C.I
c/EPA
United States
Environmental Protection
Agency
Office of Research and
Development
Washington, DC 20460
EPA/600/6-88/005B
August 1992
Workshop Review Draft
Estimating
Exposure to
Dioxin-Like
Compounds
Review
Draft
(Do Not
Cite or
Quote)
Notice
This document is a preliminary draft. It has not been formally
released by EPA and should not at this stage be construed to
represent Agency policy. It is being circulated for comment on its
technical accuracy and policy implications.
-------
EPA/600/6-88/005B
DO NOT QUOTE OR CITE August 1992
Workshop Review Draft
ESTIMATING EXPOSURE TO DIOXIN-LIKE COMPOUNDS
NOTICE
THIS DOCUMENT IS A PRELIMINARY DRAFT. It has not been formally released by the U.S.
Environmental Protection Agency and should not at this stage be construed to represent
Agency policy. It is being circulated for comment on its technical accuracy and policy
implications.
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th Floor
Chicago, IL 60604-3590
Exposure Assessment Group
Office of Health and Environmental Assessment
U.S. Environmental Protection Agency
Washington, D.C.
Printed on Recycled Paper
-"V
-------
DRAFT-DO NOT QUOTE OR CITE
DISCLAIMER
This document is an external draft for review purposes only and does not constitute U.S.
Environmental Protection Agency policy. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
: .: U
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
CONTENTS
Tables viii
Figures xii
Foreword xiii
Preface xiv
Authors and Reviewers xv
1. INTRODUCTION 1-1
2. PHYSICAL AND CHEMICAL PROPERTIES 2-1
2.1. INTRODUCTION 2-1
2.2. GENERAL INFORMATION 2-2
2.3. PHYSICAL/CHEMICAL PROPERTIES - CHLORINATED COMPOUNDS 2-4
2.3.1. Water Solubility 2-5
2.3.2. Vapor Pressure 2-6
2.3.3. Henry's Law Constant 2-7
2.3.4. Octanol/Water Partition Coefficient 2-8
2.3.5. Photo Quantum Yields 2-9
2.4. PHYSICAL/CHEMICAL PROPERTIES - BROMINATED COMPOUNDS 2-10
2.5. FATE - CHLORINATED COMPOUNDS 2-10
2.5.1. Polychlorinated Dibenzo-p-dioxins (PCDDs) 2-10
2.5.2. Polychlorinated Dibenzofurans (PCDFs) 2-16
2.5.3. Coplanar PCBs 2-18
2.6. FATE - BROMINATED COMPOUNDS 2-20
3. ENVIRONMENTAL LEVELS OF PCDD, PCDF, AND PCB CONGENERS ... 3-1
3.1. INTRODUCTION 3-1
3.2. CONCENTRATIONS IN SOIL 3-2
3.3. CONCENTRATIONS IN WATER 3-5
3.4. CONCENTRATIONS IN SEDIMENT 3-7
3.5. CONCENTRATIONS IN FISH AND SHELLFISH 3-12
3.6. CONCENTRATIONS IN FOOD PRODUCTS 3-18
3.7. CONCENTRATIONS IN AIR 3-19
3.8. HISTORICAL TRENDS 3-20
3.9. ASSESSMENT OF BACKGROUND EXPOSURES 3-21
3.9.1. Procedure for Estimating Background Exposures 3-21
3.9.2. Geographic Comparisons 3-21
4. ESTIMATING EXPOSURES AND RISKS 4-1
4.1. INTRODUCTION 4-1
4.2. EXPOSURE EQUATION 4-1
4.3. RISK EQUATION 4-2
4.4. PROCEDURE FOR ESTIMATING EXPOSURE 4-5
iii 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
CONTENTS (continued)
5. ESTIMATING EXPOSURE MEDIA CONCENTRATIONS 5-1
5.1. INTRODUCTION 5-1
5.2. CONSIDERATIONS FOR SOLUTION ALGORITHMS 5-7
5.3. ALGORITHMS FOR THE "ON-SITE SOIL"
SOURCE CATEGORY 5-8
5.3.1. Surface Water and Sediment Contamination 5-9
5.3.2. Vapor-Phase Air Concentrations 5-20
5.3.3. Particulate-Phase Air Concentrations 5-24
5.3.4. Biota Concentrations 5-27
5.3.4.1. Fish Concentrations 5-27
5.3.4.2. Vegetation Concentrations 5-38
5.3.4.3. Beef and Milk Concentrations 5-48
5.4. ALGORITHMS FOR THE "OFF-SITE" SOURCE CATEGORY 5-53
5.4.1. Exposure Site Soil Concentrations 5-56
5.4.2. Vapor-Phase Transport 5-62
5.5. ALGORITHMS FOR THE INCINERATOR STACK EMISSION
SOURCE CATEGORY 5-65
5.5.1. Steady-State Soil Concentrations 5-67
5.6. ALGORITHMS FOR THE ASH DISPOSAL IN LANDFILLS
SOURCE CATEGORY 5-68
5.6.1. Landfill Size 5-69
5.6.2. Fugitive Particulate Emissions at the Landfill 5-70
5.6.2.1. Vehicular Traffic Over Landfill
Roadways 5-70
5.6.2.2. Fugitive Emissions From Trucks 5-75
5.6.2.3. Emissions From Unloading 5-77
5.6.2.4. Emissions From Spreading and Compacting 5-79
5.6.2.5. Total Particulate Emissions From
Ash Landfills 5-80
6. MUNICIPAL SOLID WASTE INCINERATION 6-1
6.1. INTRODUCTION 6-1
6.2. OVERVIEW OF PRINCIPAL MUNICIPAL INCINERATION
TECHNOLOGIES IN THE UNITED STATES 6-2
6.3. THE THEORETICAL MECHANISMS GIVING RISE TO THE AIR
EMISSIONS OF DIOXIN-LIKE COMPOUNDS DURING MSW
INCINERATION 6-4
6.4. HYPOTHETICAL MSW INCINERATOR FOR PURPOSES OF
EXPOSURE ANALYSIS 6-10
6.5. DERIVATION OF MASS EMISSION FACTORS FOR THE
HYPOTHETICAL MASS BURN, HEAT RECOVERY MSW
INCINERATOR 6-12
iv 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
CONTENTS (continued)
6.5.1. Congener-Specific Stack Gas Emission Factors
for Dioxin-Like Compounds 6-12
6.5.2. Estimation of Emission of Dioxin-Like Compounds
in MSW Incineration Ash 6-25
6.6. AIR DISPERSION MODELING OF THE STACK GAS
EMISSIONS OF DIOXIN-LIKE COMPOUNDS FROM
THE HYPOTHETICAL MSW INCINERATOR 6-28
6.6.1. Estimation of Distribution in the Stack Emissions
to the Hypothetical MSW Incinerator 6-34
6.7. RESULTS OF THE AIR DISPERSION MODELING OF
CONGENER-SPECIFIC EMISSIONS FROM THE HYPOTHETICAL
MSW INCINERATOR IN TAMPA, FLORIDA 6-41
7. EXPOSURE SCENARIO DEVELOPMENT 7-1
7.1. INTRODUCTION 7-1
7.2. STRATEGY FOR DEVISING EXPOSURE SCENARIOS 7-1
7.3. EXPOSURE PATHWAYS 7-3
7.3.1. Soil Ingestion 7-5
7.3.2. Soil Dermal Contact 7-6
7.3.3. Vapor and Dust Inhalation 7-7
7.3.4. Water Ingestion 7-8
7.3.5. Beef and Dairy Product Ingestion 7-8
7.3.6. Fish Ingestion 7-10
7.3.7. Fruits and Vegetables 7-11
8. PHARMACOKINETICS 8-1
8.1. INTRODUCTION 8-1
8.2. DAILY BACKGROUND LEVELS 8-2
8.2.1. Basis for Calculation 8-2
8.2.2. Daily Intakes 8-6
8.3. COMPARTMENTAL MODELING 8-11
8.3.1. Pharmacokinetic Model 8-12
8.3.2. Model Utilization 8-14
8.3.3. Determining Liver Concentrations From Fat Levels 8-18
8.4. BIOAVAILABILITY AND TISSUE DISTRIBUTION 8-20
8.4.1. Bioavailability 8-21
8.4.1.1 Bioavailability Data 8-21
8.4.1.2. Summary of Bioavailability 8-33
8.4.2. Distribution 8-35
8.4.3. Determination of Daily Intake Dose From
Exposure Concentrations 8-36
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
CONTENTS (continued)
9. DEMONSTRATION OF METHODOLOGY 9-1
9.1. INTRODUCTION 9-1
9.2. STRATEGY FOR DEVISING EXPOSURE SCENARIOS
FOR DEMONSTRATION PURPOSES 9-2
9.3. EXAMPLE EXPOSURE SCENARIOS 9-8
9.4. EXAMPLE COMPOUNDS 9-11
9.5. SOURCE TERMS 9-12
9.6. RESULTS 9-16
9.6.1. Observations Concerning Exposure Media
Concentrations 9-16
9.6.2. Observations Concerning LADD Exposure Estimates .... 9-27
10. UNCERTAINTY 10-1
10.1. INTRODUCTION 10-1
10.2. UNCERTAINTIES IN SPECIFIC METHODS APPLIED 10-3
10.2.1. Environmental Chemistry of Dioxin-Like Compounds .... 10-3
10.2.2. Lifetime, Body Weights, and Exposure Durations 10-6
10.2.3. Soil Erosion Algorithm 10-7
10.2.3.1. Comparison with Literature Data 10-8
10.2.3.2. Key Parameters and Assumptions of
the Erosion Algorithm 10-13
10.2.4. Surface Water Suspended and Bottom Sediments 10-18
10.2.4.1. Comparison with Literature Data 10-18
10.2.4.2. An Alternate Approach 10-21
10.2.4.3. Sensitivity to Key Parameters and Assumptions 10-22
10.2.5. Soil Ingestion Exposure 10-26
10.2.6. Soil Dermal Contact Pathway 10-28
10.2.7. Water Ingestion 10-31
10.2.7.1. Exposure Parameters 10-31
10.2.7.2. Comparison of Estimated Water Concentrations
With Observed Data 10-34
10.2.7.3. An Alternate Modeling Approach for
Estimating Water Concentrations 10-35
10.2.7.4. Sensitivity to Model Parameters 10-38
10.2.8. Fish Ingestion Exposure 10-39
10.2.8.1. Fish Ingestion Rates 10-39
10.2.8.2. Comparison of Estimated Fish Concentrations
With Literature Values 10-41
10.2.8.3. Alternate Modeling Approaches 10-45
10.2.8.4. Sensitivity to Key Model Parameters 10-49
10.2.9. Vapor Phase Inhalation Exposures 10-51
10.2.9.1. Comparison of Model Results With Measured
Concentrations 10-52
vi 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
CONTENTS (continued)
10.2.9.2. Other Modeling Approaches and Considerations
for Vapor-Phase Emissions 10-55
10.2.9.4. Uncertainty of Key Model Parameters 10-59
10.2.10. Particulate Phase Inhalation 10-62
10.2.10.1. Wind Erosion Modeling 10-63
10.2.10.2. Fugitive Ash Emissions 10-65
10.2.11. Uncertainty in Fruit and Vegetable Ingestion 10-66
10.2.11.1. Fruit and Vegetable Ingestion Rates 10-67
10.2.11.2. Comparing Predicted Versus Measured Plant
Concentrations 10-70
10.2.11.3. Uncertainty Evaluation of the Vegetation
Concentration Algorithm 10-75
10.2.12. Beef and Milk Ingestion 10-82
10.2.12.1. Beef and Milk Ingestion Rates 10-83
10.2.12.2. Comparison of Modeled Beef and Milk
Concentrations With Concentrations Found . 10-85
10.2.12.3. Alternate Modeling Approaches for
Estimating Beef and Milk Concentrations . . . 10-88
10.2.12.4. Uncertainty of Model Parameters Estimating
Beef and Milk Concentrations 10-95
10.3. UNCERTAINTY AND VARIABILITY IN ANALYSIS OF
DIOXIN-LIKE COMPOUNDS EMITTED FROM MUNICIPAL
SOLID WASTE INCINERATION 10-98
10.3.1. The Probability of Emission of Dioxin-Like Compounds . . . 10-99
10.3.2. The Quantity of Release of Dioxin-Like Compounds from
the Stack of the Incinerator 10-101
10.3.3. The Representativeness of the Hypothetical Incinerator
in Comparison to Actual Facilities 10-105
10.3.4. Air Dispersion Analysis of Incinerator Emissions 10-109
10.3.5. Comparison of Predicted Ambient Concentrations to
Ambient Measurements 10-112
11. CONCLUSIONS AND RECOMMENDATIONS 11-1
11.1. CONCLUSIONS 11-1
1 1.2. RECOMMENDATIONS 11-7
APPENDIX A: ENVIRONMENTAL CHEMISTRY A-1
APPENDIX B: ENVIRONMENTAL CONCENTRATIONS B-1
APPENDIX C: SPREADSHEET ANALYSIS AND MODEL PARAMETERS C-1
APPENDIX D: MODELING THE IMPACT OF EFFLUENT DISCHARGES INTO
SURFACE WATERS D-1
vii 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
TABLES
1-1 Toxicity equivalency factors (TEF) 1-4
1-2 Dioxin-Like PCBs 1-5
1-3 Nomenclature for dioxin-like compounds 1-6
2-1 Possible number of positional CDD (or BDD) and CDF (or BDF)
congeners 2-3
3-1 Regional comparisons of toxic equivalent concentrations 3-23
3-2 Background exposures 3-24
5-1 Summary and fate and transport solution algorithms for
estimating exposure media concentrations for the four
source categories 5-3
5-2 Available Biota to Sediment Accumulation Factors, BSAF,
for dioxin-like compounds 5-33
5-3 Summary of Biota Sediment Accumulation Factors, BSAFs, for
PCBs 5-36
6-1 Concentration of PCDD/PCDF (ng/g) on municipal
incinerator fly ash at varying temperatures 6-7
6-2 Capacity distribution (TOP) of existing and projected mass burn,
heat recovery MSW incinerators in the United States 6-11
6-3 Modeling parameters for the hypothetical MSW incinerator 6-12
6-4 Estimation of PCDD congener-specific emission factors
(grams/metric ton of MSW burned) for representative mass burn
heat recovery incinerators 6-1 6
6-5 Estimation of PCDF congener-specific emission factors
(grams/metric ton of MSW burned) for representative mass
burn heat recovery incinerators 6-18
6-6 Coplanar polychlorinated biphenyl congeners having
dioxin-like biological activity; number of possible isomers;
ratio of toxic isomer to total isomers in homologue group 6-22
6-7 Emission factors (grams/metric ton) of homologue groups of
polychlorinated biphenyls from emission tests of MSW incinerators . . 6-23
viii 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
TABLES (continued)
6-8 Estimation of congener-specific emission factors of coplanar PCBs . . 6-24
6-9 Emission factors (grams/metric ton) of PCDDs/PCDFs
for collected fly ash in the ESP control scenario 6-29
6-10 Emission factors (grams/metric ton of MSW incinerated)
of coplanar PCBs for collected fly ash in the ESP control scenario . . . 6-29
6-11 Emission factors (grams/metric ton of MSW incinerated)
of PCDDs/PCDFs for collected fly ash in the dry-scrubber, fabric
filter control scenario 6-30
6-12 Emission factors (grams/metric ton of MSW incinerated)
of coplanar PCBs for collected fly ash in the dry-scrubber, fabric
filter control scenario 6-30
6-13 Comparison of particle-size distribution in particulate emissions
resulting from ESP controls with particulate emissions from fabric
filter controls 6-35
6-14 Congener-specific emission factors (grams/sec) as input to the ISC air
dispersion model for both ESP and DSFF control scenarios 6-39
6-15 Congener-specific emission factors by particle size (grams/sec) for both
the ESP and dry scrubber-fabric filter (DSFF) control scenarios to the
hypothetical MSW incinerator 6-40
6-16 Predicted annual average ambient air concentrations (grams/m3)
of specific dioxin-like congeners at specified distances from the
hypothetical MSW incinerator-ESP controls 6-44
6-17 Predicted annual average air concentrations (grams/m3)
of specific dioxin-like congeners at specified distances from
the hypothetical MSW incinerator-DSFF controls 6-45
6-18 Wet + dry surface deposition (grams/m2-yr) of specific congeners
at specified distances in the vicinity of the hypothetical MSW
incinerator--ESP controls 6-47
6-19 Wet + dry surface deposition (grams/m2-yr) of specific congeners
at specified distances in the vicinity of the hypothetical MSW
incinerator-DSFF controls 6-48
7-1 Summary of exposure pathway parameters 7-13
ix 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
TABLES (continued)
8-1 Calculated Daily Intakes for 2,3,7,8-TCDD 8-7
8-2 Half life calculations 8-11
8-3 Model determined daily intakes 8-18
9-1 Summary of key source terms for the six exposure scenarios
and the three example compounds 9-13
9-2 Exposure media concentrations estimated for all scenarios
and pathways 9-1 7
9-3 Lifetime average daily dose (LADD) estimates for all scenarios
and exposure pathways (all results in mg/kg-day) 9-20
9-4 Percent contribution of the different exposure pathways
within each exposure scenario 9-29
9-5 Exposures to low soil concentrations of 2,3,7,8-TCDD assuming
lifetime exposure durations and unlimited contact with impacted
media, compared with exposures assuming limited durations and
limited contact 9-31
9-6 Comparison of exposure pathway contributions to total daily
exposure as estimated in example Scenario #2 and in Travis and
Hattemer-Frey (1 991) 9-34
10-1 Uncertainties associated with the lifetime, body weight,
and exposure duration parameters 10-8
10-2 Summary of uncertainties associated with the soil
delivery algorithm 10-9
10-3 Summary of off-site soil contamination from Tier 1 and 2 sites
of the National Dioxin Study 10-12
10-4 Summary of uncertainties associated with the surface water
sediment algorithms 10-19
10-5 Uncertainties associated with the soil ingestion pathway 10-29
10-6 Uncertainties associated with the dermal exposure pathway 10-32
10-7 Uncertainties associated with the water ingestion pathway 10-33
x 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
TABLES (continued)
10-8 Uncertainties associated with the fish ingestion pathway 10-40
10-9 Uncertainties and sensitivities associated with estimating
vapor-phase air concentrations from contaminated soils 10-53
10-10 Uncertainties and sensitivities associated with estimating
particulate-phase air concentrations from contaminated soils 10-64
10-11 Uncertainties associated with vegetable and fruit ingestion
exposure algorithms 10-68
10-12 Summary of plant concentration vs. soil concentration data
for 2,3,7,8-TCDD 10-72
10-13 Contribution of above ground vegetation concentrations from
air-to-leaf transfers and particulate depositions 10-78
10-14 Uncertainties associated with beef and milk ingestion
exposure algorithms 10-84
10-15 Estimated relative importance of soil, pasture grass, and
fodder ingestion to the estimation of beef and milk concentration
when soil and incinerator emissions are the source of 2,3,7,8-TCDD
contamination 10-92
10-1 6 Basic statistical analysis of emissions of dioxin used in the
hypothetical MSW incinerator scenarios 10-104
10-17 Comparison between measured ambient air concentrations
with predicted ambient air impacts of the hypothetical MSW
incinerator (values as pg/m3) 10-114
xi 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
FIGURES
4-1 Roadmap for assessing exposure and risk to dioxin
and dioxin-like compounds 4-6
5-1 Watershed delivery ratio, SDW, as a function of
watershed size 5-17
6-1 Principal physical processes used by the ISC dispersion
model to estimate the atmospheric transport and deposition of
pollutants discharged from the smokestack 6-31
6-2 Dispersion gradient of the ground-level concentration with
distance from the incinerator--ESP control scenario 6-43
6-3 Dispersion gradient of the ground-level concentration with
distance from the incinerator-DSFF control scenario 6-43
8-1 Sample Calculation of Daily Intake for 2,3,7,8-TCDD 8-7
8-2 Model estimates of elimination of 2,3,7,8-TCDD from fat 8-15
8-3 Accumulation of TCDD in fat with 0.44 pg/kg/day dose - human .... 8-16
8-4 Accumulation of TCDD in fat with 0.30 pg/kg/day dose - human .... 8-17
10-1 Comparison of relative distributions of MSW stack emissions
of PCS congeners as derived by Tiernan 10-102
10-2 Percent distribution of existing incinerators 10-107
10-3 Percent distribution of incinerators by 1995 10-107
10-4 Capacity distribution of existing and planned mass burn
incinerators 10-108
10-5 Scatter plot of stack height relative to capacity 10-108
xii 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
FOREWORD
The Exposure Assessment Group (EAG) within the Office of Health and Environmental
Assessment of EPA's Office of Research and Development has three main functions: (1) to
conduct exposure assessments, (2) to review assessments and related documents, and (3) to
develop guidelines for exposure assessments. The activities under each of these functions
are supported by and respond to the needs of the various program offices. In relation to the
third function, EAG sponsors projects aimed at developing or refining techniques used in
exposure assessments.
The purpose of this document is to present procedures for conducting site-specific
exposure assessments to dioxin-like compounds. It serves as a final version of the 1988
draft document titled "Estimating Exposure to 2,3,7,8-TCDD." This effort represents a
substantial expansion in scope to include all compounds that exhibit dioxin-like toxicity. The
types of sites covered in this document include incinerators, landfills, and other areas
involving contaminated soils. The procedures identify possible exposure pathways associated
with these sites, present fate models to estimate media concentrations at the point of
exposure, and suggest ways to estimate contact rates and resulting exposure levels.
Hypothetical examples are used to illustrate the procedures.
The document is intended to be used as a companion to the health reassessment of
dioxin-like compounds that the Agency is publishing concurrently. It is hoped that these two
documents will improve the accuracy and validity of risk assessments involving this important
family of compounds.
Michael A. Callahan
Director
Exposure Assessment Group
xiii 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
PREFACE
The Exposure Assessment Group of the Office of Health and Environmental
Assessment has prepared this guidance document for general use throughout the agency.
The purpose of this document is to present procedures for conducting site-specific exposure
assessments to dioxin-like compounds. It serves as a final version of the 1988 draft
document titled "Estimating Exposure to 2,3,7,8-TCDD" (U.S. EPA, 1988).
xiv 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
AUTHORS AND REVIEWERS
The Exposure Assessment Group (EAG) within EPA's Office of Health and
Environmental Assessment was responsible for the preparation of this document. General
support was provided by VERSAR Inc. under EPA Contract Number 68-DO 0101. Matthew
Lorber of EAG served as EPA task manager (as well as contributing author) providing overall
direction and coordination of the production effort as well as technical assistance and
guidance.
AUTHORS
Primary authors of each chapter are listed below.
Jerry Blancato
U.S. Environmental Protection Agency
Las Vegas, NV
David Cleverly
U.S. Environmental Protection Agency
Washington, DC
Robert J. Fares
Versar, Inc.
Springfield, VA
Geoffrey Huse
Versar, Inc.
Springfield, VA
Matthew Lorber
U.S. Environmental Protection Agency
Washington, DC
John L. Schaum
U.S. Environmental Protection Agency
Washington, DC
Greg Schweer
Versar, Inc.
Springfield
Paul White
U.S. Environmental Protection Agency
Washington, DC
Chapter 8
Chapter 6, 10
Chapter 3
Chapters 2, 3
Chapter 4, 5, 7, 9, 10 and
11 and Appendices C, D
Chapter 1,11
Chapter 2
Chapter 10
xv
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
REVIEWERS
The original draft of this document was reviewed by the Science Advisory Board in
1988. The new draft was reviewed by the following experts outside of EPA:
James Falco, Ph.D.
Battelle, NW
Richland, WA
Thomas E. McKone, Ph.D.
Lawrence Livermore National Laboratory
Livermore, CA
Thomas 0. Tiernan, Ph.D.
Wright State University
Dayton, OH
Curtis C. Travis, Ph.D.
Oak Ridge National Laboratory
Oak Ridge, TN
G.R. Barrie Webster, Ph.D.
University of Manitoba
Winnipeg, Canada
George Fries, Ph.D
United States Department of Agriculture
Beltsville Agricultural Research Center
Beltsville, MD
Thomas Parkerton, Ph.D
Manhattan College
Riverdale, NY
Derek Muir, Ph.D
Freshwater Institute
Department of Fisheries and Oceans
Winnipeg, MB, Canada
Christopher Rappe, Ph.D.
University of Umea
Institute of Environmental Chemistry
Umea, Sweden
Thomas Umbreit, Ph.D.
ATSDR
Atlanta, GA
Vlado Ozvacic, Ph.D.
