomagnification factor

BRAC

Base Realignment and Closure

BTAG

Biological Technical Assistance Group

BTEX

benzene, toluene, ethylbenzene, and total xylenes

body weight of receptor

chemical concentration

CAA

Clean Air Act

CAMU

Corrective Action Management Unit

CDDs

chlorinated dibenzo-p-dioxins

CDFs

chlorinated dibenzofurans

CERCLA

Comprehensive Environmental Response, Compensation, and

Liability Act

CFR

Code of Federal Regulations

COEC

Contaminants of ecological concern

COPEC

contaminant of potential ecological concern

CTE

central tendency exposure

CWA

Clean Water Act

dietary composition

DDT

p,p’-dichlorodiphenyltrichloroethane

DERA

Defense Environmental Restoration Account

DERP

Defense Environmental Restoration Program

DQO

data quality objective

DSMOA/CA

Department of Defense and State Memorandum of

Agreement/Cooperative Agreement Program

ECAS

Environmental Compliance Assessment System

EM 200-1-4

31 Dec 10

Glossary-2

Eco-SSL

ecological soil screening level

ECSM

ecological conceptual site model

oxidation-reduction potential

Engineer Manual

EMF

exposure modifying factors

Engineer Pamphlet

equivalents (normally associated with CDDs or PCBs)

ERA

ecological risk assessment

ERAGS

Ecological Risk Assessment Guidance for Superfund (USEPA

1997)

ER-L

effects range-low

ESA

Endangered Species Act

ETAG

Ecological Technical Assistant Group

FDA

U.S. Food and Drug Administration

FFA

Federal Facility Agreement

food ingestion/food intake

Feasibility Study

FUDS

formerly used defense sites

GLWQI

Great Lakes Water Quality Initiative (USEPA 1995b)

HHRA

human health risk assessment

hazard index

hazard quotient

HTRW

Hazardous, Toxic, and Radioactive Waste

IAG

Interagency Agreement

IBI

Index of Biological Integrity

ingestion factor

IRIS

Integrated Risk Information System

IRP

Installation Restoration Program

Organic Carbon Partition Coefficient

median lethal concentration

median lethal dose

LEL

lower effects level

LOAEL

lowest observed adverse effect level

Log K

logarithm of the organic carbon partition coefficient

Log K

logarithm of the n-octanol-water partition coefficient

Log P

see log K

NAPL

non-aqueous phase liquids

NAS

National Academy of Science

NCP

National Oil and Hazardous Substances Pollution Contingency Plan

not detected

NFA

no further action

NOAA

National Oceanic and Atmospheric Administration

NOAEL

no observed adverse effect level

NPL

National Priorities List

NRC

National Research Council

Glossary-3

EM 200-1-4

31 Dec 10

NRDA

Natural Resource Damage Assessment

NRI

Natural Resource Injury

OCPs

organochlorine pesticides

Ordnance and Explosives

ORNL

Oak Ridge National Laboratory

OSWER

Office of Solid Waste and Emergency Response (U.S. EPA)

operable unit

Preliminary Assessment

PAHs

polycyclic aromatic hydrocarbons

PCBs

polychlorinated biphenyls

PDF

portable document format

project manager

PNA

polynuclear aromatics

PRA

probabilistic risk assessment

PRGs

Preliminary Remediation Goals

QSAR

quantitative structure activity relationship

remedial action

RAGS

Risk Assessment Guidance for Superfund, Volume I, Human

Health Evaluation Manual (USEPA, 1989)

RAO

Remedial Action Objective

RCRA

Resource Conservation and Recovery Act

Remedial Design

Remedial Investigation

RMDM

risk management decision-making

RME

reasonable maximum exposure

ROD

Record of Decision

RTECS

Registry of Toxic Effects of Chemical Substances

SEEM

spatially explicit exposure model

SEM

simultaneously extracted metal

Site Inspection

SLERA

Screening-Level Ecological Risk Assessment

SOW

Statement of Work or Scope of Work

SSL

soil screening level

SVOC

semivolatile organic compound

SWMU

Solid Waste Management Unit

TBC

to-be-considered

TCDD

2,3,7,8-tetrachlorodibenzo-p-dioxin

TCL

Target Cleanup Levels

TEF

toxicity equivalence factor (normally associated with CDDs or

PCBs)

TEL

threshold effects level

TPH

total petroleum hydrocarbons

TPP

Technical Project Planning (USACE 1998)

TRV

toxicity reference value

TSERAWG

Tri-Services Environmental Risk Assessment Working Group

EM 200-1-4

31 Dec 10

Glossary-4

TTD

terrestrial toxicity database

temporary unit

TWEM

terrestrial wildlife exposure model

UCL

upper confidence limit

uncertainty factor

USACE

U.S. Army Corps of Engineers

USAECBC

U.S. Army Edgewood Chemical Biological Center (formerly

USAERDEC)

USAPHC (Prov)

U.S. Army Public Health Command (Provisional)

(Formerly U.S. Army Center for Health Promotion and Preventive

Medicine [USACHPPM] )

USDA

U.S. Department of Agriculture

USDHHS

U.S. Department of Human Health Services

USDOC

U.S. Department of Commerce

USDOD

U.S. Department of Defense

USDOE

U.S. Department of Energy

USDOI

U.S. Department of the Interior

USEPA

U.S. Environmental Protection Agency

USFS

U.S. Forest Service

USFWS

U.S. Fish and Wildlife Service

VOC

volatile organic compound

WER

water effect ratio

WERF

Water Environment Research Foundation

WET

wetland evaluation technique

water ingestion/water intake

WTA

wildlife toxicity assessment

THIS PAGE INTENTIONALLY LEFT BLANK