Ministry of the Environment
Toronto, ON, Canada
Dale Hattis, Ph.D.
Clark University
Worcester, MA
Jeffrey Wong, Ph.D.
California EPA
Sacramento, CA
John Hawley, Ph.D.
NYS Health Department
Albany, NY
XVI
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
1. INTRODUCTION
The primary purpose of this document is to present procedures for conducting site-
specific exposure assessments to dioxin-like compounds. In addition, information is also
provided on the levels of these compounds found in various media, identification of the
possible associated sources, and estimation of the resulting exposure levels.
This document serves as a final version of the 1988 draft document entitled
"Estimating Exposure to 2,3,7,8-TCDD" (EPA, 1988). However, due to the expansion in
scope to cover all dioxin-like compounds and numerous updates, it appears significantly
different than the original draft. The draft document was reviewed by the Science
Advisory Board in 1988 and their comments have been addressed in this assessment.
Additional peer reviews of earlier drafts of this report were conducted in 1992.
Dioxin-like compounds are defined to include those compounds with nonzero
Toxicity Equivalency Factor (TEF) values as defined in the 1989 International scheme. As
shown in Table 1-1, this TEF scheme assigns nonzero values to all chlorinated
dibenzodioxins (CDDs) and chlorinated dibenzofurans (CDFs) with chlorines substituted in
the 2,3,7,8 positions. Additionally, the analogous brominated compounds (BDDs and
BDFs) and certain polychlorinated biphenyls (PCBs, see Table 1-2) have recently been
identified as having dioxin-like toxicity (EPA, 1992a) and thus are also included in the
definition of dioxin-like compounds. The nomenclature adapted here for purposes of
describing these compounds is summarized in Table 1-3.
The types of sites covered in this document include incinerators, landfills and other
areas involving contaminated soils. The procedures identify possible exposure pathways
associated with these sites, present fate models to estimate media concentrations at the
point of exposure and identify ways to estimate contact rates and resulting exposure
levels. Hypothetical examples are used to illustrate the procedures. A brief discussion is
presented on procedures specific to effluent discharges to water bodies. For further
details on assessing these sources, the reader is directed to EPA's recent document titled
"Analysis of Populations at Risk from the Consumption of Dioxin-Contaminated Fish
Caught Near Bleached Pulp and Paper Mills" (EPA, 1992b).
The end products of the exposure assessment procedures presented in this
document are estimates of potential dose expressed in mg of toxicity equivalents/kg-day.
1-1 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Pharmacokinetic models are presented to predict dioxin levels in blood, adipose tissue or
other body compartments. The procedures for converting these dose estimates to risk
estimates are provided in a companion document (EPA, 1992a) on health assessment
which EPA is publishing concurrently and which addresses the same compounds.
The scope of each chapter is summarized below.
Chapter 2, Physical and Chemical Properties, summarizes the information on
physical and chemical properties of the dioxin-like compounds.
Chapter 3, Occurrence, summarizes the levels of dioxin-like compounds found in
various media.
Chapter 4, Estimating Exposure and Risks, presents overall framework for
conducting exposure assessments. Procedures for using the Toxicity Equivalency Factors
in exposure assessments are discussed here.
Chapter 5, Estimating Exposure Media Concentrations, provides procedures for
estimating concentrations of the dioxin-like compounds in exposure media (soil, air, water,
biota) resulting from soil contamination and nearby incinerators.
Chapter 6, Municipal Solid Waste Incineration, provides procedures to estimate the
emission rates of dioxin-like compounds from municipal waste incinerators, including stack
and fly ash emissions. It also demonstrates the use of the Industrial Source Complex
model on a hypothetical incinerator and lists the associated atmospheric dispersion and
deposition estimates from that model exercise.
Chapter 7, Exposure Scenario Development, provides procedures for identifying
exposure pathways, estimating contact rates and resulting exposure levels. Approaches
for defining exposure scenarios are presented.
Chapter 8, Pharmacokinetics, summarizes information about uptake and distribution
of dioxin-like compounds in the body and presents pharmacokinetic models to predict
blood levels resulting from exposure. Additionally, information on blood levels is used to
back calculate associated exposure levels.
Chapter 9, Demonstration of Methodology, develops hypothetical scenarios and
generates exposure estimates to demonstrate the methodologies of this document.
Chapter 10, Uncertainty, discusses the sources and possible magnitude of
uncertainty in the exposure assessment procedures. Uncertainty and variability of fate and
transport, and exposure parameters, are discussed. Modeled exposure media
1-2 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
concentrations are compared with concentrations that have been found in the literature,
and alternate modeling approaches are demonstrated and compared with modeling
approaches discussed in Chapter 5 and demonstrated in Chapter 9.
Chapter 11, Conclusions and Recommendations, presents conclusions concerning
exposure to dioxin-like arrived at through the analyses conducted throughout the
document and recommends areas for further research.
Appendix A, Tables of Chemical Properties, summarizes congener specific data on a
variety of chemical properties
Appendix B, Tables of Media Levels, summarizes congener specific data on a levels
of congeners in various media.
Appendix C, Spreadsheet Analysis and Model Parameters, explains use of
spreadsheets designed to accompany this report and lists model parameters.
Appendix D, Modeling the Impact of Effluent Discharges Into Surface Waters,
discusses procedures for modeling the fate and transport of dioxin-like-compounds
discharged via pipes to water bodies.
1-3 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 1-1. Toxicity equivalency factors (TEF)
Compound TEF
Mono-, Di-, and Tri-CDDs 0
2,3,7,8-TCDD 1
Other TCDDs 0
2,3,7,8-PeCDD 0.5
Other PeCDDs 0
2,3,7,8-HxCDD 0.1
Other HcCDDs 0
2,3,7,8-HpCDD 0.01
Other HpCDDs 0
OCDD 0.001
Mono-, Di-, and Tri-CDFs 0
2,3,7,8-TCDF 0.1
Other TCDFs 0
1,2,3,7,8-PeCDF 0.05
2,3,4,7,8-PeCDF 0.5
Other PeCDFs 0
2,3,7,8-HxCDF 0.1
Other HcCDFs 0
2,3,7,8-HpCDF 0.01
Other HpCDFs 0
OCDF 0.001
Source: EPA, 1989.
1-4 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 1-2. Dioxin-Like PCBs
IUPAC No.
Congener
77
81
105
114
118
126
156
157
167
169
189
3,3',4,4'-tetra PCB
3,4,4',5-tetra PCB
2,3,3',4,4'-penta PCB
2,3,4,4',5-penta PCB
2,3',4,4',5-penta PCB
3,3',4,4',5-penta PCB
2,3,3',4,4',5-hexa PCB
2,3,3',4/4',5'-hexa PCB
2,3',4,4',5,5'-hexa PCB
3,3',4,4',5,5'-hexa PCB
2,3,3',4,4',5,5'-hepta PCB
1-5
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 1-3. Nomenclature for dioxin-like compounds
Term/Symbol
Definition
Congener
Homologue
Isomer
Specific
Congener
D
F
M
D
Tr
T
Pe
Hx
Hp
0
CDD
CDF
PCB
2378
Any one particular member of the same chemical family; e.g., there are 75
congeners of chlorinated dibenzo-p-dioxins.
Group of structurally related chemicals that have the same degree of chlorination.
For example, there are eight homologues of CDDs, monochlorinated through
octochlorinated.
Substances that belong to the same homologous class. For example, there are 22
isomers that constitute the homologues of TCDDs.
Denoted by unique chemical notation. For example, 2,4,8,9-
tetrachlorodibenzofuran is referred to as 2,4,8,9-TCDF.
Symbol for homologous class: dibenzo-p-dioxin
Symbol for homologous class: dibenzofuran
Symbol for mono, i.e., one halogen substitution
Symbol for di, i.e., two halogen substitution
Symbol for tri, i.e., three halogen substitution
Symbol for tetra, i.e., four halogen substitution
Symbol for penta, i.e., five halogen substitution
Symbol for hexa, i.e., six halogen substitution
Symbol for hepta, i.e., seven halogen substitution
Symbol for octo, i.e., eight halogen substitution
Chlorinated dibenzo-p-dioxins, halogens substituted in any position
Chlorinated dibenzofurans, halogens substituted in any position
Polychlorinated biphenyls
Halogen substitutions in the 2,3,7,8 positions
Source: EPA, 1989.
1-6
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
REFERENCES FOR CHAPTER 1
U.S. Environmental Protection Agency. (1988) Estimating exposure to 2,3,7,8-TCDD.
U.S. Environmental Protection Agency, Office of Health and Environmental
Assessment, Washington, DC; EPA/600/6-88/OOSA.
U.S. Environmental Protection Agency. (1989) Interim procedures for estimating risks
associated with exposures to mixtures of chlorinated dibenzo-p-dioxins and
-dibenzofurans (CDDs and CDFs) and 1989 update. U.S. Environmental Protection
Agency, Risk Assessment Forum, Washington, DC; EPA/625/3-89/016.
U.S. Environmental Protection Agency. (1992a) Health reassessment of dioxin-like
compounds. U.S. Environmental Protection Agency, Office of Health and
Environmental Assessment, Washington, DC.
U.S. Environmental Protection Agency. (1992b) Analysis of populations at risk from the
consumption of dioxin-contaminated fish caught near bleached pulp and paper mills.
U.S. Environmental Protection Agency, Office of Technology Transfer, Washington,
DC.
1-7 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
2. PHYSICAL AND CHEMICAL PROPERTIES AND FATE
2.1. INTRODUCTION
This chapter summarizes available information regarding tha physical and chemical
properties and fate of the CDDs, CDFs, BDDs, BDFs, and coplanar PCBs, with an emphasis
on the subset of these chemicals defined as dioxin-like in Chapter 1. Physical/chemical
properties addressed in this chapter include melting point, water solubility, vapor pressure,
Henry's Law constant, octanol/water partition coefficient and photochemical quantum
yield. Fate and transport processes addressed in this chapter include photolysis, oxidation,
hydrolysis, biodegradation, volatilization, and sorption. Biologically-mediated transport
properties (i.e., bioconcentration, plant uptake, etc.) are covered in Chapter 5.
Knowledge of basic physical and chemical properties is essential to understanding
and modeling environmental transport and fate as well as pharmacokinetic and toxicologic
behavior. The most essential parameters for the dioxin and dioxin-like compounds appear
to be water solubility (WS), vapor pressure (VP), octanol/water partition coefficient (Kow)
and photochemical quantum yield. The ratio of VP to WS (VP/WS) yields the Henry's Law
constant (Hc) for dilute solutions of organic compounds, an index of partitioning for a
compound between the atmospheric and the aqueous phase (Mackay et al., 1982).
To maximize and optimize the identification of information on the physical/chemical
properties of these compounds, a thorough search of the recent literature was conducted.
A computer literature search was conducted in the on-line Chemical Abstracts (CA)
database maintained by the Scientific Technical Network (STN), printed abstracts were
obtained and screened, selected hard-copy of literature were retrieved and critically
evaluated, and the appropriate value for each physical/ chemical property for each
congener was selected. These property values are summarized in Tables A-1 and A-2 in
Appendix A. Table A-1 lists the property values for the dioxin-like compounds, and Table
A-2 lists the property values for all CDDs, CDFs, and coplanar PCBs. Where technically
feasible, estimation procedures have been used to provide values where measured data are
not available. Where data could not be found and estimates are not appropriate, the field
is left blank and a homologue group average is presented as the property for that
homologue group.
2-1 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Brief summaries of the scientific literature on the physical-chemical properties and
fate of the polychlorinated and polybrominated dibenzodioxins, dibenzofurans, and
biphenyls are provided in the following sections. The data presented have been chosen
based on the adequacy of the experimental methods, the reliability of the estimation
procedures, and the credibility of the obtained results. No detailed discussions relative to
earlier reported data and scientific findings are included here, although such discussions
are usually offered in the original papers.
2.2. GENERAL INFORMATION
Polychlorinated dibenzodioxins (CDDs), polychlorinated dibenzofurans (CDFs), and
polychlorinated biphenyls (RGBs) are chemically classified as halogenated aromatic
hydrocarbons. CDDs and CDFs are formed as by-products through a variety of chemical
reactions (e.g., dimerization of chlorophenols and phenoxy-acids to form chlorodioxins).
Both compound classes have a triple-ring structure that consists of two benzene rings
connected by a third ring. For dibenzodioxins, the connection of the benzene rings is
through a pair of oxygen atoms as opposed to one oxygen atom for the dibenzofurans (see
structures below). PCBs are a class of compounds formed by the chlorination of a
biphenyl molecule.
The chlorinated and brominated dibenzodioxins and dibenzofurans are tricyclic
aromatic compounds with similar physical and chemical properties, and both classes are
quite similar structurally. There are 75 possible different positional congeners of CDDs and
135 different congeners of CDFs. Likewise, there are 75 possible different positional
congeners of BDDs and 135 different congeners of BDFs (see Table 2-1). The basic
structure and numbering of each chemical class is shown below.
2-2 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
PCDDs
PCDFs
X = 1 to 4, Y = 1 to 4, X + Y >_ 1
Table 2-1. Possible number of positional CDD (or BDD) and CDF (or BDF) congeners
Halogen substitution
Mono
Di
Tri
Tetra
Penta
Hexa
Hepta
Octa
Nona
Deca
CDDs (or BDDs)
2
10
14
22
14
10
2
1
0
0
Number of Congeners
CDFs (or BDFs)
4
16
28
38
28
16
4
1
0
0
PCBs
3
12
24
42
46
42
24
12
3
1
2-3
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
There are 209 possible PCB congeners (see Table 2-1). The physical/chemical
properties of each congener vary according to the degree and position of chlorine
substitution. The list of coplanar PCBs can be found in Table 1-2. PCBs assume a
coplanar structure when the two benzene rings rotate into a position where the two rings
are in the same plane. The PCBs assume a dioxin-like structure when the substituent
chlorines occupy the 3, 3', 4, 4', 5, or 5' positions, or possibly, one of the 2 or 2'
positions, and the structure is not hindered from assuming the preferred planar
configuration. The basic structure and numbering of each chemical class is shown below.
2'
X = 1 to 5, Y = 1 to 5. X + Y >. 1
2.3. PHYSICAL/CHEMICAL PROPERTIES - CHLORINATED COMPOUNDS
Limited research has been carried out to determine physical and chemical properties
of PCDFs and PCDDs. The congeners having 2,3,7,8-chlorination have received the most
attention with 2,3,7,8,-TCDD being the most intensely studied compound. Of the large
number of possible isomers, very few are available commercially and preparation and
synthesis can be both time consuming and difficult. In addition, many of these isomers
have not been prepared in pure form. The isomers that have been prepared may not be
available in sufficient quantities for testing. Another factor to be considered is the high
toxicity of these compounds, which necessitates extreme precautions to prevent adverse
effects if their syntheses are to be attempted.
2-4
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
2.3.1. Water Solubility
Very few measured water solubility values are available in the literature. Marple et
al. (1986a) reported the water solubility of 2,3,7,8-TCDD as 19.3 ± 3.7 parts per trillion
(nanograms per liter, ng/L) at 22°C (Table A-1). Marple et al. (1986a) used a procedure of
equilibrating thin films of resublimed 2,3,7,8-TCDD with a small volume of water followed
by gas chromatography (GO analysis with 63Ni electron capture detection. Other water
solubility values for 2,3,7,8-TCDD have been reported in the literature and are summarized
in U.S. EPA (1990). Values ranging from 7.9 ng/L to 483 ng/L are reported in U.S. EPA
(1990) with 19.3 ng/L selected as the recommended value.
Friesen et al. (1985) used a HPLC generator column to measure the water
solubilities of a series of chlorinated dioxins (1,2,3,7- and 1,3,6,8-TCDD; 1,2,3,4,7-PCDD;
1,2,3,4,7,8-HxCDD; 1,2,3,4,6,7,8-HpCDD; and 1,2,3,4,6,7,8,9-OCDD) and reported
water solubilities ranging from 430 ng/L to 0.4 ng/L for the 1,2,3,7-TCDD and
1,2,3,4,6,7,8,9-OCDD isomers, respectively (Table A-1). Friesen et al. (1990) used a gas
chromatography/mass spectrometry detection (GC/MSD) generator column technique to
measure the water solubilities of a series of chlorinated furans (2,3,7,8-TCDF; 2,3,4,7,8-
PeCDF; 1,2,3,6,7,8- and 1,2,3,4,7,8-HxCDF; and 1,2,3,4,6,7,8-HpCDF) and again
reported a decrease in water solubility with an increase in the number of chlorine
substituents. The reported water solubility values ranged from 1.37 x 10~9 mol/L (419
ng/L) for the 2,3,7,8-TCDF isomer to 3.30 x 10~12 mol/L (1.35 ng/L) for the 1,2,3,4,6,7,8-
HpCDF isomer (Table A-1).
The reported water solubility values for the PCB compounds are higher than those
for the CDD and CDF compounds (Table A-1). The reported values range from 36.1 ng/L
for 3,3',4,4',5,5'-HxCB to 2,920 ng/L for 3,4,4',5-TeCB. However, only one of the values
listed in Table A-1 is a measured value, 569 ng/L for 3,3',4,4'-TeCB, and the other values
are only estimates reported in the literature. The measured water solubility value was
reported by Dickhut et al. (1986) based on a modified generator column method.
For those compounds without reported measured water solubility values,
estimations were calculated by the homologue-average method. For example, for the
tetra-chlorinated dioxins, values reported in the literature were averaged to yield an
estimated water solubility value for the tetra-chlorinated dioxin homologue group. A
similar procedure was used to develop the average value for each of the other
2-5 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
polychlorinated dioxin, and furan homologue groups. Estimating the water solubility values
from measured log Kow values by using the estimation procedure of Lyman et al. (1982)
did not yield satisfactory results; the estimated water solubilities for 2,3,7,8-TCDD and
1,3,6,8-TCDD were at least two orders of magnitude greater from the measured values in
Tables A-1 and A-2. Compounds which have water solubility values in the ranges reported
for these chlorinated compounds are considered to be very poorly soluble in water.
2.3.2. Vapor Pressure
Very few measured vapor pressure values are available in the literature for the
CDDs and CDFs. The only measured vapor pressures available are for 2,3,7,8-TCDD,
1,2,3,4-TCDD, and 1,3,6,8-TCDD (Tables A-1 and A-2).
Similar to water solubility, U.S. EPA (1990) presented a variety of vapor pressure
data for 2,3,7,8-TCDD, and selected a recommended value of 7.4 x 10~10 mm Hg at
25°C. This value was reported by Podoll et al. (1986) who used radiolabeled 2,3,7,8-
TCDD and a gas saturation technique with combustion to 14C02. Rordorf (1987, 1989)
reported a higher vapor pressure value for 2,3,7,8-TCDD, 1.49 x 10"9mm Hg. SRC (1991)
reported this same value by extrapolating the vapor pressures measured by Schroy et al.
(1985b) at four higher temperatures, 30°, 55°, 62°, and 71 °C. The value recommended
in U.S. EPA (1990) is reported in Table A-1. Webster et al. (1985) used a gas saturation
method to determine the vapor pressure for 1,3,6,8-TCDD, 4.03 x 10~6 mm Hg, but this
value appears to be out of line with the other reported values, so it is not used in Table
A-1.
Rordorf (1987, 1989) reported an experimental vapor pressure value for 1,2,3,4-
TCDD, 4.8 x 10'8 mm Hg, 1,2,3,4,6,7,8,9-OCDD, 8.25 x 10-13mm Hg, and
1,2,3,4,6,7,8,9-OCDF, 3.75 x lO'12 mm Hg (Table A-1). Rordorf (1987, 1989) used a
gas flow method in a saturation oven, with integrated gas chromatographic analysis, to
measure vapor pressure values for 10 PCDDs and 4 CDFs. Rordorf (1987, 1989) also
used a vapor pressure correlation method to predict the vapor pressures of 1 5 other CDDs
and 55 CDFs based on the measured vapor pressures for the 10 PCDDs, 4 CDFs, and the
deduced boiling point and enthalpy data for the larger series of CDDs and CDFs. Measured
boiling point and enthalpy data are in good agreement with the deduced data used in the
correlation method. Of the CDDs studied by Rordorf (1987, 1989), only two of the ten,
2-6 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
1,2,3,4-TCDD and 2,3,7,8-TCDD, were in the dioxin-like compound group of chemicals
studied in this report. The other CDDs with measured values were monochloro-, dichloro-,
and trichloro-dibenzo-p-dioxins.
The vapor pressure values reported for the PCBs are higher than those reported for
the CDD and CDF compounds (Table A-1), although all of the PCB and CDF values and
most of the CDD values are calculated-estimated values reported in the literature. The
values reported in Tables A-2 and A-3 are an average of the OV-101 Rl and Dexsil 410 Rl
correlation methods because both methods were determined to be equally valid. As with
the other groups, the vapor pressures of the PCBs decrease with an increase in the number
of chlorine substituents. The highest value for PCBs in Table A-2 is 1.77 x 10~5 mm Hg
for 3,4,4',5-TeCB and the lowest value reported is 3.04 x 10~7 mm Hg for a heptachloro-
PCB.
Vapor pressure values for CDDs and CDFs that were not found in the literature
were also calculated by the homologue-average method using the literature-reported values
within a homologue group. For example, the literature values for the TCDDs were
averaged to obtain an estimated vapor pressure assumed to apply to the TCDD congeners
that did not have literature values. A similar procedure was used to develop a homologue-
average for each of the other homologue groups. Compounds with vapor pressures in the
ranges reported for these compounds are considered to have very low vapor pressures.
However, volatilization is considered to be one of the important transport properties for
chlorinated dioxins, furans, and biphenyls.
2.3.3. Henry's Law Constant
Measured data for Henry's Law constant have been reported for only two
compounds: 1,3,6,8-TCDD at 6.81 x 10"5 atm-m3/mol (Webster et al. 1985), and
3,3',4,4'-PCB at 9.4 x 10~5 atm-m3/mol (Dunnivant and Elzerman 1988) (Tables A-1 and
A-2). Both values were determined by the gas purging technique. All other values
reported in the literature for CDDs, CDFs, and PCBs were calculated by the vapor
pressure/water solubility ratio technique.
Group-average Henry's Law constants are estimated for each homologue group
based on the reported data for that group. The Henry's Law constant values for the PCBs
are similar to those for the CDDs and CDFs, except for the value estimated for
2-7 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
2/3,3',4,4',5,5'-HpCB (3.0 x 1CT3 atm -m3/mol) which is higher than most of the other
reported values and homologue-averages.
Lyman et al. (1982) offers guidelines, though not specific to these compounds, for
comparing the degree to which organic compounds volatilize from water. These guidelines
suggest that volatilization of polycyclic aromatic hydrocarbons and halogenated aromatics
(which includes all the dioxin-like compounds) from water represents a significant transfer
mechanism from the aqueous to the atmospheric phase.
2.3.4. Octanol/Water Partition Coefficient
Marple et al. (1986b) reported the octanol/water partition coefficient of
2,3,7,8-TCDD as 4.24 (± 2.73) x 106 at 22 ± 1 °C, yielding a log Kow of 6.64 (Table
A-1). Two similar experimental techniques were used, but the more reliable method
involved equilibration of water-saturated octanol, containing the 2,3,7,8-TCDD, with
octanol-saturated water, over 6 to 31 days. U.S. EPA (1990) reported that the available
low Kow data ranged from 6.15 to approximately 8.5, and the 6.64 value reported by
Marple et al. (1988b) was the value recommended in that report.
Burkhard and Kuehl (1986) used reverse-phase High Pressure Liquid
Chromatography (HPLC) and Liquid Chromatography/Mass Spectrometry (LCMS) detection
to determine octanol/water partition coefficients for 2,3,7,8-TCDD and a series of seven
other tetrachlorinated planar molecules, including three other TCDD isomers (1,2,3,4-
TCDD; 1,3,7,9-TCDD; 1,3,6,8-TCDD), 2,3,7,8-TCDF and 3,3',4,4'-tetrachlorobiphenyl.
The log Kow values for the four TCDD isomers ranged from 7.02 to 7.20, the log Kow for
the 2,3,7,8-TCDF was 5.82, and the log Kow for 3,3',4,4'-TCB was 5.81.
Burkhard and Kuehl (1986) also reevaluated data on 13 CDDs and CDFs previously
reported by Sarna et al. (1984) under similar experimental techniques. In the reevaluation,
Burkhard and Kuehl (1986) used experimental rather than estimated log Kow values in
correlations with gas chromatographic retention times. This approach yielded log
octanol-water partition coefficients ranging from about 4.0 for the non-chlorinated parent
molecules to about 8.78 for the octa-chlorinated compounds, much lower than the values
originally reported by Sarna et al. (1984).
Sijm et al. (1989) used a slow stirring method to obtain log Kow values for 73 CDD
and CDF congeners. Values ranged from 6.10 to 7.92.
2-8 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
The measured and literature-estimated log Kow values for the PCBs are similar to
those reported for the CDDs and CDFs (Table A-1). The values range from 6.2 (measured)
for 3,3',4,4'-TeCB to 7.72 (literature-estimate) for 2/3,3',4,4',5,5'-HpCB. The log Kow
values increase with an increase in the number of chlorine substituents. The log Kow for
the 3,3',4,4'-TeCB was measured by Hawker and Connell (1988) using the generator
column technique against the linear relationship of relative retention time on a nonselective
gas chromatograph stationary phase. Log Kow values for the HxCBs were measured by
Risley et al. (1990) using a high-performance liquid chromatographic (HPLC) system. The
values reported in Tables A-2 and A-3 are an average of the two HPLC techniques because
both methods were determined to be equally valid.
Partition coefficient values that were not found in the literature were calculated by
averaging the literature values within homologue groups, as was done for vapor pressure
and water solubility. Partition coefficients in the ranges of these reported values indicate
that the substances tend to adsorb strongly to organic components in the soil and may
bioconcentrate in those organisms exposed to the compounds.
Literature values for the hexachlorodibenzofurans could not be found.
2.3.5. Photo Quantum Yields
Photo quantum yields have been reported for only eight compounds: 1,2,3,7-
TCDD at 5.42 x 10"4 (Choudhry and Webster, 1989)
1,3,6,8-TCDD at 2.17 x 10"3 (Choudhry and Webster, 1989)
2,3,7,8-TCDD at 2.2 x 10"3 (Dulin et al. 1986)
1,2,3.4,7-PeCDD at 9.8 x 1(T6 (Choudhry and Webster, 1987)
1,2,3,4,7,8-HxCDD at 1.1 x 10"4 (Choudhry and Webster, 1987)
1,2,3,4,6,7,8-HpCDD at 1.53 x 1Cr5 (Choudhry and Webster, 1987)
1,2,3,4,6,7,8,9-OCDD at 2.26 x 1CT5 (Choudhry and Webster, 1987)
1,2,4,7,8-PeCDF at 1.3 x 10'2 (Choudhry et al. 1 990)
1,2,3,4,7,8-HxCDF at 6.96 x 10"4 (Choudhry et al. 1990)
All of the quantum yields were measured in a water-acetonitrile solution at 313 nm. No
values were found for the PCBs.
2-9 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Homologue group averages were not calculated because photo quantum yields are
very sensitive to chlorine position and the solvent system used in the experiments.
Different water to acetonitrile volume ratios were used in these experiments.
The photo quantum yields can be used to estimate the photolysis half-life of these
compounds. This is further discussed in Section 2.4.
2.4. PHYSICAL CHEMICAL PROPERTIES - BROMINATED COMPOUNDS
Information on the physical and chemical properties of the polybrominated dioxins
and furans is very limited. Until very recently, only estimated values were available, but in
early 1992 Webster et al. (personal communication) will publish measured results for
testing with brominated compounds. The data for these brominated compounds are listed
in Table A-1.
2.5 FATE - CHLORINATED COMPOUNDS
2.5.1 Polychlorinated Dibenzo-p-dioxins (PCDDs)
With the exception of 2,3,7,8-TCDD, the environmental behavior of PCDDs has
been studied very little. To what extent these studies with 2,3,7,8-TCDD apply to other
PCDDs and, conversely, to what extent the results of studies on other PCDDs apply to
2,3,7,8-TCDD is uncertain in many instances. In general, PCDDs are extremely stable
compounds under normal environmental conditions, are relatively immobile in the
environment, and are primarily associated with particulate and organic materials. The only
environmentally significant path for destruction of PCDDs is photodegradation.
Upon deposition of PCDDs onto surfaces, there can be a relatively high initial loss
due to photodegradation and/or volatilization. Once PCDDs adsorb onto or move into soils
or sediments, however, they are apparently strongly sorbed. Erosion and aquatic transport
of sediment appears to be the dominant physical transport mechanism for sorbed PCDDs.
Some studies have shown that there may be slow rates of vapor phase transport out of
soils.
If discharged to water, PCDDs are expected to preferentially sorb to solids.
Volatilization may also be a significant transport mechanism for non-sorbed PCDDs even
though PCDDs have negligible vapor pressures. In the atmosphere, PCDDs are expected
2-10 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
to also preferentially sorb to airborne particulate matter, particularly the more chlorinated
congeners. Removal mechanisms include dry and wet deposition and photodegradation.
• Photodegradation
The photochemistry of PCDDs has been reviewed by Miller and Zepp (1987),
Choudry and Webster (1987), and Esposito et al. (1980). This section summarizes the
key findings of those reviews and the results of recent environmentally significant studies.
Photodegradation appears to be the most significant natural degradation mechanism
for PCDDs. PCDDs absorb electromagnetic radiation above 290 nm wavelength (i.e., the
lower bound of the sun's radiation reaching the earth's surface) and, therefore, can be
expected to be subject to photolysis by sunlight (EPRI 1983). It is expected that all
PCDDs can be dechlorinated photolytically in the presence of a suitable hydrogen donor
(Crosby et al. 1973, Crosby 1978). The major products of photolysis are lower
chlorinated PCDDs. Numerous studies have demonstrated that photolysis is slow in water
and on dry surfaces but is dramatically increased when organic solvents are present.
Dulin et al. (1986) studied the photolysis of 2,3,7,8-TCDD in various solutions
under sunlight and artificial light. Using the results obtained in a water:acetonitrile solution
(1:1, v/v) under sunlight conditions, Dulin et al. (1986) calculated the half-life of 2,3,7,8-
TCDD in surface water in summer at 40 degrees north latitude to be 4.6 days. The
quantum yield for photodegradation of 2,3,7,8-TCDD in water was three times greater
under artificial light at 313 nm than under sunlight, and the artificial light photolysis
quantum yield for hexane, a good hydrogen donor, was 20 times greater than for the
watenacetonitrile solution, a poor hydrogen donor.
Podoll et al. (1986) used the Dulin et al. (1986) quantum yield data for the
watenacetonitrile solution to calculate seasonal half-life values for dissolved 2,3,7,8-TCDD
at 40 degrees north latitude in clear near-surface water. The seasonal values for half-lives
were calculated to be 21 hours in summer, 51 hours in fall, 118 hours in winter, and 27 hours
in spring. The authors noted, as do the authors of the studies described below, that sorption
of 2,3,7,8-TCDD to suspended particulate would have the effect of slowing the photolysis
rates.
Choudry and Webster (1989) studied the photolytic behavior under 313 nm light of
a series of PCDDs in a watenacetonitrile solution (2:3, v/v). Assuming that the quantum
2-11 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
yields observed in these studies are the same as would be observed in natural waters,
Choudry and Webster (1989) estimated the mid-summer half-life values at 40 degrees
north latitude in clear near-surface water to be as follows: 1,2,3,7-TCDD (1.8 days);
1,3,6,8-TCDD (0.31 days); 1,2,3,4,7-PeCDD (15 days); 1,2,3,4,7,8-HxCDD (6.3 days);
1,2,3,4,6,7,8-HpCDD (47 days); and 1,2,3,4,6,7,8,9-OCDD (18 days). In addition, the
authors also experimentally determined the sunlight photolysis half-life of 1,3,6,8-TCDD in
pond water to be 3.5 days.
A recent study by Friesen et al. (1990) examined the photolytic behavior of
1,2,3,4,7-PeCDD and 1,2,3,4,6,7,8-HpCDD in watenacetonitrile (2:3, v/v) and in pond
water under sunlight conditions at 50 degrees north latitude. The observed half-lives of
these two compounds in the acetonitrile solution were 1 2 and 37 days, respectively, and
were 0.94 and 2.5 days in pond water, respectively. Based on these results, the authors
conclude that an indirect photolytic mechanism may dominate the photodegradation of
PCDDs in natural waters.
Substantial research on the environmental persistence of 2,3,7,8-TCDD has been
done as part of the decontamination of the area around the ICMESA chemical plant in
Seveso, Italy, which was contaminated when a trichloropheno! reaction vessel overheated
in 1976, blowing out the safety devices and spraying dioxin contaminated material into the
environment. The levels of dioxin in the soil decreased substantially during the first six
months following the accident, reaching a steady state of 1/5 to 1/11 of the initial levels
(diDomenico et al. 1982). An experiment was conducted at this site to determine the
effectiveness of photolysis in decontaminating surface deposits on foliage. Test plots
were sprayed with olive oil to act as a hydrogen donor and the levels of dioxin on grass
were found to be reduced by over 80 percent within nine days (Crosby 1981). The
2,3,7,8-TCDD in contaminated soil was also found to be photolabile in sunlight when the
soil was suspended in an aqueous solution of a surfactant. The destruction of 8 ug/ml of
2,3,7,8-TCDD in 0.02 M hexadecylpyridinium chloride could be accomplished in 4 hours
(Botre et al. 1978).
Photolysis in the gas phase also appears to be a degradative mechanism for PCDDs.
However, because of the low volatility of PCDDs few studies have been attempted to
measure actual rates of photodegradation. Also, if airborne PCDDs are predominantly
sorbed to particulates, then vapor phase photolysis measurements may be unimportant.
2-12 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
In studies conducted at Stanford Research Institute, the half-life of gaseous
2,3,7,8-TCDD in sunlight was found to be 5 to 24 days, and may be the result of indirect
photolysis through the attack of hydroxyl radicals on the dioxin. This mode of destruction
would be similar to the attack of hydroxyl radicals on PCBs (Leifer et al. 1983).
Mill et al. (1987) reported preliminary photolysis experiments with vapor phase
2,3,7,8-TCDD. The half-life for vapor phase 2,3,7,8-TCDD in simulated sun was several
hours. The photolysis of 2,3,7,8-TCDD sorbed onto small diameter fly ash particulates
suspended in air was also measured. The results indicated that fly ash appears to confer
photo-stability on 2,3,7,8-TCDD. There was little (8 percent) to no loss observed on the
two fly ash samples after 40 hours of illumination.
Orth et al. (1989) conducted photolysis experiments with vapor-phase
2,3,7,8-TCDD under illumination with a light source and filters to achieve radiation in the
UV region from 250 nm to 340 nm. Carrier gases included air, helium, and an
isobutane/helium mixture. The rate constants in helium and air were very similar, 5.4 x
10~3 sec"1 and 5.9 x 10~3 sec"1, which corresponds to a quantum yield in air of 0.033 +
0.046. No products could be observed in the mass spectrometer, so Orth et al. (1989)
postulated that the product might be sorbing to the surface of the photolysis cell and being
lost from potential analysis. Further studies were suggested to study product sorption to
surfaces and to determine any wave length dependence of the photoinduced loss across
the absorption band studied.
Podoll et al. (1986) estimated the photolysis rate of 2,3,7,8-TCDD vapors in the
atmosphere. Based on the quantum yield for photolysis in hexane, the half-life in summer
sunlight at 40 degrees north latitude was calculated to be 58 minutes, but Podoll et al.
(1986) stated this estimate is an upper limit.
• Oxidation
Stehl (1973) has suggested that 2,3,7,8-TCDD is probably stable to oxidation in
the ambient environment. The reaction rates of hydroxyl (OH) radicals with PCDDs have
not been measured. The low vapor pressures of these compounds makes direct
measurements very difficult with currently available techniques. However, Podall et al.
(1986) estimated the half-life of 2,3,7,8-TCDD vapor via OH oxidation in the atmosphere
2-13 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
to be 200 hours. Atkinson (1987) estimated the atmospheric lifetime for 2,3,7,8-TCDD
due to the OH radical reaction to be about 3 days.
• Hydrolysis
There is no available evidence indicating that hydrolysis would be an operative
environmental process for degradation of PCDDs (EPRI 1983, Miller and Zepp 1987).
• Volatilization and Sorption
Based on their very low vapor pressures and high Koc values, PCDDs would be
expected to sorb to rather than volatilize from soils, sediments, and other solids; however,
due to the stability and persistence of PCDDs via other transformation and transport
pathways, volatilization should not be ignored as a transport mechanism. Freeman and
Schroy (1985) point out that low volatility chemicals may bind strongly with dry soil but,
once a molecular microlayer of water covers the soil particles, the chemical should become
more volatile; in addition, water vaporization may enhance the rate of chemical
vaporization from a soil column.
Observations from the Seveso incident indicate that when 2,3,7,8-TCDD is
deposited on the soil surface, both volatilization and photodegradation are initially rapid
(DiDomenico et al. 1982). Once it has been washed into the soil, however, 2,3,7,8-TCDD
is unaffected by photolysis, and volatilization is greatly reduced (Plimmer 1978).
Concerning the volatility of 2,3,7,8-TCDD from water, Podoll et al. (1986) utilized
data for the molecular diffusivity in air and in water to calculate volatilization half-lives.
The calculated volatilization half-life was about 32 days for ponds and lakes and about 1 5
days for rivers.
Palausky et al. (1986) injected 2,3,7,8-TCDD dissolved in various organic solvents
into soil columns to determine the extent of vapor phase diffusion in soil and the effects of
carrier medium on the extent of migration. No noticeable changes in 2,3,7,8-TCDD
concentration profiles were observed after 30 days incubation at temperatures ranging
from 0° to 20°; however, a measurable change was observed at 40°C. The extent of
migration was related to solubility of 2,3,7,8-TCDD in the solvent, but a direct correlation
between solubility and extent of migration was not found.
2-14 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Puri et al. (1989) of the same lab as Palausky reported continuing research which
indicates that 2,3,7,8-TCDD is highly sorptive, but that organic co-contaminants such as
waste oil and surfactants may act to enhance the translocation of 2,3,7,8-TCDD. Also,
the distribution coefficient of 2,3,7,8-TCDD in soils is dependent on the total organic
matter, with higher sorption at higher soil organic levels, but the degree of sorption is also
dependent on the types of organic matter present and their stability and solubility in the
presence of the various co-contaminants.
Kapila et al. (1989) of the same lab as Palausky and Puri tested the soil migration of
2,3,7,8-TCDD in waste crankcase oil. Field experiments showed that 2,3,7,8-TCDD
moved downward in the soil column, and this movement was attributed to application of
water to simulate rainfall. The water application resulted in displacement of the waste oil
components and 2,3,7,8-TCDD from the macropore spaces in the soil column. The
formation of colloidal suspensions and the presence of 2,3,7,8-TCDD in these suspensions
may play a significant role in the downward movement of 2,3,7,8-TCDD. Since the
amount of 2,3,7,8-TCDD recovered from each column after 12 months was
"approximately the same as the amount initially applied," it was inferred by Kapila et al.
(1989) that there was little loss of 2,3,7,8-TCDD from the soil surface via volatilization or
photolysis during the one-year period.
Muir et al. (1985) studied the fate of radiolabeled 1,3,6,8-TCDD in sandy loam soil
under field conditions and in silty-clay pond and lake sediments under laboratory
conditions. Under field conditions, the time for disappearance of a surface-applied dilute
solution of 1,3,6,8-TCDD was in the 130 to 400 day range. The authors attributed the
losses to volatilization and/or erosion because no downward movement in the soil and no
biodegradation could be detected. In sediment, 80 percent of the intact chemical was still
present after 675 days.
• Biotransformation and Biodegradation
Investigations on the biodegradability of PCDDs have focused on the microbial
degradation of 2,3,7,8-TCDD. Arthur and Frea (1989) provide a comprehensive review of
studies conducted during the 1970s and 1980s. Arthur and Frea (1989) conclude that
2,3,7,8-TCDD is recalcitrant to microbial degradation. Several of the major studies
conducted during this period are discussed below.
2-15 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Matsumura and Beriezet (1973) tested approximately 100 strains of
micro-organisms which had previously shown the ability to degrade persistent pesticides,
and only five strains showed any ability to degrade 2,3,7,8-TCDD, based on
autoradiographs of thin-layer chromatograms. Although it is possible that the less
chlorinated dioxins are more susceptible to biodegradation, microbial action on
2,3,7,8-TCDD is very slow under optimum conditions (Mutter and Philippi 1982).
Long-term incubations of radiolabeled 2,3,7,8-TCDD yielded no radioactivity in carbon
dioxide traps after one year, and analyses of the cultures showed that at most, 1 to 2
percent of a potential metabolite (assumed to be an hydroxylated derivative of 2,3,7,8-
TCDD) could be detected. Camoni et al. (1982) added organic compost to contaminated
soil from the Seveso area in an attempt to enrich the soil and enhance the 2,3,7,8-TCDD
biodegradation rate, but the soil amendment had no clear effect on degradation.
Bumpus et al. (1985) tested the white rot fungus, Phanerochaete chrysosporium,
which secretes a unique H202-dependent extracellular lignin-degrading enzyme system
capable of generating carbon-centered free radicals. Lignin is resistant to attack by all
microorganisms except some species of fungi and a relatively small number of species of
bacteria. Radioiabeled 2,3,7,8-TCDD was oxidized to labeled C02 by nitrogen-deficient,
ligninolytic cultures of P. chrysosporium, and since the label was restricted to the ring, it
was concluded that the strain was able to degrade halogenated aromatic rings. In 10 ml
cultures containing 1,250 pmol of substrate, 27.9 pmol of 2,3,7,8-TCDD were converted
to Iabeled-C02 during the 30-day incubation period, thus only about 2 percent of the
starting material was converted.
2.5.2. Polychlorinatecf Dibenzofurans (PCDFs)
There is little information on the environmental transport and fate of PCDFs.
However, the available information on the physical/chemical properties of PCDFs indicate
that PCDFs should resemble PCDDs in environmental behavior. In general, PCDFs are
expected to be extremely stable compounds under normal environmental conditions and to
be strongly sorbed to soils, sediments and particulate matter. Erosion and aquatic
transport of sediment is expected to be the main transport mechanism; the potential for
significant leaching and volatilization are minimal. The only environmentally significant
path for destruction of PCDFs is apparently photodegradation.
2-16 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
• Photodegradation
PCDFs absorb electromagnetic radiation at wavelengths above 290 nm and should
be capable, therefore, of undergoing photolysis when subjected to sunlight (EPRI 1983).
The potential for photodegradation of PCDFs in the environment appears to be similar to
the photodegradation potential of PCDDs; in the presence of a hydrogen donor and
sunlight, PCDFs will dechlorinate. The major photoproducts are lower chlorinated PCDFs.
Crosby et al. (1973) report that polychlorinated dibenzofurans undergo photolytic
dechlorination in the presence of a hydrogen donor, with more highly chlorinated
congeners being more stable. In contrast, Hutzinger (1973) and Buser (1976) report that
the more highly chlorinated congeners undergo photodegradation at a rate similar to that of
lower chlorinated PCDFs. Hutzinger (1973) found that both 2,8-DCDF and OCDF
photolyze rapidly in methanol and hexane.
Buser (1988) studied the photolytic decomposition rates of 2,3,7,8-TCDF, 1,2,3,4-
TCDF, and 1,2,7,8-TCDF. Studies were performed in dilute isooctane solutions and as
solid phases on quartz surfaces under sunlight and artificial laboratory illumination (fluores-
cent lights). When the solutions were illuminated with sunlight, the estimated half-lives
were 220 minutes for a solution containing 3 ng/ul of 2,3,7,8-TCDF, 180 minutes for a
solution containing 2 ng/ul of 1,2,3,4-TCDF, and 600 minutes for a solution containing 0.3
ng/ul of 1,2,7,8-TCDF. For the same solutions illuminated with artificial light, the half lives
were greater than 28 days. When the same solutions dispersed and dried as thin films in
quartz vials and exposed to sunlight, the estimated half-lives were reported to be 120
hours, 95 hours, and 35 hours, respectively.
• Oxidation
Although there is no information available on the oxidation of PCDFs under
environmentally relevant conditions, PCDFs are expected to be stable to oxidation (EPRI
1983).
• Hydrolysis
No information is available indicating that hydrolysis would be an operative
environmental process for degradation of PCDFs (EPRI 1983).
2-17 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
• Volatilization and Sorption
Based on their very low vapor pressures and high Koc values, PCDDs would be
expected to sorb to rather than volatilize from soils, sediments, and other solids; however,
due to the stability and persistence of PCDDs via other transformation and transport
pathways, volatilization should not be ignored as a transport mechanism.
• Biotransformation and Biodegradation
No information was found relating specifically to the biodegradability of PCDFs.
However, structurally similar PCDDs are considered to be essentially nonbiodegradable in
the environment (EPRI 1983) and, like other persistent compounds, PCDFs are found in
soils, sediments, and biota. Therefore, PCDFs are expected to behave in a like manner and
remain persistent to attack by microorganisms and other types of biotransformation
processes.
2.5.3 Coplanar PCBs
Little information exists on the environmental transport and fate of the specific
coplanar PCBs. However, the available information on the physical/chemical properties of
coplanar PCBs coupled with the body of information available on the widespread
occurrence and persistence of PCBs in the environment indicates that these coplanar PCBs
are likely to be strongly sorbed by soils and sediments, and to be thermally and chemically
stable. Photodegradation of the more highly chlorinated congeners followed by slow
aerobic biodegradation is believed to be the principal path for destruction of PCBs.
The following sections present brief summaries of readily available information on
the fate of the 11 coplanar PCBs. Because few studies have been performed, much of the
available information was based on estimation techniques.
• Photodegradation
Leifer et al. (1983) summarized available data on photolysis of PCB isomers. Based
on the available data, Leifer et al. (1983) concluded that all PCBs, especially the more
highly chlorinated congeners and those that contain two or more chlorines in the ortho
position, photodechlorinate. In general, as the chlorine content increases, the photolysis
rate increases and the half-life decreases. The products of photolysis are predominantly
2-18 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
lower chlorinated PCBs. In addition, Choudry and Webster (1987) report that the
photolysis of PCBs in certain solvent systems yields PCDFs.
• Oxidation
PCBs are extremely resistant to oxidation as evidenced by the high temperatures
required to achieve thermal destruction. However, reaction in the atmosphere with OH
radicals may be a significant degradation mechanism. Leifer et al. (1983) estimated that
atmospheric transformation occurs reasonably fast for those PCB congeners containing
either a small number of chlorines or which have all or most of the chlorines on one ring.
For the 11 coplanar PCBs, the predicted half-lives are: for the TeCBs, 11 to 20 days; for
the PeCBs, 12 to 31 days; for the HxCBs, 32 to 62 days; and for the HpCBs, 94 days.
• Hydrolysis
There are no experimental data published on the hydrolysis of PCBs under
environmental conditions (Leifer et al. 1983). However, it is expected that the attachment
of chlorines directly to the aromatic ring in PCBs confers hydrolytic stability (Leifer et al.
1983).
• Volatilization and Sorption
Based on their very low vapor pressures and high Koc values, PCDDs would be
expected to sorb to rather than volatilize from soils, sediments, and other solids; however,
due to the stability and persistence of PCDDs via other transformation and transport
pathways, volatilization should not be ignored as a transport mechanism. Freeman and
Schroy (1985) point out that low volatility chemicals may bind strongly with dry soil but,
once a molecular microlayer of water covers the soil particles, the chemical should become
more volatile; in addition, water vaporization may enhance the rate of chemical
vaporization from a soil column.
• Biotransformation and Biodegradation
Leifer et al. (1983) summarized the available information on the degradation of
PCBs by microorganisms. The following general findings were made. There are numerous
aerobic microorganisms in the environment that are capable of degrading most PCBs and
2-19 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
that such organisms are widely distributed in the environment. In general, the rate of
aerobic biodegradation decreases with increasing chlorination. For example, the half-lives
resulting from biodegradation for TeCBs in fresh surface water and soil are 7 to 60+ days
and 12 to 30 days, respectively. For PeCBs and higher chlorinated PCBs, the half-lives in
fresh surface water and soil are likely to exceed one year. Also, PCBs with all or most
chlorines on one ring and PCBs with fewer than two chlorines in the ortho position tend to
degrade more rapidly. Finally, there is no evidence for biodegradation of PCBs under
anaerobic conditions.
2.6 FATE - BROMINATED COMPOUNDS
Although there are no available published studies documenting measured fate rate
constants, relatively few studies with measured physical/chemical property data, and few
relevant environmental monitoring studies, it is possible to estimate the environmental
transport and transformation processes for major PBDDs, PBDFs, and PBBs using structure
activity and property estimation methods. Mill (1989) performed such an assessment and
much of what is reported in this section is a summary of that review paper.
Mill (1989) concluded that the estimated physical/chemical properties of these
compounds indicate that will behave in a similar fashion to their chlorinated analogs. In
general, these chemicals are expected to be stable under normal environmental
conditions, relatively immobile in the environment, and primarily associated with
particulate and organic materials. The only environmentally significant path for destruction
is photodegradation. If discharged to the atmosphere, any vapor phase compounds will
probably be rapidly photolyzed.
Upon deposition onto surfaces, there can be an initial loss due to photodegradation
and/or volatilization. Once sorbed onto soils or sediments, however, they are expected to
be strongly sorbed with erosion and aquatic transport of sediment the dominant physical
transport mechanism. If discharged to water, they are expected to preferentially sorb to
solids. Volatilization may also be a significant transport mechanism for non-sorbed
chemicals even though they have negligible estimate vapor pressures.
2-20 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
• Photodegradation
Photolysis in the atmosphere appears to be a major pathway for loss of PBDDs and
PBDFs based on recent studies by Buser. Buser (1988) studied the photolytic
decomposition rates of the following compounds in dilute isooctane solutions and as solid
phases on quartz surfaces under sunlight and artificial laboratory illumination: 1,2,3,4-
TBDD; 2,3,7,8-TBDD; 2,3,7,8-TBDF; and mono- and dibrominated 2,3,7,8-TCDD and
2,3,7,8-TCDF. Under natural sunlight, estimated half-lives were very short, on the order
of minutes. Solid phase photolysis was significantly slower, in the range of 7 to 35 hours.
The major photolytic pathway was reductive dehalogenation with the formation of lower
halogenated or unsubstituted dibenzo-p-dioxins and dibenzofurans.
Mill (1989) used the results obtained by Buser (1988) together with assumptions to
overcome the lack of quantum yield data from Buser (1988) to estimate the photolysis
half-lives of the three brominated-only compounds tested by Buser (1988). Mill (1989)
estimated the following half-lives in water (top one meter) and for vapor in air (first
kilometer above surface) for clear-sky conditions in mid-summer at 40 degrees north
latitude:
Half-life Half-life
Compound in water (hrs) in air (min)
1,2,3,4-TBDD 7 <1
2,3,7,8-TBDD 2 0.3
2,3,7,8-TBDF 1.7 0.2
• Oxidation
The reaction rates of OH radicals with PBDDs, PBDFs, and PBBs have not been
measured. The low vapor pressures of these compounds makes direct measurements very
difficult with the current techniques. However, Mill (1989), using a structure activity
relationship developed by Atkinson (1987), has estimated the half-lives of OH oxidation for
the tetra- through octa- PBDDs and PBDFs. The estimated half-lives listed below indicate
that OH oxidation is probably too slow to compete with photolysis.
2-21 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
PBDD PBDF
No. of Br Half-life (hrs) Half-life (hrs)
4 50 420
5 50 430
6 100 960
7 200 1900
8 770 3800
• Hydrolysis
There is no available evidence indicating that hydrolysis would be a significant
degradation process for these compounds.
• Volatilization and Sorption
Little information exists on the environmental transport of PBDDs, PBDFs, and
PBBs. However, the available information on the physical/chemical properties of these
compounds and their chlorinated analogs coupled with the body of information available on
the widespread occurrence and persistence of the chlorinated analogs in the environment
indicates that these compounds are likely to be strongly sorbed by soils and sediments and
to be resistant to leaching and volatilization.
• Biotransformation and Biodegradation
Although there are no data available concerning the biodegradability of the
brominated analogs of PCDDs, PCDFs, and PCBs, it is expected that these brominated
analogs, especially the more halogenated congeners, will be recalcitrant to biodegradation.
2-22 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
REFERENCES FOR CHAPTER 2
Abramowitz, R.; Yalkowsky, S.H. (1990) Estimation of aqueous solubility and melting
point for PCB congeners. Chemosphere 21(10/11): 1221-1229.
Arthur, M.F., Frea, J.I. (1989) 2,3,7,8-Tetrachlorodibenzo-p-dioxin: aspects of its
important properties and its potential biodegradation. J. Environ. Qual. 18:1-11.
Atkinson, R. (1987) Estimation of OH radical reaction rate constants and atmospheric
lifetimes for polychlorobiphenyls, dibenzo-p-dioxins, and dibenzofurans. Environ.
Sci. Technol. 21:305-307.
Botre, C., Memoli, A., AI-Haique, F. (1978) TCDD solubilization and photodecomposition in
aqueous solutions. Environ. Sci. Technol. 12:335-336.
Bumpus, J.A., Tien, M., Wright, D., Aust, S.D. (1985) Oxidation of persistent
environmental pollutants by a white rot fungus. Science 228:1434-1436.
Burkhard, L.P.; Kuehl, D.W. (1986) N-octanol/water partition coefficients by reverse
phase liquid chromatography/mass spectrometry for eight tetrachlorinated planar
molecules. Chemosphere 15(2): 163-167.
Buser, H.R. (1976) Preparation of qualitative standard mixtures of polychlorinated dibenzo-
p-dioxins and dibenzofurans by ultraviolet and gamma irradiation of the
octachlorocompounds. J. Chromatogr. 192:303-307.
Buser, H.R. (1988) Rapid photolytic decomposition of brominated and
brominated/chlorinated dibenzodioxins and dibenzofurans. Chemosphere 17:889-
903.
Camoni, I., Dimuccio, A., Pontecorvo, D., Taggi, F., Vergori, I. (1982) Laboratory
investigation for the microbial degradation of 2,3,7,8-tetrachlorodibenzo-p-dioxin in
soil by addition of organic compost. Pergammon Ser. Environ. Sci. 5:95-103.
Choudhry, G.G., Webster, G.R.B. (1987) Environmental photochemistry of polychlorinated
dibenzofurans (PCDFs) and dibenzo-p-dioxins (PCDDs): A review. Toxicol. Environ.
Chem. 14:43-61.
Choudhry, G.G., Webster, G.R.B. (1989) Environmental photochemistry of PCDDs. 2.
Quantum yields of direct phototransformation of 1,2,3,7-tetra-, 1,3,6,8-tetra-,
1,2,3,4,6,7,8-hepta-, and 1,2,3,4,6,7,8,9-octachlorodibenzo-p-dioxin in aqueous
acetonitrile and their sunlight half-lives. J. Agric. Food Chem. 37:254-261.
2-23 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Choudhry, G.G.; Foga, M.; Webster, G.R.B.; Muir, D.C.G.; Friesen, K. (1990) Quantum
yields of the direct phototransformation of 1,2,4,7,8-penta- and 1,2,3,4,7,8-hexa
chlorodibenzofuran in aqueous acenitrile and their sunlight half-lives. Toxicology
and Environ, norua! Chemistry 26:181-195.
Crosby, D.G., Moilanen, K.W., Wong, A.S. (1973) Environmental generation and
degradation of dibenzodioxins and dibenzofurans. Environ. Health Perspect., Exp.
Issue. 5:259-266.
Crosby, D.G. (1978): Conquering the monster - the photochemical destruction of
chlorodioxins. In: Disposal and decontamination of pesticides. M.V.Kennedy, Ed.
ACS Symposium Series 73:1-12.
Crosby, D.G. (1981) Methods of photochemical degradation of halogenated dioxins in view
of environmental reclamation. Paper presented on "Human health aspects of
accidental exposure to dioxins. Strategy for environmental reclamation and
community protection," Bethesda, MD, October 5-7, 1982.
Davis, C.A.: University of California, Department of Environmental Toxicology.
Dickhut, R.M.; Andren, A.W.; Armstrong, D.E. (1986) Aqueous solubilities of six
polychlorinated biphenyl congeners at four temperatures. Environ. Sci. Technol.
20(8): 807-810.
DiDomenico, A., Viviano, G., Zapponi, G. (1982) Environmental persistence of 2,3,7,8-
TCDD at Seveso. In: Chlorinated dioxins and related compounds, impact on the
environment. O. Hutzinger et al., Eds. Elmsford, NY: Pergamon Press, pp. 105-
113. (1983).
Dulin, D., Drossman, H., Mill, T. (1986) Products and quantum yields for photolysis of
chloroaromatics in water. Environ. Sci. Technol. 20:72-77.
Dunnivant, P.M.; Elzerman, A.W. (1988) Aqueous solubility and henry's law
constant data for PCB congeners for evaluation of quantitative structure-property
relationships (QSPRs). Chemosphere 17(3): 525-541.
EPRI (1983) Electric Power Research Institute. State-of-the-art review; PCDDs and PCDFs
in utility fluid. Palo Alto, CA: EPRI-CS-3308.
Esposito, M.P., Teirnan, T.D., Dryden, F.E. (1980) Dioxins. U.S. EPA, IERL, Office of
Research and Development, Cincinnati, OH. EPA-600/2-80-197.
Foreman, W.T.; Bidleman, T.F. (1985) Vapor pressure estimates of individual
polychlorinated biphenyls and commercial fluids using gas chromatographic
retention data. J. Chromatog. 330: 203-216.
Freeman, R.A., Schroy, J.M. (1985) Environmental mobility of TCDD. Chemosphere
14:873-876.
2-24 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Friesen, K.J.; Vilk, J.; Muir, D.C.G. (1990) Aqueous solubilities of selected 2,3,7,8-
substituted polychlorinated dibenzofurans (PCDFs). Chemosphere 20(1-2): 27-32.
Friesen, K.J.; Sarna, L.P.; Webster, G.R.B. (1985) Aqueous solubility of polychlorinated
dibenzo-p-dioxins determined by high pressure liquid chromatography chemosphere.
14(9): 1267-1274.
Friesen, J.K., Muir, D.C.G., Webster, G.R.B. (1990) Evidence of sensitized photolysis of
polychlorinated dibenzo-p-dioxins in natural waters under sunlight conditions.
Environ. Sci. Technol. 24(11):1739-1744.
Hawker, D.W.; Connell, D.W. (1988) Octanol-water partition coefficients of
polychlorinated biphenyl congeners. Environ. Sci. Technol. 22(4): 382-387.
Mutter, R., Philippi, M. (1982) Studies in microbial metabolism of TCDD under laboratory
conditions, in: Chlorinated dioxins and related compounds, impact on the
environment. 0. Hutzinger et al. eds. Elmsford, NY: Pergammon Press, pp. 87-93
as cited in Vuceta et al. (1983).
Hutzinger, 0. (1973) Photochemical degradation of di and octachlorodibenzofuran.
Environ. Health. Perspect. Exp. Issue. 5:253-256.
Kapila, S., Yanders, A.F., Orazio, C.E., Meadows, J.E., Cerlesi, S., Clevenger, T.E. (1989)
Field and laboratory studies on the movement and fate of tetrachlorodibenzo-p-
dioxin in soil. Chemosphere 18:1297-1304.
Leifer, A., Brink, R.H., Thorn, G.C., Partymiller, K.G. (1983) Environmental transport and
transformation of polychlorinated biphenyls. Washington, D.C.: U.S.
Environmental Protection Agency, Office of Toxic Substances. EPA-560/5-83-025.
Lyman, W.J.; Reehl, W.F.; Rosenblatt, D.H. (1982) Handbook of chemical property
estimation methods. New York, NY: McGraw-Hill.
Mackay, D.; Bobra, A.; Chan, D.W.; Shiu, W.Y. (1982) Vapor pressure correlations for
low-volatility environmental chemicals. Environ. Sci. Technol. 16(10): 645-649.
Marple, L.; Brunck, R.; Throop, L. (1986a) Water solubility of 2,3,7,8-tetrachlorodibenzo-
p-dioxin. Environ. Sci. Technol. 20(2): 180-182.
Marple, L.; Berridge, B.; Throop, L. (1986b) Measurement of the water-octanol partition
coefficient of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Environ. Sci. Technol. 20(4):
397-399.
Matsumura, F., Benezet, J.H. (1973) Studies on the bioaccumulation and microbial
degradation of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Environ. Health Perspect. Sept.
253-258.
2-25 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Mill, T., Rossi, M., McMillen, D., Coville, M., Leung, D., Spang, J. (1987) Photolysis of
tetrachlorodioxin and PCBs under atmospheric conditions. Internal report prepared
by SRI International for USEPA, Office of Health and Environmental Assessment,
Washington, D.C.
Mill, T. (1989) Environmental fate of polybrominated dibenzodioxins and dibenzofurans.
Office of Toxic Substances, Washington, D.C.
Miller, G.C., Zepp, R.G. (1987). 2,3,7,8-Tetrchlorodibenzo-p-dioxin: environmental
chemistry. In: Exner, J.H. ed. Solving hazardous waste problems -- learning from
dioxins. Washington, D.C.: American Chemical Society.
Muir, C.G., Yarechewski, A.L., Corbet, R.L., Webster, G.R.B., Smith, A.E. (1985)
Laboratory and field studies on the fate of 1,3,6,8 - tetrachlorodibenzo-p-dioxin in
soil and sediments. J. Agric. Food Chem. 33:518-523.
Orth, R.G., Ritchie, C., Hileman, F. (1989) Measurement of the photoinduced loss of vapor
phase TCDD. Chemosphere 18:1275-1282.
Palausky, J., Kapila, S., Manahan, S.E., Yanders, A.F., Malhotra, R.K., Clevenger, T.E.
(1986) Studies on vapor phase transport and role of dispersing medium on mobility
of 2,3,7,8-TCDD in soil. Chemosphere 15:1387-1396.
Plimmer, J.R. (1978) Photolysis of TCDD and trifluralin on silica and soil. Bull. Environ.
Contam.Toxicol. 20:87-92.
Podoll, R.T., Jaber, H.M., Mill, T. (1986) Tetrachlorodibenzodioxin: Rates of volatilization
and photolysis in the environment. Environ. Sci. Technol. 20:490-492.
Podoll, R.T.; Jaber, H.M.; Mill, T. (1986) Tetrachlorodibenzodioxin rates of volatilization
and photolysis in the environment. Envrion. Sci. Technol. 20(5): 490-492.
Puri, R.K., Clevenger, T.E., Kapila, S., Yanders, A.F., Malhotra, R.K. (1989) Studies of
parameters affecting translocation of tetrachlorodibenzo-p-dioxin in soil.
Chemosphere 18:1291-1296.
Risby, T.H.; Hsu, T.B.; Sehnert, S.S.; Bhan, P. (1990) Physiochemical parameters of
individual hexachlorobiphenyl congeners. Environ. Sci. Technol. 24(11): 1680-
1687.
Rordorf, B.F. (1989) Prediction of vapor pressures, boiling points and enthalpies of fusion
for twenty-nine halogenated dibenzo-p-dioxins and fifty-five dibenzofurans by a
vapor pressure correlation method. Chemosphere 18(1-6): 783-788.
Rordorf, B.F. (1987) Prediction of vapor pressures, boiling points, and enthalpies of
fusion for twenty-nine halogenated dibenzo-p-dioxins. Thermochimica Acta 112:
117-122.
2-26 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Sabljic", A.; Glisten, H. (1989) Predicting Henry's Law constants for polychlorinated
biphenyls. Chemosphere 19(10/11): 1503-1511.
Sarna, L.P.; Hodge, P.E.; Webster, G.R.B. (1984) Octanol-water partition coefficients of
chlorinated dioxins and dibenzofurans by reversed-phase HPLC using several C18
columns. Chemosphere 13(9): 975-983.
Shiu, W.Y.; Doucette, W.; Gobas, F.A.P.C.; Andren, A.; Mackay, D. (1988) Physical-
chemical properties of chlorinated dibenzo-p-dioxins. Environ. Sci. Technol. 22(6):
651-658.
Sijm, D.T.H.M.; Wever, H.; de Vries, P.J.; Opperhuizen, A. (1989) Octan-1-oi/water
partition coefficients of polychlorinated dibenzo-p-dioxins and dibenzofurans:
experimental values determined with a stirring method. Chemosphere 19(1-6):
263-266.
SRC. (1991) Syracuse Research Corporation calculated values from chemfate database
literature search.
Stehl, R.H. (1973) The stability of pentachlorophenol and chlorinated dioxins
to sunlight, heat, and combustion. In: Chlorodioxins: origin and fate. Blair, E.H.,
ed. Adv. Chem. Ser. 120:119-125.
U.S. Environmental Protection Agency. (1990) Background document to the integrated
risk assessment for dioxins and furans from chlorine bleaching in pulp and paper
mills. Office of Toxic Substances, Washington, DC; EPA 560/5-90-014.
Webster, G.R.B.; Friesen, K.J.; Sarna, L.P.; Muir, D.C.G. (1985) Environmental fate
modelling of chlorodioxins: determination of physical constants. Chemosphere
14(6/7):
2-27 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
3. ENVIRONMENTAL LEVELS OF PCDD, PCDF, AND PCB CONGENERS
3.1. INTRODUCTION
Polychlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs),
and polychlorinated biphenyls (PCBs) have been found throughout the world in practically
all media including air, soil, water, sediment, fish and shellfish, such other food products
as meat, milk, and vegetation. PCDDs and PCDFs are contaminants that can be released
to the environment as by-products resulting from the manufacture of such chlorinated
compounds as polychlorinated phenols, PCBs, phenoxy herbicides, hexachlorobenzene, and
chlorodiphenyl ethers. Although the manufacture of most chlorinated phenolic
intermediates and products, including PCBs, was terminated in the late 1970s, continued
use and disposal of those compounds can result in releases of PCDDs, PCDFs, and PCBs
to the environment. Significant releases may also result from the combustion of municipal
and chemical wastes, and burning such non-chlorinated organic substances as polystyrene,
cellulose, lignin, and coal in the presence of a chlorine donor. Dioxins can also be released
through the condensation of naturally occurring phenolic compounds such as that resulting
from chlorination of municipal water or the contribution of household bleaches flushed into
sewer systems. In addition, paper mills that use chlorine bleaching processes have been
identified recently as a source of both PCDDs and PCDFs. Consequently, PCDDs, PCDFs,
and PCBs become available for human exposure via the pathways mentioned previously.
There are 75 PCDD congeners and 135 PCDF congeners (based on the positioning
of between 1 and 8 chlorine atoms in these compounds). Of the 210 PCDD/PCDF
congeners, those that pose the greatest human health risk are the 2,3,7,8-substituted
compounds, and the congeners of greatest concern are 2,3,7,8-tetrachlorodibenzo-p-dioxin
(2,3,7,8-TCDD) and 2,3,7,8-tetrachlorodibenzofuran (2,3,7,8-TCDF). With regard to
PCBs, some congeners exhibit dioxin-like attributes in their biologic processes and toxic
effects. For the purposes of this report, particular attention will be paid to the distribution
of 2,3,7,8-TCDD and 2,3,7,8-TCDF congeners, and total CI4 to CI8 homologues in the
environment. In addition, distribution of the following PCB congeners in the various media
will be discussed:
3-1 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
IUPAC No. Congener
77 3,3',4,4'-tetra PCB
81 3,4,4',5-tetra PCB
105 2,3,3',4,4'-penta PCB
114 2,3,4,4',5-penta PCB
118 2,3',4,4',5-penta PCB
126 3,3',4,4',5-penta PCB
156 2,3,3',4,4',5-hexa PCB
157 2,3,3',4,4',5'-hexa PCB
167 2,3',4,4',5,5'-hexa PCB
169 3,3',4,4',5,5'-hexa PCB
189 2,3,3',4,4',5,5'-hepta PCB
The sections that follow present ranges of the aforementioned compounds in air, soil,
water, sediment, fish and shellfish, and food throughout the world. This literature
summary is not all inclusive, but is meant to present the reader with a general idea of
worldwide values reported in the literature.
3.2. CONCENTRATIONS IN SOIL
Tables B-1 and B-2 (Appendix B) contain summaries of data from the published
literature regarding concentrations of PCDDs and PCDFs in soil. Data on coplanar PCB
congener soil concentrations were not found in the literature; the PCB soil concentration
data that were found in the literature were reported as either total PCB concentrations or
concentrations of Aroclor PCB mixtures.
Soil samples from rural and semi-urban sites in England, Wales, and lowland
Scotland showed a general increase in concentration from the tetra (CI4) to the octa (CI8)
homologues of PCDD, whereas the PCDF levels showed very little variation between the
homologue groups (Greaser et al., 1989). Concentrations of 2,3,7,8-TCDD at those sites
ranged from <0.5 to 2.1 parts per trillion (ppt). The median values for the CI4 to CI8 CDD
homologues were 6.0, 4.6, 31, 55, and 143 ppt, respectively. The median values for the
CI4 to CI8 CDF homologues, on the other hand, were 16, 17, 32, 15, and 1 5 ppt.
Evaluation of soil data from urban sites in the same geographical area showed that the
mean levels for the PCDD and PCDF homologues were significantly greater (p<0.01) than
those for rural and semi-urban background soils (Greaser et al., 1990). Concentrations of
2,3,7,8-TCDD at those sites ranged from <0.5 to 4.2 ppt. The median values for the CI4
3-2 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
to CI8 CDD homologues were 40, 63, 141, 256, and 469 ppt, respectively. The median
values for the CI4 to CI8 CDF homologues were 140, 103, 103, 81, and 40 ppt. Although
the same argument was held as for the previous study that the median values for the data
set probably better reflect the typical background concentrations, the elevated levels of the
lower congeners, together with higher overall concentrations, are indicative that local
sources and short range transport mechanisms are major contributors of PCDDs and
PCDFs to urban soils.
Analysis of four sites in Hamburg, Germany, contaminated by an organochlorine
pesticide manufacturing company showed patterns of PCDD and PCDF homologue
distribution that are similar to the urban and industrial sites examined in England, Wales,
and Scotland (Sievers and Friesel, 1989). The study indicated that PCDDs and PCDFs
showed a regular increase from the CI4 to CI8 homologues (although individual data points
were not presented). The maximum concentrations of 2,3,7,8-TCDD ranged from 900 ppt
to 874,000 ppt. The very high concentrations of 2,3,7,8-TCDD at the sites were
attributed to an admixture of wastes from 2,4,5-T production.
Soil sampled in 1987 from the vicinity of a sewage sludge incinerator was
compared with soil from rural and urban sites in Ontario, Canada (Pearson et al., 1990).
Soil in the vicinity of the incinerator showed a general increase in concentration from the
CI4 to CI8 CDD homologues, whereas only the CI8 CDF homologue was detected (mean
concentration 43 ppt). Rural woodlot soil samples only contained the CI8 CDD homologue
(mean concentration of 30 ppt). Soil samples from undisturbed urban parkland settings
revealed only CI7 and CI8 CDD homologues, but all CDF homologues (CI4 to CI8) were
present. Those samples showed an increase in concentration from the CI7 to CI8 CDD
homologues and CI5 to CI8 CDF homologues. The CI4 homologue had the highest mean
value (29 ppt) of all the CDF homologues. Resampling of one of the urban sites in 1988,
however, showed high variability in the concentrations of PCDD and PCDF.
Data were collected on PCDD and PCDF homologues in soil samples from industrial,
urban, and rural sites in Ontario and some U.S. Midwestern states (Birmingham, 1990).
The levels of PCDD/PCDF in rural soils were primarily nondetected (ND), although the CI7
and CI8 CDD homologues were found in a few samples. In urban soils, the CI4 to CI8
homologues were measured for both CDDs and CDFs. The CI7 and CI8 CDD homologues
dominated the homologue profile, and were two orders of magnitude greater than in the
3-3 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
rural soils. These soils also contained measurable quantities of the CI4 and CI5 CDD
homologues. Industrial soils did not contain any CI4 or CI5 CDD homologues, but they
contained the highest levels of the CI4, CI7, and CI8 CDF homologues. In another study,
soils from industrialized areas of a group of cities from Midwestern and Mid-Atlantic states
(Ml, IL, OH, TN, PA, NY, WV, VA) were analyzed for levels of 2,3,7,8-TCDD (Nestrick et
al., 1986). Many of the samples were taken within one mile of major steel, automotive or
chemical manufacturing facilities, or municipal solid waste incinerators. Concentrations of
2,3,7,8-TCDD measured in this study ranged from ND to 9.4 ppt.
EPA conducted a 2-year nationwide study to investigate the national extent of
2,3,7,8-TCDD contamination (USEPA, 1987). The results of this large study were
summarized broadly in the primary reference (i.e., the number and types of samples per
site, and range of detection). Moreover, sampling for five of the seven "tiers" of the study
had detection limits in soil, sediment, and water of 1 part per billion (ppb). Only Tier 5
(sites where pesticides derived from 2,4,5-TCP have been or are being used for
commercial purposes) and Tier 7 (ambient sampling for fish and soil) had detection limits
of 1 ppt. Subsequently, the data from this study are not included in the tables, but some
observations from this study with regard to soil contamination are discussed below.
The 100 Tier 1 and 2 sites were investigated by the Superfund program as sites
already on or expected to be on the NPL list. Soil concentrations found in most of the
sites were in the ppb range, although in a few sites where concentrated 2,4,5-TCP
production wastes were stored or disposed of, concentrations were as high as 2,000 parts
per million (ppm). Off-site soil contamination of concern was confirmed in 7 of the 100
Tier 1 and 2 sites, with soil concentrations in the ppb range. Eleven of 64 Tier 3 sites
(facilities and associated disposal sites where 2,4,5-TCP and its derivatives were
formulated into pesticide products) were found to have soil concentrations exceeding 1
ppb, and in 7 of 11 sites where contamination was found, only one or two soil samples
were above 1 ppb. Fifteen of 26 Tier 5 sites (areas where 2,4,5-TCP and pesticide
derivatives were or are being used) had concentrations above 1 ppt, and one of those had a
single detection of 6 ppb. Two-thirds of all detections at the Tier 5 sites were below 5
ppt. Three of 18 Tier 6 sites (organic chemical and pesticide manufacturing facilities where
improper quality control on production processes could have resulted in 2,3,7,8-TCDD
being introduced into the waste streams) had soil concentrations that exceeded the
3-4 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
detection limit of 1 ppb, although these levels were limited to one or two samples per site.
Seventeen of the 221 urban soil sites and 1 of the 138 rural sites from Tier 7 (background
sites not expected to have contamination) had soil concentrations exceeding 1 ppt. The
highest concentration detected (11.2 ppt) was found in an urban sample. The results from
Tier 7 are consistent with the other studies discussed above regarding soil concentrations
of 2,3,7,8-TCDD in non-industrial settings.
Some general observations for PCDD and PCDF levels in soils are possible from the
data presented in the various soil studies discussed above:
• Generalizations about the prevalence of specific congeners within a
homologue group are not possible.
• As the degree of chlorination increases, the concentrations increase.
Concentrations of the CI7 and CI8 CDD and CDF homologues generally are
higher than the CI4, CI5, and CI6 homologues.
• Concentrations associated with industrial sites clearly are the highest, with
concentrations in the hundreds to thousands of parts per trillion.
Concentrations in settings identified as urban are higher than those in areas
identified as rural.
3.3. CONCENTRATIONS IN WATER
Tables B-3 and B-4 (Appendix B) contain summaries of data from the published
literature regarding concentrations of PCDDs and PCDFs in water. Data on coplanar PCB
congener water concentrations were not found in the literature.
PCDDs in surface water samples collected from the Eman River in southern Sweden
generally increased in concentration from the CI4 to the CI8 homologues, whereas the
PCDF levels showed very little variation between the homologue groups (Rappe et al.,
1989b). In general, however, the levels of PCDFs were higher than the level of PCDDs.
Concentrations of 2,3,7,8-TCDF were 0.022 parts per quadrillion (ppq) in Jarnsjon and
0.026 ppq in Fliseryd. The filtered water, before chlorination and distribution as drinking
water had no detectable CI4, CI5, or CI6 homologues for CDDs or CDFs, but the CI7 and
CI8 homologues for CDDs were detected at 120 and 170 ppq, respectively. However, the
concentration of the CI8 CDD homologue was close to the level found in the blank.
3-5 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
A survey of 49 drinking water supplies in Ontario, including supplies in the vicinity
of chemical industries and pulp and paper mills was initiated in 1983 (Jobb et al., 1990).
As of February 1989, 4,347 results were received for 399 raw and treated water samples.
The CI8 CDD homologue was detected in 36 of 37 positive results, and ranged from 9 to
175 ppq in raw samples and 19 to 46 ppq in treated samples. These low concentrations
primarily were found in raw water located downstream of industrialized areas in the St.
Clair/Detroit River system. The 2,3,7,8-TCDD congener was not detected in any sample.
Since PCDDs and PCDFs are hydrophobic compounds and consequently have a tendency
to sorb onto particulate matter in water, conventional water treatment processes can be
effective in removing the contaminants along with the particulates. This is substantiated
by the fact that 33 of the 37 positive results were raw water samples. Because of the
relatively low levels of PCDDs detected in the samples, however, it is difficult to ascertain
whether the PCDDs were particulate-associated or dissolved.
A survey of 20 community water systems throughout New York State was
conducted in 1986 (Meyer et al., 1989). The sampling sites were representative of the
major surface source waters in New York and included sources receiving industrial
discharges or known to contain dioxin-contaminated fish as well as waters in more remote
areas. The CI4 CDF homologue was detected in the finished water at the Lockport
(duplicate samples had concentrations of 2.1 and 2.6 ppq). Except for a trace of the CI8
CDF homologue detected at one location, no other PCDDs/PCDFs were detected in
finished water at any of the other 19 community water systems surveyed. Raw water
sampled at the Lockport facility contained concentrations of the CI4 CDD homologue (1.7
ppq) as well as CI4 to CI8 CDF homologues (18, 27, 85, 210, and 230 ppq, respectively).
As can be seen from the data, the PCDF homologue group concentrations increased with
increasing chlorine number.
Some general observations for PCDD and PCDF levels are possible from the data
presented in the various water studies above:
• Raw water samples generally have higher concentrations of PCDDs/PCDFs
than finished water samples.
• The concentration of PCDDs and PCDFs in surface water generally increase
from the CI4 to the CI8 homologues.
3-6 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
3.4. CONCENTRATIONS IN SEDIMENT
Tables B-5 through B-7 (Appendix B) contain summaries of data from the published
literature regarding concentrations of PCDDs, PCDFs, and coplanar PCB congeners in
sediment.
Sediment samples from the vicinity of a magnesium production plant were analyzed
for PCDDs and PCDFs (Oehme et al., 1989). The concentration distribution of PCDD and
PCDF congeners was rather homogeneous except for a slight decrease at a sampling
station further downstream of the plant. However, the deeper sediments (4-6 and 11-13
cm depth) at that site had somewhat higher levels. Another sampling station even further
downstream had concentrations that were a factor of 4 to 10 lower, thereby indicating
substantial transport of PCDDs and PCDFs. The CI4 CDF congener profiles were the same
as those for magnesium production. In addition, the CI5 CDF congener profiles were very
similar to those found in the waste water. Trapped sediments from the archipelago of
Stockholm, Sweden displayed PCDD and PCDF homologue distribution patterns that were
very similar to those exhibited in total air and air particulates (Rappe and Kjeller, 1987).
The CI7 and CI8 CDDs and CI7 CDF were the dominant homologues in the sediment.
Bottom surface sediment samples collected from the Baltic Sea showed interesting PCDD
and PCDF distribution patterns (Rappe et al.,1989a). The background samples, one
between the Swedish and Soviet coasts and the other between the Swedish and Finnish
coasts, contained similar levels and distribution profiles. The study indicated that the
pattern of the CI4 CDF congeners at these sites were typical of the "incineration pattern"
(i.e., patterns resulting from MSW incineration, car exhausts, steel mills, etc.) which also
had been found in samples of air and air particulates. However, sediment samples
collected at a distance of 4 to 30 km from a pulp mill revealed a congener distribution
pattern typical of bleaching mills. The CI4 CDF homologue in the sediment 4 km from the
pulp mill contained only two major congeners, and even the sediment collected 30 km
from the mill displayed the same pattern. Surface sediments collected from 18 lake areas
in central Finland were analyzed for PCDDs, PCDFs, and PCBs. Although 2,3,7,8-TCDD
was not detected in any of the samples, two other CI4 CDD congeners that are common
and abundant in pulp mill effluents were detected. In addition, CI6 to CI8 CDD congeners
as well as CI7 and CI8 CDF congeners that are not linked to pulp mills also were detected.
The study suggested that these may have resulted from combustion operations in the
3-7 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
more densely populated and industrialized areas in south Finland. Coplanar PCBs were
detected at low background levels. The most toxic PCB congener (IUPAC No. 126) was
found only in one area that is located nearest to a PCB leakage point. The concentration
of PCB 1 26 at that site was 110 ppt. Evaluation of sediments in Hamburg Harbor in
Germany revealed high concentrations of CI4 to CI8 CDD homologues (mean
concentrations of 564, 1112, 2744, 4040, and 7560 ppt, respectively) and CI4 to CI8
CDF homologues (mean concentrations of 526, 2980, 4106, 2358, and 2712 ppt) (Gotz
et al., 1990). The average concentration of 2,3,7,8-TCDD was 375.3 ppt. The high
concentrations of 2,3,7,8-TCDD, especially in the Moorfleeter Canal and the Auserer
Vering Canal was attributed to discharges from an organochlorine pesticide manufacturing
plant. The patterns of 2,3,7,8-TCDD and the other CI4 CDD congeners is characteristic of
the patterns resulting from the production of 2,4,5-T and 2,4,6-Trichlorophenol. In
addition, the pattern of the CI7 CDF congeners can be linked to emissions from thermal
processes employed by chemical industries in the production of chlorinated organic
chemicals. The high concentrations of CI7 and CI8 CDD and CDF homologues may also be
the result of other industrial combustion processes in the Hamburg area.
In sediment samples collected from estuaries adjacent to an industrial site in
Newark, New Jersey, where chlorinated phenols had been produced, the level of the CI8
CDD homologue was many times higher than that of 2,3,7,8-TCDD. The study indicated
that there probably is a significant regional source (i.e., combustion and use of a common
wood preservative, pentachlorophenol) for the CI8 CDD homologue, depleted in 2,3,7,8-
TCDD relative to the local industrial source. A high correlation was found between
2,3,7,8-TCDD and 2,3,7,8-TCDF (R2 = 0.87), which suggests that the industrial site was a
major source of 2,3,7,8-TCDF to the natural waters of the study area. An interesting note
is that the bottom section (108-111 cm) of one of the sediment cores contained 2,3,7,8-
TCDD at a concentration of 21,000 ppt, the highest concentration measured in the study.
This value was consistent with deposition of that sample during the mid to later stages of
active 2,4,5-T production at the site from the late 1950s to early 1960s. PCDD and PCDF
in Hudson River sediment samples primarily comprised the higher chlorinated CI6 to CI8
CDD and CDF homologues (Petty et al., 1982). Concentrations of the CI7 and CI8 CDD
homologues ranged from 5 to 15 ppb, and the CI8 CDD homologue in most instances
3-8 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
accounted for more than half of the total PCDD residue. Likewise, the CI7 and CI8 CDF
homologues occurred at the highest levels (ca. 1 ppb).
Surface sediment samples were collected from several estuaries in the United
States (Norwood et al., 1989). The sampling sites included Black Rock Harbor in
Bridgeport, Connecticut (an industrialized urban estuary); central Long Island Sound (a
relatively clean reference site); Narragansett Bay, Rhode Island (where chemical industries
may have contributed to the input); New Bedford Harbor, Massachusetts (a section of
which is a National Superfund Site because of PCB contamination); and Eagle Harbor,
Washington (the site of a creosote wood treatment facility). The sediments in New
Bedford Harbor were reported to be more heavily contaminated with PCDFs, especially
with regard to the CI6 congeners which were greater by a factor of 40 (although individual
data points were not presented). Sediments from Eagle Harbor, on the other hand,
practically were devoid of PCDFs, and showed a large increase in the CI7 and CI8 CDD
congeners closer to the treatment facility. Whereas Narragansett Bay and Black Rock
Harbor were quite similar in both concentration and distribution of PCDDs and PCDFs,
Black Rock Harbor contained slightly greater levels of the CI4 to CI6 CDD and CDF
congeners. Sediment from Long Island Sound was cleaner and had a distribution of PCDFs
between that of Narragansett Bay and Black Rock Harbor. Sediment with the least
contamination was collected in New Bedford Harbor, up-river from the PCB facilities, and
the highest CI8 CDD congener concentration (1400 ppt) was detected in Eagle Harbor.
Sediment samples from Siskiwit Lake, on Isle Royale, Lake Superior were examined
to evaluate the atmospheric input of PCDDs and PCDFs to the lake (Czuczwa et al., 1984).
The water level in Siskiwit Lake is 17 meters higher than that in Lake Superior, and in
addition, there are no anthropogenic inputs in the drainage basin of Siskiwit Lake.
Consequently, the atmosphere is the only source of anthropogenic chemicals in that lake.
The CI8 CDD homologue was most predominant, and the CI7 CDD and CDF homologues
also were abundant. The study indicated that the considerable decrease in concentration
of all PCDD and PCDF between 6 and 8 cm of the sediment core depth was the result of
the "1940 horizon."
Surficial sediments collected from Jackfish Bay on the north shore of Lake Superior
contained moderate concentrations of the CI4 CDF and CI8 CDD homologues, with trace
concentrations of other congeners (Sherman et al., 1990). The magnitude of the CI8 CDD
3-9 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
homologue was similar to that found in the Siskiwit Lake sediment samples. The CI4 CDF
and CI8 CDD profile for a sediment core collected from Moberly Bay was similar to the
surficial sediment pattern. Congeners from those homologues predominated at all depths
where detectable concentrations occurred. In addition, low concentrations of the CI7
CDD, and CI6 and CI7 CDF homologues were detected. The concentration profile of the
CI4 CDF homologue showed a relatively high value which dropped abruptly to non-
detectable «60 ppt) below a depth of 10 cm. This abrupt change corresponded to a
section date 1973 which reflects an operational change at the pulp mill, resulting in the
formation of CI4 CDF congeners in the mill effluent some time after 1973.
A survey of surficial harbor sediments collected near a wood preserving plant in
Thunder Bay, Ontario, Canada, on the north shore of Lake Superior, found PCDDs and
PCDFs, the highest concentrations of which occurred at stations closest to the plant dock,
and lower concentrations at locations further from the source (Mckee et al., 1990). No
PCDDs or PCDFs were detected below the surficial layer. CI4 and CI5 CDD homologues
were below analytical detection limits in all samples. However, the concentrations of the
CI6 to CI8 CDD homologues increased with the degree of chlorination. The maximum
concentrations of the CI6 to CI8 CDD homologues ranged from 5,700 ppt for the CI6
homologue to 980,000 ppt for the CI8 homologue. As with the PCDD homologue
distribution profile, the CI6 to CI8 CDF homologues increased with the degree of
chlorination.
Bottom surficial sediments (0-3 cm) were collected from the sedimentation basins
of Lake Ontario to assess the levels of the various PCB congeners there (Oliver and Niimi,
1988). Concentrations of PCB congeners 118, 105, and 156 in the sediment were 15,
10, and 2.1 ppb, respectively. A baseline assessment of PCDDs and PCDFs was
performed on the Elk River, a semi-rural area located about 25 miles northwest of
Minneapolis-St. Paul, Minnesota (Reed et al., 1990). Sediment samples were collected
from Lake Orono, a reservoir on the Elk River, and from an abandoned gravel pit. Although
none of the sediment samples contained 2,3,7,8-TCDD, the gravel pit sediments contained
measurable concentrations of the CI4 CDF homologue. Only one of the Lake Orono
samples contained measurable concentrations of 2,3,7,8-TCDF (0.31 ppt) and total CI4
CDF (0.54 ppt) homologue. The gravel pit samples also contained CI6 to CI8 CDD and CI5
to CI8 CDF homologues. Lake Orono samples contained CI7 and CI8 CDD and CI7 CDF
3-10 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
homologues. The CI7 CDD homologue ranged from 7.3 ppt in the lake inlet to 110 ppt in
the gravel pit and the lake, near the dam. The CI8 CDD homologue ranged from 450 ppt in
the gravel pit to 600 ppt in the middle of Lake Orono. The sediment profiles were
reminiscent of combustion source influences.
The Sheboygan River, a Wisconsin tributary to Lake Michigan, is polluted with PCBs
from the mouth to about 14 miles upstream (Sonzogni et al., 1991). That portion of the
river is a Superfund site as well as one of the Great Lakes "Areas of Concern." Sediment
cores were collected at Rochester Park, near the original source of the PCBs, about 14
miles upstream from the mouth. The PCS congeners 118, 105, and 77 were detected in
all samples, and ranged from about 5 to 1500 ppb. The remaining toxic PCB congeners,
81, 114, 167, 126, and 169 were detected less frequently and ranged from non-
detectable to slightly over 100 ppb. PCB congener 118 appears to be the most common
toxic congener in environmental samples, and was found in the Sheboygan River
sediments in the highest weight percent. The eight toxic PCBs presented in this study
were present in relatively low concentrations compared to total PCBs or other more
abundant congeners. The concentrations of these congeners were also low relative to the
congeners with which they co-elute.
Sediments collected from Waukegan Harbor in Lake Michigan contained PCB
congeners 77 and 105 (Huckins et al., 1988). The percentage of PCB 77 (0.16 +/- 0.15)
in the samples varied by 1.4 orders of magnitude, with concentrations ranging from 13 to
27,500 ppb. The percentage of PCB 105 averaged 0.66 +/- 0.37, with concentrations
ranging from 102 to 131,000 ppb. In another Lake Michigan study, sediment samples
collected from Green Bay contained concentrations of the toxic PCB congeners 81, 77,
118, 114, 105, 126, 167, 156, 157, 169, and 189 (Smith et al., 1990). PCB 118 and
105 were the dominant congeners, with concentrations of 11 and 5.8 ppb, respectively.
Some general observations for PCDD and PCDF levels are possible from the data
presented in the various sediment studies above:
• The PCDD and PCDF homologue distribution patterns in sediment generally
follow those exhibited by the contaminant source.
3-11 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
• The concentration of CI6 to CI8 PCDD and PCDF homologues in sediment are
usually the result of industrial combustion processes and generally increase
with the degree of chlorination, but decrease uniformly with distance from
the source.
3.5. CONCENTRATIONS IN FISH AND SHELLFISH
Tables B-8 through B-10 (Appendix B) contain summaries of data from the
published literature regarding concentrations of PCDDs, PCDFs, and coplanar PCB
congeners in fish and shellfish. PCB congener data were found only for North American
species.
Evaluation of fish in the Baltic Sea (Gulf of Bothnia) and Northern Atlantic Ocean in
the vicinity of Sweden revealed that concentrations of PCDDs and PCDFs in herring from
the Atlantic Ocean were lower than those in the Gulf of Bothnia (Rappe et al., 1989b).
Detectable levels of 2,3,7,8-TCDD in salmon were found in both wild homing (4.6 to 19
ppt) and hatchery-reared (0.2 to 0.3 ppt) varieties in the Gulf of Bothnia. In addition,
concentrations of the same representative congeners of the CI5 to Cl8 CDD and CDF
homologues found in herring were found in both varieties of salmon. Levels of those
congeners in the wild salmon, however, were five to ten times higher than the herring
levels, while the levels in the hatched salmon essentially were the same as in the herring
samples. Perch collected at a distance of 1-6 km from a pulp mill in the southern part of
the Gulf of Bothnia contained 2,3,7,8-TCDD and 2,3,7,8-TCDF, the levels of which were
higher in the samples collected closer to the pulp mill. Those two isomers have been
identified in bleaching effluents from pulp mills as well as in bleached pulp. Arctic char
collected from Lake Vattern, a popular fishing lake in southern Sweden, contained levels of
2,3,7,8-TCDD (6.5 to 25 ppt), 2,3,7,8-TCDF (21 to 75 ppt), and representative congeners
of the CI5 CDD and CDF homologues. There was a good correlation between the weight
of the fish and the levels of PCDDs and PCDFs. The main general pollution sources of the
long, deep, narrow lake are two pulp mills.
Fish (cod, haddock, pole flounder, plaice, flounder, and eel), mussels, and edible
shrimps from a fjord area contaminated by wastewater from a magnesium factory in
Norway were analyzed for PCDDs and PCDFs (Oehme et al., 1989). Magnesium
production results in the formation of substantial amounts of PCDDs and PCDFs as
3-12 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
byproducts. The pattern of CI4 and CI5 CDD and CDF homologues released in wastewater
during the magnesium production process is very characteristic, and is dominated by
congeners with chlorine in the positions 1,2,3,7, and/or 8. Fish and shellfish differ in
their ability to bioconcentrate PCDD and PCDF congeners. For example, fish generally only
concentrate the most toxic 2,3,7,8-substituted congeners, whereas shellfish can usually
concentrate most of the congeners. Nearly all congeners were present in the shrimp and
mussel samples. Whereas these organisms displayed the very characteristic CI5 CDF
congener pattern of the magnesium production process, some deviations were found in the
CI4 CDF congener distribution within those species. For fish, the concentrations of PCDDs
and PCDFs are dependent on the exposure level, fat content, living habit, and how
stationary the species is. The highest PCDD and PCDF levels were found in comparatively
fat bottom fish collected close to the source. Cod and haddock, leaner, non-stationary
fishes, had much lower concentrations, even in the vicinity of the magnesium production
factory. An interesting note is that the main stream of the fjord follows the west coast,
and subsequently, cod and eel samples collected along the west coast of the fjord had
considerably higher levels of PCDDs and PCDFs than those collected from the eastern fjord
entrance. Similarly, the level of 2,3,7,8-TCDD in mussels decreased by one order of
magnitude from the vicinity of the magnesium production factory to the outer region of the
fjord system.
Brown trout, grayling, barbel, carp, and chub collected in the Neckar River in
southwest Germany contained much higher levels of 2,3,7,8-TCDF than in eels collected
from the same river and the Rhine River (Frommberger, 1991). In addition, eels from both
rivers showed very similar patterns for PCDD and PCDF congener distribution, whereas the
patterns of PCDD and PCDF distribution generally showed some degree of difference
among the other fishes collected from the Neckar River. Perch and bream collected from
various locations in the vicinity of Hamburg Harbor, however, showed similar patterns in
the distribution of the CI4 to CI8 CDD and CDF homologues (Gotz et al., 1990). In general,
however, the levels of PCDFs were higher than the level of PCDDs in these fishes,
especially with regard to the CI4 to CI6 CDF homologues. Pooled samples of eels collected
at six different localities in the Netherlands contained low levels of PCDDs and PCDFs, the
major congeners of which were 2,3,7,8-chlorine substituted (Van den Berg et al., 1987).
The concentrations of the various congeners identified in the eel samples ranged from 0.1
3-13 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
to 9.1 ppt. The sample with the highest concentration of 2,3,7,8-TCDD (9.1 ppt} was
collected from Broekervaart in a location that was not far from a chemical waste dump
that contained high concentrations of the same congener.
Samples of striped bass, blue crabs, and lobsters collected from Newark Bay and
the New York Bight all contained high levels {up to 6200 ppt) of 2,3,7,8-chlorine
substituted tetra- and penta-CDDs and CDFs (Rappe et al., 1991). The levels of 2,3,7,8-
TCDD were higher than any other New Jersey samples, and the highest sample in this
study may be the highest level of 2,3,7,8-TCDD ever reported for aquatic animals. The
crustaceans resembled one another in congener pattern. Specifically, they all contained
both a large number and large amounts of PCDD and PCDF congeners in addition to the
2,3,7,8-chlorine substituted compounds. The striped bass samples, on the other hand,
primarily contained only the 2,3,7,8-chlorine substituted congeners.
Carp, catfish, striped bass, large mouth bass and lake trout were collected from
sites in the Hudson River and the Great Lakes Basin that were contaminated with industrial
chemicals or contained known or suspected levels of PCBs (Gardner and White, 1990).
The congener 2,3,7,8-TCDF was detected in 12 fish at levels which ranged from 3 to 93
ppt. A 2,3,7,8-chlorine substituted penta-CDF was detected in 14 fish at levels ranging
from 4 to 113 ppt. An interesting observation in this study was that 2,4,6-chlorine
substituted CDFs were detected in four of the fish samples, suggesting that those fish
may have been exposed to chlorinated phenols. The study indicated that the 2,4,6-
chlorine substituted CDFs occurred in the fish at levels similar to those of the 2,3,7,8-
chlorine substituted CDFs, but with less frequency.
Samples of lake trout or walleye collected from each of the Great Lakes and Lake
St. Clair were analyzed for PCDDs and PCDFs (De Vault et al., 1989). PCDF and PCDD
concentrations in lake trout were substantially different from each lake and between sites
in Lake Michigan, probably reflecting differences in types and amounts of loadings to the
lakes.
In all of the sampling sites except Lake Ontario, 2,3,7,8-TCDF was the dominant
furan congener in lake trout and ranged from 14.8 ppt in Lake Superior to 42.3 ppt in Lake
Michigan. In Lake Ontario, the dominant congener in lake trout was a 2,3,7,8-chlorine
substituted penta-CDF. The distribution of PCDF congeners in the Lake Erie walleye were
very similar to that of the lake trout from Lake Superior. With regard to dioxins, the
3-14 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
concentrations of 2,3,7,8-TCDD ranged from 1 ppt in Lake Superior to 48.9 ppt in Lake
Ontario. With the exception of Lake Ontario, the dominant dioxin congener was a 2,3,7,8-
chlorine substituted penta-CDD. A 2,3,7,8-chlorine substituted hexa-CDD also contributed
significantly to the total PCDD concentrations. As with PCDFs, the distribution of PCDD
congeners in the Lake Erie walleye was very similar to that of the lake trout from Lake
Superior.
In another study, PCDDs and PCDFs were measured in four species of salmonids
(lake trout, coho salmon, rainbow trout, and brown trout) that were collected from Lake
Ontario (Niimi and Oliver, 1989a). Levels of 2,3,7,8-TCDD in whole fish ranged from 6 to
20 ppt, and the CL6 CDD homologue was most dominant in all fishes. High levels of the
CL8 CDD homologue also were detected in lake trout and coho salmon, but not in rainbow
trout or brown trout. Although the total PCDF levels were about 25 percent lower than
the total PCDD concentrations, the levels of 2,3,7,8-TCDF which was the dominant
component of the CL4 CDF homologue was the same range as 2,3,7,8-TCDD (6 to 20
ppt). However, the study suggested that whereas collection sites can influence chemical
levels and congener composition, comparisons of chemical levels and congener frequencies
may not be suitable because of differences resulting from localized factors. The study also
indicated that the importance of the various PCDD and PCDF congeners can differ with
location (i.e., the same species of fish collected at different locations in a study area may
reveal that the most common congener is different at each site).
Channel catfish, carp, yellow perch, small mouth bass, sucker, and lake trout were
collected from the Tittabawassee, Grand, and Saginaw Rivers, Lake Michigan, and
Saginaw Bay, and analyzed for 2,3,7,8-TCDD residues (Harless and Lewis, 1980).
Detectable quantities were found in 26 of 36 samples, and ranged from 4 to 695 ppt. The
bottom feeders (catfish and carp) contained the highest concentrations, and the lowest
levels were found in bass, perch, and sucker. Because the Grand and Saginaw Rivers are
not connected to the Tittabawassee River, the presence of 2,3,7,8-TCDD in fish from
those rivers suggests that several separate bodies of water are contaminated with TCDD
residues.
Travis and Hattemer-Frey (1991) evaluated the data generated as part of the
National Dioxin Study regarding 2,3,7,8-TCDD concentrations in fish. The fish were
collected from 304 urban sites in the vicinity of population centers or areas with known
3-15 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
commercial fishing activity, including sites from the Great Lakes Region. Data from that
study indicated that concentrations of 2,3,7,8-TCDD in whole fish from urban sites ranged
from non-detectable to 85 ppt. In addition, only 29 percent of the fillets from urban sites
had detectable levels of 2,3,7,8-TCDD, with a geometric mean concentration of 0.3 ppt.
Whole fish samples from the Great Lakes Region had higher 2,3,7,8-TCDD levels than fish
from urban areas (e.g., 80% vs 35% detectable levels). In the Great Lakes Region,
2,3,7,8-TCDD concentrations in whole fish samples ranged from nondetectable to 24 ppt,
with a geometric mean of 3.8 ppt. These levels were ten times higher than the
concentration in whole fish from urban areas. Likewise, the mean concentration of
2,3,7,8-TCDD in Great Lakes Region fish fillets (2.3 ppt) was about seven times higher
than the levels in the fillets from urban areas (0.3 ppt). As with the whole fish samples,
fish fillet samples from the Great Lakes Region had higher 2,3,7,8-TCDD levels than fillets
from background urban areas (e.g., 67% vs 29% detectable levels). Comparable levels of
2,3,7,8-TCDD were detected in whole bottom feeders and predators from the Great Lakes
Region.
Samples from all trophic levels in the Lake Ontario ecosystem were analyzed for
PCB congeners (Oliver and Niimi, 1988). Analysis revealed that the PCB concentration
increased from water to lower organisms to small fish to salmonids, demonstrating the
classical biomagnification process. In addition, the chlorine content of the PCBs increased
at the higher trophic levels. PCBs with the highest chlorine content (57%) were found in
sculpin, small bottom-living fish that feed on benthic invertebrates. The CL3 and CL4 PCB
homologues comprised a much higher percentage of the PCBs in the lower trophic levels
than in salmonids and small fish. The percentage of CL5 and CL8 PCB homologues in all
samples was fairly uniform, but the CL6 and CL7 PCB homologues comprised a much
larger fraction of the PCBs in the small fish and salmonids than in the lower trophic levels.
A study regarding the distribution of PCBs in Lake Ontario salmonids (brown trout,
lake trout, rainbow trout, and coho salmon) showed that the CL5 and CL6 PCB
homologues were dominant in all species (Niimi and Oliver, 1989b). The 10 most common
PCB congeners represented about 52 percent of the total content and did not appear to be
influenced by species or total concentration. The homologues observed averaged
approximately 56 percent chlorine by weight in whole fish and muscle. The analysis of the
3-16 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
chlorine content suggested that the more persistent congeners tend to behave as a
homogeneous mixture instead of as individual congeners.
A large quantity of fish data were collected as part of EPA's National
Bioaccumulation Study during the period of 1986 to 1989. The report associated with
this study has been through several reviews, is currently in final Agency clearance
procedures and is expected to be released in 1993. Although the report has not yet been
released, the raw data is public information (Personal communication from Richard Healy,
US EPA Office of Water in Washington, DC, July 22, 1992). Based on this raw data,
several summaries were prepared and are presented here. Tables B-8 and B-9 include the
dioxin and furan data collected from a wide variety of sites across the U.S. including 314
sites thought to be influenced by point or nonpoint sources, and 34 sites identified as
relatively free of influence from point and nonpoint sources. This latter group of sites can
be characterized as background per the definition used in this document. Finally, Table B-
10 includes similar data for the various PCB congener groups from 362 sites. Since, the
specific PCB congeners could not be identified, it is not known what percentage of these
concentrations represent the PCBs identified as dioxin-like. Twenty of these sites were
identified as background sites. The total PCB, all 209 congeners, mean concentration for
these background sites was 46,900 ppt. Since the dioxin-like PCBs consist of only 11 of
the 209 possible PCB congeners, it may be that they are a small percentage of the total.
However, only congener specific analysis can ultimately confirm this.
Some general observations for PCDD and PCDF levels are possible from the data
presented in the various fish and shellfish studies above:
• Fish and shellfish differ in their ability to bioconcentrate PCDD and PCDF
congeners. Fish generally concentrate the most toxic 2,3,7,8-substituted
congeners, but shellfish can usually concentrate most of the congeners.
• For fish, the concentrations of PCDDs and PCDFs are dependent on the
exposure level, fat content, living habit, and the degree of movement of the
species. Comparatively fat bottom fish collected close to the contaminant
source generally have the highest PCDD/PCDF levels, whereas leaner, non-
3-17 • 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
stationary fishes have much lower concentrations, even in the vicinity of the
contaminant source.
• The National Dioxin Study indicated that the levels of 2,3,7,8-TCDD in fish
from the Great Lakes Region were higher than those from urban areas.
Comparable levels were detected in whole bottom feeders and predators
from the Great Lakes Region.
• With regard to PCBs, concentrations increase from water to lower organisms
to small fish to salmonids, and the chlorine content of the PCBs increase at
the higher trophic levels.
3.6. CONCENTRATIONS IN FOOD PRODUCTS
Tables B-11 and B-12 (Appendix B) contain summaries of data from the recent
published literature regarding concentrations of PCDDs and PCDFs in food products. Most
of the selected studies investigated "background" levels of PCDDs and PCDFs rather than
studies targeted at areas of known contamination. Data on relevant coplanar PCB
congener concentrations in food products were not found in the literature.
The studies summarized in Tables B-11 and B-12 primarily examined PCDD and
PCDF levels in products of animal origin (i.e., fish, meat, eggs, and dairy products).
Because of their lipophilic nature, PCDDs and PCDFs are expected to accumulate in these
food groups. The data in the tables indicate that PCDDs and PCDFs are found at levels
ranging from the intermediate ppq up to the low ppt range. As expected, the highest
levels reported are those measured in fat. The highest reported congener concentrations
are for the CI7 and CI8 CDDs. In general, for the less-chlorinated homologue groups (i.e.,
CI4 - CI6), the CDF levels measured were larger than the CDD levels but were still within
an order of magnitude. The situation is reversed for the CI7 and CI8 homologue groups.
Few data have been reported on levels of PCDDs and PCDFs in foods of vegetative
origin. Beck et al. (1989) found no PCDDs or PCDFs in samples of vegetable oil,
cauliflower, lettuce, cherries, and apples bought in stores in Berlin. The detection limit
was approximately 0.01 ppt for each isomer on a whole-weight basis. Analyses of three
food baskets collected from stores in Stockholm showed PCDD and PCDF levels close to
or below detection limits (0.1 to 0.5 pg/g detection limits) (de Wit et al., 1990). The only
congeners found were 2,3,7,8-TCDF (0.1 to 0.4 pg/g) and CI8 CDD (1.0 to 1.3 pg/g).
3-18 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
It is generally assumed that consumption of food containing trace levels of PCDDs
and PCDFs is responsible for the majority of human exposure to these compounds. Travis
and Hattemer-Frey (1991), using a fugacity model, estimated that the food chain,
especially meat and dairy products, accounts for 99 percent of human exposure to
2,3,7,8-TCDD. Despite these assumptions and conclusions, few data on actual levels of
PCDDs, PCDFs, and dioxin-like compounds in food products have been published. Those
data that have been published have generally resulted from studies of a specific food
product(s) in a specific location(s) rather than from survey studies designed to allow
estimation of daily intake of the chemicals for a population. The detection limits used in
large survey studies, such as the FDA foodbasket survey, have generally been too high to
detect the low levels of these chemicals likely to be present. Analysis of trace levels of
these chemicals has in the past been hindered by lack of sensitive analytical detection
methods, extraction difficulties from the high-lipid content food products in which these
chemicals are most often found, and the presence of potentially interfering other
organochlorine compounds. As these difficulties have been and continue to be overcome,
the results of more narrow scope studies, as well as broad scope surveys, are likely to be
reported. For example, relatively extensive multiyear surveys of the levels of PCDDs and
PCDFs in food are being undertaken in the United Kingdom and Sweden (de Wit et al.,
1990 and Startin et al., 1990).
3.7. CONCENTRATIONS IN AIR
This section summarizes measurements of dioxin-like congeners in ambient air. The
purpose of sampling and monitoring the ambient air was to provide an approximate
indication of the background concentration of PCDD and PCDF compounds in urban
environs. Analysis of ambient air is extremely difficult because of the low analytical
detection limits required to detect specific congeners. Minimum limits of analytical
detection are in the range of 0.05 to 0.4 picograms of the specific congener per cubic
meter of air. These limits of detection in ambient air samples were not achieved until the
mid 1980's. In general, an air sampling apparatus consisting of a quartz-fiber filter and
polyurethane foam (PUF) is used for collection of samples. The quartz-fiber filter collects
airborne particulate, and the PUF collects vapors. The quartz-fiber filter is usually spiked
with a carbon-labelled isotope of the specific congener prior to operation to determine the
3-19 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
collection and retention efficiency for the PCDDs and PCDFS. To obtain
subparts-per-trillion levels of analytical detection requires sampling relatively large volumes
of air over a 24-hour period, e.g., 350 to 450 cubic meters of ambient air.
Tables B-13 and B-14 (Appendix B) provide a general overview of the limited
ambient air measurements that have been made in selected cities in the U.S. and Europe.
It is interesting to note that there appears to be good agreement with respect to the
magnitude of specific congeners of PCDDs and PCDFs in urbanized areas in the U.S. and
Europe, and specific dioxin-like congeners are quantified from 1/10 to 1/100 picograms per
cubic meter of air sample. Most of these measurements tend to be very close to the
current analytical detection limit. This increases the probability that congeners indicated in
Table 3-13 as Not Detected (ND) may actually be present, but will not be observed until
further advances are made in analytical methods. The value in parentheses after the
designation ND on the Table was the reported detection limit (pg/M3). The designation NA
in Table 3-1 3 means the investigator did not analyze that specific congener in the sample.
3.8 HISTORICAL TRENDS
Small amounts of dioxin-like compounds may be formed during natural fires
suggesting that these compounds may have always been present in the environment.
However, it is generally believed that much more of these compounds have been produced
in association with industrial processes and therefore environmental levels are likely to be
higher in modern times.
Rappe (1991) reports testing of archived soils and plants collected in southeast
England between 1846 and 1986. PCDDs and PCDFs were found in all samples and
showed generally increasing levels of dioxins. Rappe further notes that the congener
pattern is typical of those for combustion sources until about 1950 when the pattern
becomes more dominated by hepta- and octa-CDDs corresponding to increases in
production of chlorinated compounds.
Schecter (1991) analyzed ancient liver tissues (estimated to be 100 to 400 years
old) recovered from frozen bodies of Native American (Eskimo) women. He found that the
dioxin levels were much lower than those commonly found in livers of people currently
living in industrial areas.
3-20 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Travis and Hattemer-Frey (1991) used the Fugacity Food Chain (FFC) model to
predict the contribution of municipal solid waste incinerators, motor vehicles, hospital
waste incinerators, residential wood burning, and pulp and paper mill effluents, to the U.S.
total environmental input of 2,3,7,8-TCDD. It was estimated that the total input from all
five sources combined accounted for only 11 percent of the total TCDD found in the
different media in the U.S. The authors concluded that this low value indicated: (1) that
the source term used in the FFC modeling exercise for 2,3,7,8-TCDD may have been too
high; (2) some unidentified major source(s) of 2,3,7,8-TCDD exist; or (3) there are multiple
environmental sources of TCDD, and no one source dominates the total input.
Rappe (1991) compared the known emission sources of PCDD and PCDF inputs
from Sweden with an estimated aerial deposition rate into the Swedish environment. The
aerial deposition rate was found to be 10-20 times higher than the total emissions sources
originating from Sweden. In an earlier publication, Rappe (1987) characterized sources of
PCDD and PCDF sources as well as environmental and human samples. He concluded that
there was a poor correlation between the isomeric distribution found in emissions from
known sources and human and environmental samples. He speculated that the pattern in
human and environmental samples was the result of a combination of sources, coupled
with environmental and biological degradation.
3.9 ASSESSMENT OF BACKGROUND EXPOSURES
The occurrence data were analyzed to evaluate background levels and differences
between contaminant levels in Europe and North America.
3.9.1. Procedures for Estimating Background Exposures
The term: "background" as applied to exposure can be confusing since it can be
used to represent very different concepts and is commonly not defined by the users. The
two most common uses are 1) the level of exposure that would occur in an area without
known sources of the contaminant of concern or 2) the average level of exposure
occurring in an area whether sources are present or not. For purposes of this document
we will define background as suggested in the first definition above and use "average"
(with a specification of the area of concern) to describe the second concept. The data
collected in rural areas far from known contaminant sources are probably the most useful
3-21 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
for describing background since the fewest sources of dioxin-like compounds are likeiy to
be present.
Using the above definition, background data was considered to be those collected
in areas defined as rural, pristine or residential unless known sources were also reported in
the vicinity. Tables B-15 through B-27 in Appendix B show the geometric and arithmetic
averages of these background data for each media. The geometric averages are
consistently lower than the arithmetic levels. This results from the fact that the data
spans several orders of magnitude with the distribution skewed toward the lower end (due
to the large number of nondetects). If the data were collected in a truly random fashion,
and human contact occurred randomly, then an arithmetic average would be more
appropriate for estimating average exposures. However, most of these data cannot be
considered random and since the geometric average should provide a better indication of
the central tendency (due to the skewed distribution), it was selected as the best way to
estimate a level representative of typical exposure levels.
3.9.2. Geographic Comparisons
Another issue associated with background levels of the dioxins, concerns how the
levels differ between various geographic areas, particularly North America and Europe. As
shown in Table 3-1, the available data were divided into these areas and averaged.
Considering the similar levels of industrialization and lifestyle between Europe and North
America, one may expect similar environmental levels of the dioxin-like compounds.
However, the geographic averages shown on Table 3-1, indicate some discrepancies. It
should be emphasized that it is not really known how representative these averages are of
such large areas. The North America data were collected from approximately 20 different
sites (all media) and the European data were collected from about 18 sites (all media), but
almost all were located in the United Kingdom. Thus, the significance of these
discrepancies is largely unknown due to the relatively sparse database and uncertainties
regarding representativeness.
The exposures associated with these levels can be estimated by multiplying the
levels by typical contact rates. Table 3-2 shows the results of this analysis. Bearing in
mind the limitations of the occurrence database noted above, some observations from this
table include:
3-22 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 3-1. Regional Comparisons of Toxic Equivalent Concentrations
Media
Soil, ppt: CDD
(TEq)
CDF (TEq)
Total CDD + CDF (TEq)
Sediment, ppt: CDD (TEq)
CDF (TEq)
Total CDD + CDF (TEq)
PCB (ng/kg)2
Fish, ppt: CDD
(TEq)
CDF (TEq)
Total CDD + CDF (TEq)
Air, pg/m3: CDD
(TEq)
CDF (TEq)
Total CDD + CDF (TEq)
Water, ppq: CDD
(TEq)
CDF (TEq)
Total CDD + CDF (TEq)
Dairy, ppt3: CDD
(TEq)
CDF (TEq)
Total CDD + CDF (TEq)
Beef, ppt3: CDD (TEq)
CDF (TEq)
Total CDD + CDF (TEq)
Worldwide1
4.12
2.91
7.03
16.2
1.18
17.4
7.42-14,881
0.90
0.46
1.36
0.018
0.031
0.049
0.86
7.40
8.26
0.0427
0.0568
0.0995
0.18
0.19
0.37
North America1
5.24
2.45
7.69
3.24
0.03
3.27
250-14,881
0.94
0.43
1.37
0.019
0.032
0.051
0.86
7.40
8.26
NDA
NDA
NDA
NDA
NDA
NDA
Europe1
0.64
2.93
3.57
25.0
3.13
28.1
6.1-250
0.19
0.87
1.06
0.0059
0.012
0.018
NDA
NDA
NDA
0.0427
0.0568
0.0995
0.18
0.19
0.37
Footnotes
NDA = No data available.
1 Values are the geometric mean TEqs from Tables B-15 to B-27.
2 Values are the ranges of the geometric means across all PCBs
from Table B-20.
3 Values calculated for beef and dairy only from Beck et al. (1989), Furst et al. (1990) and Schecter et al.
(1990), and expressed in table on whole product basis - fat basis given by authors converted to whole product
basis by multiplication by fat content: beef fat by 0.22, and dairy fat by 0.035.
3-23
7/31/92
-------
Table 3-2. Background Exposures
Media
Soil ingestion
Fish ingestion
Inhalation
Water ingestion
Dairy ingestion
Beef ingestion
Worldwide
Concentra-
tion
TEq1
7.03x
TO'12
1.36x
io-9
4.9 x
ID'11
8.26 x
io-9
9.95x
10-"
3.70 x
10-10
Contact
rate2
100
mg/day
6.5 g/day
23
m3/day
1 .4 L/day
300
g/day
100
g/day
Total
Daily
intake
mg/day
7.03 x
10-10
8.84x
io-9
1.13x
io-9
1.16x
10'8
2.98 x
10'8
3.70 x
io-8
8.91 x
io-8
%
of
total
0.79
9.92
1.27
13.0
33.5
41.5
North America
Concentra-
tion
TEq1
7.69 x
io-12
1.37 x
io-9
5.1 x
io-11
8.26x
10'9
NDA
NDA
Contact
rate2
100
mg/day
6.5 g/day
23
m3/day
1 .4 L/day
300
g/day
100
g/day
Total
Daily
intake
mg/day
7.69x
io-10
8.91 x
io-9
1.17 x
io-9
1.16x
io-8
NA
NA
2.24x
1C'8
%
Of
total
3.43
39.7
5.21
51.7
NA
NA
Europe
Concentra-
tion
TEq1
3.57 x
10'12
1.06x
io-9
1.8x
io-11
NDA
9.95x
10-"
3.70 x
10-io
Contact
rate2
100
mg/day
6.5 g/day
23
m3/day
1 .4 L/day
300
g/day
100
g/day
Total
Daily
intake
mg/day
3.57 x
10-10
6.89 x
10'9
4.14 x
io-10
NA
2.98 x
io-8
2.94x
io-8
7.45x
io-8
%
of
total
0.48
9.25
0.57
NA
40.0
49.7
D
33
D
O
D
C
O
H
m
O
33
O
H
m
to
ro
Footnotes: NA = Not applicable, NDA = No Data Available.
1 Values from Table 3-1, and units were converted for calculation of daily intake rate.
2 Values from Exposure Factors Handbook (USEPA, 1989)
to
(D
ro
-------
DRAFT-DO NOT QUOTE OR CITE
• Exposure via beef or dairy ingestion in North America could not be estimated due
to lack of occurrence data. Since these pathways accounts for about two thirds of the
total exposure in Europe, the North America exposure estimates may be low by this
amount.
• Exposure via fish ingestion appears much higher in North America than Europe.
• The total background exposures to TEq shown in Table 3-2 were 74, 89, and 22
pg/day for Europe, Worldwide, and North America, respectively. Because of the database
weaknesses noted above, these discrepancies cannot be considered be significant. It is
probably more appropriate to conclude that permutations of the data tested in this exercise
indicates a range, perhaps, of 20-90 pg/day TEq exposure. Other qualifying considerations
include: 1) the addition of the pathways presumes that individuals are exposed by all
pathways, 2) on the other hand, several exposures such as those from other meat
products, are not considered, and 3) the fraction absorbed into the body is assumed to be
the same for all pathways. Thus, these estimates must be considered highly uncertain.
Similar analyses have been done by other investigators. Furst et al. (1991) estimated
average exposures to the dioxin-like compounds as 70 to 200 pg of TEq/day. Travis and
Hattemer-Frey (1991), estimated typical exposures to 2,3,7,8-TCDD only may be in the
range of 35 pg/day.
• This exercise, and those undertaken by others, suggest that over 90% of the
exposure is associated with ingestion of fish, beef and dairy products, with the remainder
due to air inhalation, soil ingestion, and water ingestion.
• The Worldwide exposure estimate of 89 pg/day was examined further to
determine how much of the exposure was due to 2,3,7,8-TCDD and how much was due
to other congeners. It was found that 14 pg/day was due to 2,3,7,8-TCDD, which is 16
percent of the total exposure. Again, this is an uncertain result. It was driven by the
result for fish, dairy and beef ingestion, which was 13 pg/day 2,3,7,8-TCDD. Fish, dairy
and beef ingestion contributed approximately equal amounts to the 2,3,7,8-TCDD
exposure. Examination of the fish data from which this estimate is derived (Table B-22 in
Appendix B) shows that there were very few congener specific analyses for fish except for
2,3,7,8-TCDD and 2,3,7,8-TCDF.
Examination of body burden data may provide a more accurate way to estimate
typical exposures. However, these data may not represent true background levels as
3-25 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
defined here since the sampled individuals may have lived in urban or residential areas
where dioxins sources were present. Levels of these compounds found in human
tissue/blood appear similar in Europe and North America. Schecter (1991) compared levels
of dioxin-like compounds found in blood among people from U.S. and Germany. Although
mean levels of individual compounds differed by as much as a factor of two, the total
PCDD/Fs averaged 42 ppt of TEq in Germany and 41 ppt of TEq in the U.S. Schecter
(1991) also reports tissue levels in various countries:
Country Mean Tissue Level (ppt of TEq)
USA 24
Germany 69
China 18
Japan 38
Canada 36
S. Vietnam 30
N. Vietnam 4
The tissue data shows more variation between countries but also involved much fewer
samples, reducing confidence in the accuracy of the mean. In Chapter 8, the level of
2,3,7,8-TCDD found in human adipose tissue is reported to average about 5.0 to 6.7 ppt
in the U.S. based on data from a variety of studies. This data was used to estimate the
associated exposure levels using a simple pharmacokinetic model which back calculated
the dose needed to achieve these tissue levels under the assumption of steady state. This
model requires an estimate of the elimination rate constant. Based on available data this
was assumed to be about 5.8 to 7 years which yielded a background dose rate of about
10 to 31 pg/day. This estimate agrees very well with the background exposure estimates
(to 2,3,7,8-TCDD only) of 35 pg/day by Travis and Hattemer-Frey (1991) and 25 pg/day
by Furst et al. (1991), both derived using typical media levels and contact rates. Further
discussion on body burden data and associated exposures is presented in Chapter 8.
3-26 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
REFERENCES FOR CHAPTER 3
Beck, H.; Eckart, K.; Kellert, M.; Mathar, W.; Ruhl, C.S.; Wittkowski, R. (1987) Levels of
PCDFs and PCDDs in samples of human origin and food in the Federal Republic of
Germany. Chemosphere 16 (8/9): 1977-1982.
Beck, H.; Eckart, K.; Mathar, W.; Wittkowski, R. (1989) PCDD and PCDF body burden
from food intake in the Federal Republic of Germany. Chemosphere 18 (1-6):
417-424.
Birmingham, B. (1990) Analysis of PCDD and PCDF patterns in soil samples: use in the
estimation of the risk of exposure. Chemosphere 20(7-9): 807-814.
Bopp, R.F.; Gross, M.L.; Tong, H.; Simpson, M.J.; Monson, S.J.; Deck, B.L.; Moser, F.C.
(1991) A major incident of dioxin contamination: sediments of New Jersey
estuaries. Environ. Sci. Technol. 25: 951-956.
Branson, D.R.; Takahashi, I.T.; Parker, W.M.; Blau, G.E. (1985) Bioconcentration kinetics
of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rainbow trout. Environ. Toxicol. Chem.
4(6):779-788.
Broman, D.; Naf, C.; Rolff, C.; Zebuhr, Y. (1990) Analysis of polychlorinated dibenzo-p-
dioxins (PCDD) and polychlorinated dibenzofurans (PCDF) in soil and digested
sewage sludge from Stockholm, Sweden. Chemosphere 21: 1213-1220.
CDEP. (1988) Measurement of selected polychlorinated dibenzo-p-dioxins and
polychlorinated dibenzofurans in ambient air in the vicinity of Wallingford,
Connecticut. Connecticut Department of Environmental Protection, Air Compliance
Unit, Hartford, CT. Project report by ERT, Concord, MA., Project # 7265-001-004.
July 8,1988.
Creaser, C.S.; Fernandes, A.R.; AI-Haddad,A.; Harrad, S.J.; Homer, R.B.; Skett, P.W.;
Cox, E.A. (1989) Survey of background levels of PCDDs and PCDFs in UK soils.
Chemosphere 18(1-6): 767-776.
Creaser, C.S.; Fernandes, A.R.; Harrad, S.J.; Cox, E.A. (1990) Levels and sources of
PCDDs and PCDFs in urban British soils. Chemosphere 21: 931-938.
Czuczwa, J.M.; McVeety, B.D.; Hites, R.A. (1984) Polychlorinated dibenzo-p-dioxins and
dibenzofurans in sediments from Siskiwit Lake, Isle Royale. Science 226: 568-569.
DeVault, D.; Dunn, W.; Bergquist, P.A.; Wiberg, K.; Rappe, C. (1989) Polychlorinated
dibenzofurans and polychlorinated dibenzo-p-dioxins in Great Lakes fish: a baseline
and interlake comparison. Environ. Toxicol. Chem. 8: 1013-1022.
deWit et al. (1990) Results from the first year of the Swedish dioxin survey.
Chemosphere 2 (10-12): 1473-1480.
3-27 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Edgerton, S.A.; Czuczwa, J.M.; Rench, J.D.; Hodanbosi, R.F.; Koval, P.J. (1989)
Ambient air concentrations of polychlorinated dibenzo-p-dioxins and dibenzofurans
in Ohio: sources and health risk assessment. Chemosphere 18(9/10):1713-1730.
Frommberger, R. (1991) Polychlorinated dibenzo-p-dioxins and polychlorinated
dibenzofurans in fish from south-west Germany: river Rhine and Neckar.
Chemosphere 22(1-2): 29-38.
Furst, P.; Furst, C.; Groebel, W. (1990) Levels of PCDDs and PCDFs in food-stuffs from
the Federal Republic of Germany. Chemosphere 20 (7-9): 787-792.
Gardner, A.M.; White, K.D. (1990) Polychlorinated dibenzofurans in the edible portion of
selected fish. Chemosphere 21(1-2): 215-222.
Gotz, R.; Schumacher, E. (1990) Polychlorierte dibenzo-p-dioxine (PCDDs) und
polychlorierte dibenzofurane (PCDFs) in sedimenten und fischen Aus Dem
Hamburger Hafen. Chemosphere 20(1-2): 51-73.
Harless, R.L.; Lewis, R.G. (1982) Quantitative determination of 2,3,7,8-
tetrachlorodibenzo-p-dioxin residues by gas chromatography/mass spectrometry.
In: Hutzinger, 0.; Frei, R.W.; Merian, E.; Pocchiari, F., eds. Chlorinated dioxins and
related compounds: impact on the environment, pergamon series on environmental
science. Vol. 5. Oxford: Pergamon Press, pp. 25-35.
Harless, R.L.; McDaniel, D.D.; Dupuy, A.E. (1990) Sampling and analysis for
polychlorinated dibenzo-p-dioxins and dibenzofurans in ambient air. Proceedings of
the Tenth International Symposium on Chlorinated Dioxins and Related Compounds,
Bayreuth, Germany, September 10-14, 1990.
Harless, R.L.; Lewis, R.G. (1991) Evaluation of a sampling and analysis method for
determination of polyhalogenated dibenzo-p-dioxins and dibenzofurans in ambient
air. Presented at the llth International Symposium on Chlorinated Dioxins and
Related Compounds, Research Triangle Park, NC, September 23-27, 1991.
Huckins, J.N.; Schwartz, T.R.; Petty, J.D.; Smith, L.M. (1988) Determination, fate, and
potential significance of PCBs in fish and sediment samples with emphasis on
selected AHH-lnducing congeners. Chemosphere 17(10): 1995-2016.
Hunt, G.T.; Maisel, B. (1990) Atmospheric PCDDs/PCDFs in wintertime in a
northeastern U.S. urban coastal environment. Chemosphere 20: 14551 462.
Jobb, B.; Uza, M.; Hunsinger, R.; Roberts, K.; Tosine, H.; Clement, R.; Bobbie, B.; LeBel,
G.; Williams, D.; Lau, B. (1990) A survey of drinking water supplies in the
province of Ontario for dioxins and furans. Chemosphere 20(10-12): 1553-1558.
Kitunen, V.H.; Salkinoja-Salonen, M.S. (1990) Soil contamination at abandoned sawmill
areas. Chemosphere 20: 1671-1677.
3-28 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Kjeller, L.O.; Kulp, S.E.; Bergetk, S.; Bostrom, M.; Bergquist, P.A.: Rappe,
C. ; Jonsson, B.; de Wit, C.; Jansson, B.; Olsson, M. (1990) Levels and possible
sources of PCDD/PCDF in sediment and pike samples from Swedish lakes and
rivers. (Part one). Chemosphere 20 (10-12): 1489-1496.
Koistinen, J.; Passivirta, J.; Sarkka, J. (1990) Organic chlorine compounds in Lake
Sediments. IV. Dioxins, furans and related chloroaromatic compounds.
Chemosphere 21(12): 1371-1379.
Maisel, G.E.; Hunt, G.T. (1990) Background concentrations of PCDDs/PCDFs in
ambient air. A comparison of toxic equivalency factor (TEF) models. Chemosphere
20: 771-778.
McKee, P.; Burt, A.; McCurvin, D.; Hollinger, D.; Clement, R.; Sutherland, D.; Neaves, W.
(1990) Levels of dioxins, furans and other organic contaminants in harbour
sediments near a wood preserving plant using pentachlorophenol and creosote.
Chemosphere 20(10-12): 1679-1685.
Meyer, C.; O'Keefe, D.; Hilker, D.; Rafferty, L.; Wilson, L.; Connor, S.; Aldous, K.;
Markussen, K.; Slade, K. (1989) A survey of twenty community water systems in
New York State for PCDDs and PCDFs. Chemosphere 19(1-6): 21-26.
Naf, C; Broman, D.; Ishaq, R.; Zebuhr, Y. (1990) PCDDs and PCDFs in water, sludge and
air samples from various levels in a waste water treatment plant with respect to
composition changes and total flux. Chemosphere 20: 1503-1510.
Nestrick, T.J.; Lamparski, L.L.; Frawley, N.N.; Hummel, R.A.; Kocher, C.W.; Mahle, N.H.;
McCoy, J.W.; Miller, D.L.; Peters, T.L.; Pillepich, J.L.; Smith, W.E.; Tobey, S.W.
(1986) Perspectives of a large scale environmental survey for chlorinated dioxins:
overview and soil data. Chemosphere 15: 1453-1460.
Niimi, A.J.; Oliver, B.C. (1989a) Assessment of relative toxicity of chlorinated dibenzo-p-
dioxins, dibenzofurans, and biphenyls in Lake Ontario salmonids to mammalian
systems using toxic equivalent factors (TEF). Chemosphere 18(7-8): 1413-1423.
Niimi, A.J.; Oliver, B.C. (1989b) Distribution of polychlorinated biphenyl congeners and
other halocarbons in whole fish and muscle among Lake Ontario salmonids.
Environ. Sci. Technol. 23: 83-88.
Norwood, C.B.; Hackett, M.; Powell, R.J.; Butterworth, B.C.; Williamson, K.J.; Naumann,
S.M. (1989) Polychlorinated dibenzo-p-dioxins and dibenzofurans in selected
estuarine sediments. Chemosphere 18(1-6): 553-560.
Oehme, M.; Mane, S.; Brevik, E.M.; Knutzen, J. (1989) Determination of polychlorinated
dibenzofuran (PCDF) and dibenzo-p-dioxin (PCDD) levels and isomer patterns in fish,
Crustacea, mussel and sediment samples from a fjord region polluted by Mg-
production. Fresenius Z. Anal. Chem. 335: 987-997.
3-29 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Oliver, B.C.; Nilmi, A.J. (1988) Trophodynamic analysis of polychlorinated biphenyl
congeners and other chlorinated hydrocarbons in the Lake Ontario ecosystem.
Environ. Sci. Technol. 22: 388-397.
Pearson, R.G.; Mclaughlin, D.L.; Mcllveen, W.D. (1990) Concentrations of PCDD arid
PCDF in Ontario soils from the vicinity of refuse and sewage sludge incinerators and
remote rural and urban locations. Chemosphere 20: 1543-1548.
Petty, J.D.; Smith, C.M.; Bergquist, P.A.; Johnson, J.L.; Stalling, D.L.; Rappe, C. (1982)
Chlorinated dioxins, dibenzofurans total environment, [Proc. Symp.], meeting date
1982, 203-7. Edited by Choudhary, Keith, Rappe, and Butterworth. Boston,
Massachusetts.
Rappe, C. (1987) Global distribution of polychlorinated dioxins and dibenzofurans, in
Exner, J.H., ed.. Solving Hazardous Waste Problems Learning from Dioxins. ACS
Symposium Series 338. developed from a symposium sponsored by the Division
of Environmental Chemistry at the 191st meeting the American Chemical Society,
New York, New York, April 13-18, 1986.
Rappe, C. (1991) Sources of human exposure to PCDDs and PCDFs. In: Biological Basis
for Risk Assessment of Dioxins and Related Compounds. Banbury Report #35.
Edited by M. Gallo, R. Scheuplein and K. Van der Heijden. Cold Spring Harbor
Laboratory Press. Plainview, NY.
Rappe, C.; Nygren M.; Lindstrom G. (1987) Polychlorinated dibenzofurans and dibenzo-p-
dioxins and other chlorinated contaminants in cow milk from various locations in
Switzerland. Environ. Sci. Technol. 21 (10): 964-970.
Rappe, C.; Kjeller, L.O.; Andersson, R. (1989b) Analysis of PCDDs and PCDFs in sludge
and water samples. Chemosphere 19(1-6): 13-20.
Rappe, C.; Kjeller, L.O. (1987) PCDDs and PCDFs in environmental samples, air,
particulates, sediments, and soil. Chemosphere 16(8/9): 1775-1780.
Rappe, C.; Bergqvist, P.A.; Kjeller, L.A. (1989a) Levels, trends and patterns of PCDDs
and PCDFs in Scandinavian environmental samples. Chemosphere 18(1-6): 651-
658.
Rappe, C.; Bergquist, P.A.; Kjeller, L.O.; Swanson, S.; Belton, T.; Ruppel, B.; Lockwood,
K.; Kahn, P.C. (1991) Levels and patterns of PCDD and PCDF contamination in
fish, crabs, and lobsters from Newark Bay and the New York Bight. Chemosphere
22(3-4): 239-266.
Rappe, C.; Bergquist, P.A.; Hansson, M.; Kjeller, L.O.; Lindstom, G.; Marklund, S.; Nygren,
M. (1984) Chemistry and analysis of polychlorinated dioxins and dibenzofurans in
biological samples. In: Banbury Report 18: Biological Mechanisms of Dioxin
Action. Cold Spring Harbor Laboratory, pp. 17-25.
3-30 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Reed, L.W.; Hunt, G.T.; Maisel, B.E.; Hoyt, M.; Keefe, D.; Hackney, P. (1990) Baseline
assessment of PCDDs/PCDFs in the vicinity of the Elk River, Minnesota generating
station. Chemosphere 21(1-2): 159-171.
Ryan, J.J.; Lizotte, R.; Sakuma, T.; Mori, B. (1985) Chlorinated dibenzo-p-dioxins,
chlorinated dibenzofurans, and pentachlorophenol in Canadian chicken and pork
samples. J. Agric. Food Chem. 33: 1021-1026.
Schecter, A. (1991) Dioxins and related compounds in humans and in the environment.
In: Biological Basis for Risk Assessment of Dioxins and Related Compounds.
Banbury Report #35. Edited by M. Gallo, R. Scheuplein and K. Van der Heijden.
Cold Spring Harbor Laboratory Press. Plainview, NY.
Schecter et al. (1990) Levels of chlorinated dioxins, dibenzofurans and other chlorinated
xenobiotics in food from the Soviet Union and the South of Vietnam. Chemosphere
20 (7-9): 799-806.
Sherman, R.K.; Clement, R.E.; Tashiro, C. (1990) The distribution of polychlorinated
dibenzo-p-dioxins and dibenzofurans in Jackfish Bay, Lake Superior, in relation to a
Kraft Pulp Mill effluent. Chemosphere 20(10-12): 1641-1648.
Sievers, S.; Friesel, P. (1989) Soil contamination patterns of chlorinated organic
compounds: looking for the source. Chemosphere 19(1-6): 691-698.
Smith, R.M.; O'Keefe, P.W.; Hilker, D.R.; Aldous, K.M.; Mo, S.H.; Stelle, R.M. (1989)
Ambient air and incinerator testing for chlorinated dibenzofurans and dioxins by low
resolution mass spectrometry. Chemosphere 18: 585-592.
Smith, L.M.; Schwartz, T.R.; Feltz, K. (1990) Determination and occurence of AHH-
Active polychlorinated biphenyls, 2,3,7,8-tetrachloro-p-dioxin and 2,3,7,8-
tetrachlorodibenzofuran in Lake Michigan sediment and biota. The question of their
relative toxicological significance. Chemosphere 21(9): 1063-1085.
Sonzongni, W.; Maack, L.; Gibson, T.; Lawrence, J. (1991) Toxic polychlorinated
biphenyl congeners in Sheboygan River (USA) sediments. Bull. Environ. Contam.
Toxicol. 47: 398-405.
Stalling, D.L.; Smith, L.M.; Petty, JU.D.; Hogan, J.W.; Johnson, J.L.; Rappe,
C.; Buser, H.R. (1983) Residues of polychlorinated dibenzo-p-dioxins and
dibenzofurans in laurentian Great Lakes fish. In: Tucher, R.E.; Young, A.L.; Gray,
A.P.; eds. Human and environmental risks of chlorinated dioxins and related
compounds. Arlington, Virginia: Plenum Press, pp. 221-240.
Startin, J.R.; Rose, M.; Wright, C.; Parker, I.; Gilbert, J. (1990) Surveillance of British
foods for PCDDs and PCDFs. Chemosphere 20 (7-9): 793-798.
3-31 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Stenhouse, I.A.; Badsha, K.S. (1990) PCB, PCDD, and PCDF concentrations in soils from
the Kirk Sandall/Edenthorpe/Barnby Dun area. Chemosphere 21: 563-573.
Travis, C.C.; Hattemer-Frey, H.A. (1991) Human exposure to dioxin. Sci. Total Environ.
104: 97-127.
U.S. EPA. (1992) EPA National bioaccumulation study. Personal communication between
John Schaum, U.S. EPA Office of Health and Environmental Assessment and
Richard Healy, U.S. EPA Office of Water. July 22, 1992.
U.S. EPA. (1990) Risk assessment for 2378-TCDD and 2,3,7,8-TCDF contaminated
receiving waters from U.S. chlorine-bleaching pulp and paper mills. U.S. EPA Office
of Water Regulations and Standards. PB #90272873. August 1990.
U.S. EPA. (1987) National dioxin study. Office of Solid Waste and Emergency Response.
EPA/530-SW-87-025. August 1987.
U.S. EPA. (1985) Soil screening survey at four midwestern sites. Region V.
Enviornmental Services Division, Eastern District Office, Westlake, Ohio, EPA-
905/4-805-005, June 1985.
Van den berg, M. (1987) Presence of polychlorinated dibenzo-p-dioxins and
polychlorinated dibenzofurans in fish-eating birds and fish from the Netherlands.
Arch. Environ. Contam. Toxicol. 16: 149-158.
3-32 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
4. ESTIMATING EXPOSURES AND RISKS
4.1. INTRODUCTION
In this chapter, the framework for assessing exposure and risk to 2,3,7,8-TCDD
and related dioxin-like compounds will be described. Section 4.2 introduces the exposure
equation and its key terms. Section 4.3 describes how risk is estimated given estimates
of exposure. It also discusses the use of toxicity equivalency factors. Section 4.4
provides the overview of the procedures used in this document, and provides a roadmap
for finding pertinent information in other chapters of the document.
4.2. EXPOSURE EQUATION
This document describes procedures for conducting exposure assessments to
estimate either potential or internal dose. A potential dose is defined as a daily amount of
contaminant inhaled, ingested, or otherwise coming in contact with outer surfaces of the
body, averaged over an individual's body weight and lifetime. An internal dose is defined
as the amount of the potential dose which is absorbed into the body (EPA, 1991). Section
4.3 below discusses the relevancy of this distinction for dioxin-like compounds.
The general equation used to estimate potential dose normalized over bodyweight
and lifetime is as follows:
Lifetime Average Daily Dose (LADD) = (exposure media concentration x
contact rate x contact fraction x exposure duration ) /
(body weight x lifetime) (4-1)
This procedure is used to estimate dose in the form needed to assess cancer risks. Each
of the terms in this exposure equation is discussed briefly below:
• Exposure media concentrations: These include the concentrations in soil for
dermal contact and soil ingestion exposure pathways, in vapor and
particulate phase in air for inhalation exposure pathways, in water for a
water ingestion pathway, and in food products such as fish, fruits and
vegetables, and beef and milk, for food ingestion pathways. This document
provides procedures for estimating exposures associated with all these
pathways.
4-1 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
• Contact rate: These include the ingestion rates, inhalation rates, and soil
contact rates for the exposure pathways. These quantities are generally the
total amount of food ingested, air inhaled, etc. Only a portion of this
material may be contaminated. The next term, the contact fraction, which is
1.0 or less, reduces the total contact rate to the rate specific to the
contaminated media.
• Contact fraction: As noted, this term describes the distribution of total
contact between contaminated and uncontaminated media. For example, a
contact fraction of 0.8 for inhalation means that 80% of the air inhaled over
the exposure period contains dioxin-like compounds in vapor form or sorbed
to air-borne particulates.
• Exposure duration: This is the overall time period of exposure. Values such
as 9 years and 20 years are used in the example scenarios described in
Chapter 9. The value of 9 years corresponds to the average time spent at
one residence (EPA, 1989), and was used as an exposure duration for a non-
farming family living in a rural setting for the example scenarios in Chapter
9. Twenty years was used as the exposure duration for farming families in a
rural setting. Another exposure duration demonstrated in Chapter 10 is one
associated with a childhood pattern of soil ingestion. The exposure duration
in this case is 5 years.
• Body weight: For all the pathways, the human adult body weight of 70 kg
is assumed. This value represents the United States population average.
The body weight for child soil ingestion is 17 kg (EPA, 1989).
• Lifetime: Following traditional assumptions, the average adult lifetime
assumed throughout this document is 70 years. Even though actuarial data
indicate that the United States average lifetime now exceeds 70 years, this
convention is used to be consistent with other Agency assessments of
exposure and risk.
4.3. RISK EQUATION
Although estimation of risk is technically beyond the scope of an exposure
assessment, the exposure assessor needs some background understanding in this area.
4-2 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
The primary source of information on the health risks of the dioxin-related compounds is
the Health Reassessment that EPA is publishing concurrently with this document.
However some general considerations for using exposure estimates in support of cancer
risk assessments are summarized here. The usual procedure used to calculate an upper-
limit incremental cancer risk is as follows:
R = 1 - e-*d « gi* d (4-2)
when q/d < 10"3 and where q.,* is the 95% upper confidence limit of the linearized
cancer slope factor of the dose-response function (expressed in inverse units of the dose
quantity, typically kg-day/mg) and d is the dose (typically in mg/kg-day). The dose is
generally equal to the potential dose given above in Equation (4-1). The slope factor, q^,
for 2,3,7,8-TCDD has been previously estimated by EPA as 0.156 kg-d/ng. The derivation
of this factor is described in EPA (1984) and further background is provided in EPA
(1981). The Agency is currently reevaluating this slope factor and the reader should
consult the companion Health Reassessment for the current policy.
EPA derived the 1984 slope factor for 2,3,7,8-TCDD from animal feeding studies
on the basis of potential (i.e., administered) dose. Thus, for purposes of consistency,
when using this slope factor to estimate risk to humans, the exposure assessor should
provide the dose estimate as a potential dose. This point raises issues specific to the
various pathways.
The absorption which occurred during the animal experiments EPA used to derive
the 1984 slope factor for 2,3,7,8-TCDD was estimated to be 55% (Farland, 1987). The
review of literature on bioavailability in Chapter 8 of this assessment indicates that the gut
absorption of 2,3,7,8-TCDD in humans when the vehicle is soil is 20-40% of potential
dose. Fries and Marrow (1975) found that 50-60% of the 2,3,7,8-TCDD was absorbed by
rats from feed. Rose, et al. (1976) estimated that 86% of 2,3,7,8-TCDD in a mixture of
acetone and corn oil fed by gavage to rats was absorbed. EPA (1984), using animal data
and information on fate of particles in the respiratory system, estimated that the fraction
of 2,3,7,8-TCDD absorbed into the body ranges from 0.25 to 0.29. What this discussion
indicates is that the absorption for human ingestion and inhalation pathways might range
4-3 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
from 20-80% of potential dose, which compares to 55% found in the laboratory
experiments. If no adjustment were made to potential dose estimates, then human risk
estimates might be overestimated (when absorption is in the 20% range) or
underestimated (in the 80% range). This discrepancy is not felt to be large enough or
certain enough to warrant an absorption correction factor.
The rate of absorption of vapor-phase 2,3,7,8-TCDD into the lungs has not been
studied, but it seems reasonable to assume that the absorption in the vapor phase should
exceed that of absorption from bound 2,3,7,8-TCDD on particulates, probably above 50%.
There is also an unknown uncertainty introduced when assuming that the q1" developed
from a feeding study can be used for an inhalation pathway. Thus, it is unclear what
adjustment is needed to account for differences between the feeding study and a vapor-
phase inhalation exposure. Accordingly, it is recommended that assessors not attempt
any such adjustments at this point, but fully acknowledge the uncertainty.
Estimating risks associated with dermal exposure introduces several issues to
consider. First, use of an oral dose-response function may not be applicable to the dermal
route. Second, dermal absorption of dioxin-like compounds from soil appears to be much
lower than that which occurred in the dose-response feeding studies. EPA (1992)
indicates that 0.1 - 3% of 2,3,7,8-TCDD may be dermally absorbed from soil, which is
significantly less than the 55% absorption found in the laboratory feeding experiments. It
is assumed for this assessment that an absorption fraction of 0.03 (3%) applies to
2,3,7,8-TCDD as well as the other dioxin-like compounds. Specifically, this assessment
estimates the total amount of compounds applied to skin and then reduces it by 97% to
estimate the absorbed dose. This is the only pathway in which an absorption fraction is
used to adjust a dose (see Chapter 7 for more information on the procedures to estimate
exposure from all pathways). Because of this adjustment, an additional adjustment to the
risk equation. Equation (4-2) above, is needed when estimating risk from dermal exposure
in a manner consistent with other exposure pathways: the slope factor should be
multiplied by (100%) / (55%), or about 2, to convert it to an absorbed basis. Finally, the
assessor should acknowledge that considerable uncertainty is introduced by applying an
oral based dose-response function to dermal exposure.
Another set of issues facing the exposure/risk assessor is how to estimate
exposure to mixtures of dioxin-like compounds with differing slope factors. EPA (1989)
4-4 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
has proposed a procedure to address this issue, which is to adjust the risk estimate using a
"toxicity equivalency factor", commonly referred to as TEF. The TEF for a congener of
interest is the cancer potency of that congener divided by the cancer potency of 2,3,7,8-
TCDD. As shown in Table 2-1, the TEF for 2,3,7,8-TCDD is 1 and all other dioxin-like
compounds have TEFs less than 1. The combined risk resulting from exposure to a
mixture of dioxin-like compounds can be computed using the TEFs and assuming that the
risks are additive:
Risk = 1-e'*1 ** ""* ai> (4-3)
where q.,* is the cancer slope factor for 2,3,7,8-TCDD (kg-day/mg), TEFj is the toxicity
equivalency factor for dioxin-like compound i, dj is the potential dose for dioxin-like
compound i (mg/kg-day), and n is the total number of dioxin-like compounds to which an
individual is exposed.
4.4. PROCEDURE FOR ESTIMATING EXPOSURE
Section 4.2 described the exposure equation as it applies to dioxin and dioxin-like
compounds. Before making exposure estimates, the assessor needs to gain a more
complete understanding of the exposure setting and be prepared to estimate exposure
media concentrations. The purpose of this section is to provide guidance for the
procedures followed in this assessment to define such settings and estimate exposure
media concentrations. The approach used here is termed the exposure scenario approach.
Brief descriptions of the steps and associated document chapters are presented below and
summarized in Figure 4-1.
Step 1. Identify Source
Two principal sources are addressed in this document. The first, identified as
"soil", is called a source in that the starting point of the assessment is soil contamination.
Of course, the ultimate source for soil contamination is some unidentified cause for the soil
to become contaminated. For exposure and risk assessment purposes, the cause for
contamination is not relevant except to assume that the cause is not ongoing and that the
impact of the "initial" levels is what is being evaluated. The soil source is further
4-5 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
STEPS
Step 1. Identify Sources
A. Soil
On-site, off-site
B. Incinerators
Stack emissions, fly ash disposal
Step 2. Estimate Release Rates
A. Volatilization, erosion, etc.
B. Stack emission rates
Step 3. Estimate Exposure Point Concentrations
A. Food chain, dilution, etc.
B. Dispersion, deposition, etc.
Step 4. Characterize Exposed Individuals and
Exposure Patterns
A. Contact rates, exposure durations
Step 5. Put It Together in Terms of Exposure
Scenarios
A. Scenario concept expanded
B. Demonstration with scenarios
Step 6. Estimate Exposure and Risk
A. Equations and background
B. Results for example scenarios
Step 7. Assess Uncertainty
DOCUMENT CHAPTERS
Chapter 3
Chapter 6
Chapter 5
Chapter 6
Chapter 5
Chapter 6
Chapter 7
Chapter 7
Chapter 9
Chapter 4
Chapter 9
Chapter 10
Figure 4-1. Roadmap for assessing exposure and risk to dioxin and dioxin-like compounds.
4-6
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
characterized as off-site or on-site. Off-site implies that the soil contamination is located
some distance from the site of exposure. The site of exposure could be a residence or
farm, and the site of contamination could be a landfill, for example. On-site implies that
the soil contamination is on the site of exposure. The second principal source is
"incinerators." Unlike the soil source, the contamination is assumed to be on-going. The
two subcategories under this principal source are stack emissions and fly ash. Stack
emissions in paniculate form are assumed to deposit onto the soil of the site of exposure,
and emissions in vapor form result in air-borne concentrations for inhalation exposures.
For fly ash, the exposed individuals are those who are assumed to reside near a landfill
where the ash is being disposed of in an ongoing process. The procedures for calculating
exposures and risks for those individuals are similar to the off-site soil source category. It
is noted that individuals working at the incinerator or involved in the disposal of fly ash are
also exposed. The procedures in this document only apply to residents who are not
associated with the incinerator or its operations.
Step 2. Estimate Release Rates
Estimating the release of contaminants from the initial source is the first step
towards estimating the concentration in the exposure media. Releases from soil
contamination include volatilization, erosion, transfer to biota, and so on. Chapter 5 on
estimating exposure media concentrations describes fate and transport modeling
procedures for estimating soil releases. Emissions from incinerator stacks are the release
rate pertinent to that source. Releases from off-site soil contamination areas include
suspension of particulates via wind erosion, and soil erosion. Additional releases from a
fly ash disposal areas include fugitive dust emissions from landfill operations. Background
on incinerators including assumptions made for the methodology demonstration in Chapter
9 are provided in Chapter 6.
Step 3. Estimate Exposure Point Concentrations
Contaminants released from soils or emitted from stacks move through the
environment to points where human exposure may occur. Contaminated soil that is near
but not at the site of exposure is assumed to slowly erode and contaminate the exposure
site soil, but to a level lower than the initial source strength. The only time when the
4-7 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
source concentrations equal the exposure concentrations is for the soil pathways, soil
ingestion and dermal contact, when the soil contamination is on-site. Chapter 6 describes
the use of the Industrial Source Complex (ISC) Model used to estimate dispersion of stack
plumes to arrive at air-borne concentrations at the site of exposure as well as deposition
rates of stack emitted particulates onto exposure site soil. Chapter 5 describes how soil
concentrations are estimated given particulate deposition rates, and also how release rates
from soil initially contaminated translate to exposure point concentrations.
Step 4. Characterize Exposed Individuals and Exposure Patterns
The patterns of exposure are described in Chapter 7. Exposed individuals in the
scenarios of this assessment are individuals who are exposed in their home environments.
They are adult residents who also recreationally fish, have a home garden, farm, and are
children ages 2-6 for the soil ingestion pathway. Their exposure is characterized in terms
of exposure pathways, which have generally been alluded to in discussions above -
inhalation, ingestion, and soil dermal contact. Each pathway has the set of parameters
including contact rates, contact fractions, body weights, and lifetime. These parameters
were defined earlier in Section 4.4.
Step 5. Put It Together in Terms of Exposure Scenarios
A common framework for assessing exposure is with the use of "settings" and
"scenarios." Settings are the physical aspects of an exposure area and the scenario
characterizes the behavior of the population in the setting and determines the severity of
the exposure. A wide range of exposures are possible depending on behavior pattern
assumptions. An exposure scenario framework offers the opportunity to vary any number
of assumptions and parameters to demonstrate the impact of changes to exposure and risk
estimates. Exposure estimates for six example scenarios are demonstrated in Chapter 9.
Step 6. Estimate Exposure and Risk
Section 4.2 described the basic equation that estimates exposure for every
assumed pathway in an exposure scenario. Chapter 9 demonstrates the methodology on
six example scenarios, which includes the generation of exposure estimates for ten
different exposure pathways and three different dioxin-like compounds per scenario.
4-8 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Step 7. Assess Uncertainty
Chapter 10 provides a comprehensive discussion on possible sources of uncertainty
associated with this methodology. These uncertainties should be considered when
applying this procedures to a particular site.
4-9 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
REFERENCES FOR CHAPTER 4
Farland, W.H. (1987) Memorandum titled, "Absorption fraction when calculating upper-
limit risks due to dioxin exposure", dated September 2, 1987, to Michael Callahan,
Exposure Assessment Group, Washington, DC. from William Farland, U.S.
Environmental Protection Agency, Office of Health and Environmental Assessment,
Washington, D.C.
Fries, G.F.; Marrow, G.S. (1975) Retention and excretion of 2,3,7,8-tetrachlorodibenzo-
p-dioxin (TCDD) by rats. J. Agric. Food Chem. 23: 265-269.
Rose, J.Q.; Ramsey, J.C.; Wentzler, T.H. (1976) The fate of 2,3,7,8-tertrachloridbenzo-
p-dioxin following single and repeated oral doses to the rat. Toxicol. Appl.
Pharmacol. 36: 209-226.
U.S. Environmental Protection Agency. (1981) Risk assessment on (2,4,5-
trichlorophenoxy)acetic acid (2,4,5-T), (2,4,5-trichlorophenoxy)propionic acid
(silvex), and the 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Office of Health and
Environmental Assessment, Washington, DC; EPA-600/6-81-003. NTIS PB81-
234825.
U.S. Environmental Protection Agency. (1984) Risk analysis of TCDD contaminated soil.
U.S. Environmental Protection Agency, Office of Health and Environmental
Assessment, Washington, DC; EPA/600/8-84-031.
U.S. Environmental Protection Agency. (1989) Interim procedures for estimating risks
associated with exposures to mixtures of chlorinated dibenzo-p-dioxins and
-dibenzofurans (CDDs and CDFs) and 1989 update. U.S. Environmental Protection
Agency, Risk Assessment Forum, Washington, DC; EPA/625/3-89/016.
U.S. Environmental Protection Agency. (1991) Guidelines for exposure assessment. SAB
Draft Final. August 8, 1991. U.S. Environmental Protection Agency, Office of
Health and Environmental Assessment, Washington, DC; OHEA-E-451.
U.S. Environmental Protection Agency. (1992) Health reassessment of dioxin-like
compounds. U.S. Environmental Protection Agency, Office of Health and
Environmental Assessment, Washington, DC.
U.S. Environmental Protection Agency. (1992) Dermal exposure assessment: principles
and applications. U.S. Environmental Protection Agency, Office of Health and
Environmental Assessment, Washington, DC; EPA\600\8-91\011B.
4-10 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
5. ESTIMATING EXPOSURE MEDIA CONCENTRATIONS
5.1. INTRODUCTION
The purpose of this chapter is to describe the algorithms used to determine
exposure media concentrations of the dioxin-like compounds. Discussion of the algorithms
are structured around four "source categories." These categories roughly translate to
beginning points, or origins, of contamination. The source categories are also the basis for
the example scenarios described in Chapter 9.
Section 5.3 describes the algorithms used for the first source category, on-site soil,
where the contaminants occur in surface soils, and this contamination source and
subsequent exposure occur at the same site. The second source category, described in
Section 5.4, is termed off-site soil. The contaminated soil is remote from the exposure,
such as in a landfill impacting a nearby residence. Section 5.5 describes algorithms to
determine exposure media concentrations resulting from incinerator stack emissions, the
third source category. The Industrial Source Complex (ISC) Model (EPA, 1986) was used
in the example of this source category to generate two key quantities: air-borne
contaminant concentrations at a site of exposure, and particulate deposition rates.
Chapter 6 discusses incineration as a source of dioxin-like compounds and describes this
ISC application. The ISC model is the only complex fate and transport model used in this
document. This exception was made because the ISC model is widely available and
relatively easy to use. Section 5.5 describes how ISC-modeled deposition rates translate
to soil, vegetative, and water concentrations. Section 5.6 concludes the chapter with a
discussion of algorithms specific to the fourth source category, incinerator ash disposal in
a landfill. The contaminated ash is spread onto the surface of an active landfill, so the fate
and transport algorithms associated with this category are assumed to be the same as
those for the second category, off-site soil. Section 5.6 shows how landfill size (or the
size of the landfill portion dedicated to ash disposal) can be estimated given an amount of
ash disposed. Section 5.6 also describes algorithms to estimate fugitive ash emissions
that result from ash handling operations.
Algorithms are presented which estimate exposure media concentrations for: 1)
surface water impacts: sediment and dissolved phase concentrations, 2) air including the
vapor phase and in particulate form, and 3) biota including beef, milk, fruit and vegetables,
5-1 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
and fish. The soil at the site of exposure is the exposure media for the soil ingestion and
dermal exposure pathways. Table 5-1 summarizes the solution algorithms used to
estimate exposure media concentrations for each of the four source categories.
The purpose of this chapter is twofold. First, it provides the fate and transport
algorithms for estimating these exposure media concentrations. Second, it provides
information about all the model parameters and justification for the default values selected
for the demonstration of methodologies in Chapter 9. The four source categories outlined
in this chapter are demonstrated in Chapter 9 in what are termed exposure scenarios. An
exposure scenario includes a physical description of an area where exposure occurs,
including features such as the proximity to the soil contamination or incinerator, the
presence of nearby water bodies used for fishing or drinking, and so on. Exposure
scenario definition also includes assumptions on the behavior of individuals and the
pathways of exposures. One set of scenarios was crafted to represent "central"
exposures, and the other set, "high end" exposures. The central exposure scenario is
modeled as a residence in a rural setting, and the high end a farm in the same rural setting.
One way a residence is central in contrast to a farm is that residents are assumed to reside
a shorter time in the residence, 9 years, in contrast to the family residing at the farm, 20
years. For purposes of exposure estimation, residence time is equivalent to exposure
duration.
The introductory portion of Chapter 9 provides a complete description of the
example exposure scenarios, and other chapters in this assessment discuss issues on
exposure estimation. This introductory portion is offered only so that references below to
the example scenarios in Chapter 9 are understood.
It has been noted in earlier chapters that the assessment procedures described in
this document are site specific. This is perhaps most true for the algorithms estimating
exposure media concentrations. The values selected to demonstrate the methodologies
are not being promoted as general default values. Brief discussions and references are
provided for all model parameters so that users of the methodology can make alternate
parameter selections to better suit their purposes.
5-2 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 5-1. Summary and fate and transport solution algorithms for estimating exposure
media concentrations for the four source categories.
Fate and transport
Source category solution algorithms Exposure media
On-site soil a) Suspended and bottom sediment Surface water
concentrations estimated as a function
of soil erosion from contaminated site
and from watershed; water concentrations
estimated by partitioning from suspended
sediment.
b) Surface water bottom sediment concentrations Fish
(as estimated above) multiplied by a
Biota to Sediment Accumulation Factor, or BSAF
c) No dissipation of initial soil con- Soil
centrations
d) Volatilization assuming vapor phase Air: vapor
diffusion; near-field dispersion
estimates air concentration
e) Wind erosion estimates emissions of Air: particulate
contaminates associated with suspended
particulates; same near-field dispersion
model as used for vapor-phase diffusion
f) Below ground vegetation assume soil to Fruits and
plant transfer; above ground vegetation vegetables
assumes air-to-plant vapor phase
transfer and particulate deposition
g) Biotransfer to cattle meat and milk as a Meat and
linear function of modeled concentrations milk
in total dry matter intake including pasture
grass, fodder, and soil ingested by cattle;
different assumptions on proportion of
noted dry matter intakes made for cattle
raised for beef vs. raised for dairy;
pasture grass and fodder estimated using
above ground vegetation algorithms as
noted in "f)" above
(continued on the following page)
5-3 7/31/92
-------
Table 5-1. (continued)
DRAFT-DO NOT QUOTE OR CITE
Source category
Fate and transport
solution algorithms
Exposure media
Off-site soil
a) Same algorithm as on-site source category
b) Same algorithm as on-site source category
c) Off-site soil concentrations remain
constant; erosion onto exposure site
(residence or farm, e.g.) delivers soil-
borne contaminants; steady-state
contaminant concentrations on exposure
site soil estimated using Universal
Soil Loss Equation for erosion estimates
and assuming further dissipation of
contaminants; exposure site soil concen-
trations further reduced for tillage-
related activities of vegetable gardening
and crop growing.
d) Volatilization assuming vapor phase
diffusion as in on-site source category;
far-field dispersion model replaces
near-field dispersion model of on-site
source category to estimate exposure site
vapor phase concentrations
e) Particulates suspended via wind erosion
as in on-site source category; far-field
dispersion model estimates exposure site
air-borne paniculate phase concentrations
f) Same algorithm as on-site soil source
category using modeled media concentra-
tion; one difference is that exposure
site soil concentrations further diluted
by tillage for below ground vegetables
g) Same algorithm as on-site source
category; air concentrations for vapor
phase transfers and particulates for
deposition (for grass and fodder) now
originate at off-site contamination
Surface water
Fish
Soil
Air: vapor
Air: paniculate
Fruits and
vegetables
Meat and milk
(continued on the following page)
5-4
7/31/92
-------
Table 5-1. (continued)
DRAFT-DO NOT QUOTE OR CITE
Source category
Fate and transport
solution algorithms
Exposure media
Stack emission
Ash disposal in
a landfill
Surface water
Fish
Soil
a) Watershed average steady state soil
concentration estimated using ISC modeled
deposition rates and residue dissipation
as in algorithm c) below; suspended and
bottom sediment concentrations assumed equal
to watershed soil concentrations; surface
water concentrations then estimated
assuming partitioning as in other source
categories.
b) Same algorithm as on-site source category
c) ISC model estimates particulate deposition
rates; steady-state concentrations in
soil estimated assuming further dissipation
of deposited particles
d) ISC model estimates exposure site vapor-
phase concentrations
e) No suspended particulates assumed to
result from stack emissions; wind erosion
assumed not to occur as in other source
categories
f) Same basic algorithm as in on-site source
category using: ISC modeled deposition
rates of particle-bound contaminants,
ISC modeled vapor-phase concentrations,
and soil concentrations diluted by tillage
g) Same algorithm as for on-site soil with
pasture grass and fodder concentrations
now a function of ISC modeled
contaminant concentrations
a) Same algorithm as on-site source category Surface water
b) Same algorithm as on-site source category Fish
(continued on the following page)
Air: vapor
Not modeled
Fruits and
vegetables
Meat and milk
5-5
7/31/92
-------
Table 5-1. (continued)
DRAFT-DO NOT QUOTE OR CITE
Source category
Fate and transport
solution algorithms
Exposure media
Ash Landfill
(continued)
c) Landfill soil concentrations assumed
equal to 112 contaminant concentrations
in disposed ash; total size of landfill,
27 hectares, estimated assuming 30 years
of ash disposal; "active" portion over
a duration of exposure assumed to 174 of
total size of 109 ha required for 30-year
lifetime; exposure site soil concentrations
estimated using erosion algorithm as in
off-site source category with similar
dilution for the tillage activity of
vegetable gardening
Soil
d) Same algorithm as off-site source category
Air: vapor
e) Emissions of suspended particulate include:
wind erosion as in on-site and off-site
source categories; vehicular resuspension
of contaminated roadway dust; emissions
off trucks delivering ash; unloading;
and spreading and compacting; far-field
dispersion models exposure site air-borne
particulate phase concentrations given
emission fluxes
f) Same algorithm as off-site source category
g) Same algorithm as off-site source category
Air: particulate
Fruits and
vegetables
Meat and
milk
5-6
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
5.2. CONSIDERATIONS FOR SOLUTION ALGORITHMS
Algorithms selected for estimation of exposure media concentrations are best
described as screening level algorithms. They are approximations of complex
environmental processes. Simple assumptions are often made in order to arrive at the
desired result, which is long-term average exposure media concentrations. It cannot be
overemphasized that measured concentrations of dioxin-like compounds in exposure media
provide more appropriate information for estimating exposure than the estimations made
with the algorithms of this assessment, or even estimations that could be made with the
most sophisticated modeling approaches. In Chapter 10 on Uncertainty, the exposure
media concentrations predicted were compared with measured field concentrations
reported in the literature. Although this was not a validation exercise, the comparisons
generally showed that predictions were similar to measurements.
One important assumption made for all but one of the algorithms is that the source
strength remains constant throughout the period of exposure; no dissipation via
environmental degradation or otherwise is assumed. The tendency for the dioxin-like
compounds to resist degradation and tightly sorb to soils and sediments supports the part
of the assumption concerning degradation. However, initial concentrations in soil or
sediment will dissipate over time due to the processes which move the chemical in the
environment, such as runoff, volatilization, or leaching from sediments into overlying
surface waters. The one algorithm where this assumption was not made is the algorithm
estimating volatilization flux (and subsequently vapor phase concentrations, inhalation
exposures and risks). Contaminants available for volatilization are assumed to originate
from deeper soil depths over time reflecting the dissipation, although not necessarily
degradation, of surface residues. Because residues originate from deeper in the soil,
volatilization flux decreases over time.
Assumptions about the most likely or reasonable route between the beginning point
of contamination, on-site or off-site soil for example, and the exposure media endpoint
needed to be made for some of the algorithms. Often it was possible that exposure media
could become contaminated by more than one route. For example, volatilization was
modeled to occur for the on-site soil source category, resulting in vapor-phase air
concentrations to which an individual living at the site was then exposed. For the off-site
soil source category, vapor phase concentrations are also estimated for an exposed
5-7 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
individual who is residing at some distance from the contaminated site. The soil at this
individual's residence is assumed to become contaminated over time due to erosion from
the nearby contaminated site. Two approaches are then available to estimate vapor-phase
contaminant concentrations. One is to assume that volatilization occurs at the
contamination site, and vapors are then transported to the exposure site. Another is to
assume that the exposure site soil (and soil between the contaminated site and the
exposure site, for that matter), now assumed to contain a thin layer of contaminated soil,
emits contaminated vapors to which the individual is exposed. Should emissions to which
individuals are exposed be assumed to originate at the contaminated site or at the
exposure site, or is it most appropriate to assume emissions occur from the contaminated
site and all nearby impacted soil? Emission of vapors from the off-site contaminated soil,
followed by air transport to the exposed individual, was the strategy selected for this
assessment. This assumption was made because it may be more plausible to assume that
the deeper and more uniform soil contamination located off-site is more likely to be an
on-going source of vapor emissions in comparison to the thin layer of contaminated eroded
soil which does deposit on the site of exposure. This and other alternate solution
strategies are discussed in Chapter 10 on uncertainty.
Chapter 10 also provides other technical discussions that are not included in this
chapter, as a way of evaluating some of the algorithms described here. Where possible,
model predictions are compared with observed data. For example, algorithms to estimate
vegetative concentrations of contaminants are presented in this chapter. Chapter 10
evaluates the algorithm by comparing results with literature studies which measured soil
contamination and resulting plant contamination.
5.3. ALGORITHMS FOR THE "ON-SITE SOIL" SOURCE CATEGORY
As earlier noted, the contamination and exposure occur at the same site for this
source category. The contamination is assumed to originate at the soil surface. As such,
the soil itself is the exposure media for the dermal contact and soil ingestion pathways.
Sections 5.3.1 through 5.3.4 describe the algorithms for estimating concentrations of the
dioxin-like compounds in: bottom sediment, suspended solids, and in the dissolved phase
in the water column of surface water bodies (5.3.1), in the air in the vapor phase (5.3.2)
and particulate phase (5.3.3), and in biota including fish (5.3.4.1), home-grown vegetables
5-8 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
and fruit (5.3.4.2), and home-produced beef and milk (5.3.4.3).
5.3.1. Surface Water and Sediment Contamination
Contaminated soil can erode into and, over time, result in the contamination of
sediments of nearby surface water bodies. If soil sources are expected to supply
contaminants to the surface water body over time, then concentrations in suspended and
bottom sediments can be estimated as a function of concentrations in eroding soil. For
purposes of this assessment, it will be assumed that equilibrium is achieved between
concentrations in the dissolved phase in the water column, concentrations in the sorbed
phase in the water column, and concentrations in bottom sediments. Concentrations in
bottom sediment are desired because fish concentrations are estimated as a function of
bottom sediment concentrations (see Section 5.3.4.1). Concentrations in suspended
solids are desired because they are used to estimate bottom sediment concentrations, and
dissolved phase concentrations are needed for estimating drinking water exposures.
The principal assumption for this approach is that dioxin-like compounds enter the
surface water body, either a standing water body such as a lake or a moving water body
such as a river, only through surface erosion. Other key assumptions and simplifications
include: 1) suspended solids are simply a reservoir into which dioxin-like compounds sorb;
more complex approaches consider sorption onto more than one reservoir of suspended
materials including suspended particulates and dissolved organic matter, 2) the sorption of
dioxin-like compounds onto suspended solids and bottom sediments is principally a
function of organic carbon, 3) the water body is completely mixed; results are described as
average for the water, 4) contaminants are not subject to dissipation processes such as
photolysis or volatilization, and 5) suspended solids concentrations are assumed to be
constant; also, no deposition/suspension or other sediment transfer and transport
processes enter into the solution.
Concentrations of dioxin-like compounds entering the water body via erosion are
assumed to be a simple linear function of concentrations of dioxin-like compounds in
surface soils from all or a portion of a watershed draining into the surface water body:
5-9 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Cs SLS As SDS ) + ( Cw SLW Aw SDW )
( SLS As SDS ) + ( SLW Aw SDW )
where:
CSwb = concentration on soil entering water body, mg/kg
Cs = contaminated site soil concentration of dioxin-like compound, mg/kg
SLS = unit soil loss from contaminated site area, kg/ha-yr
As = area of contaminated site, ha
SDS = sediment delivery ratio for soil eroding from contaminated site to
water body, unitless
Cw = average concentration of dioxin-like compound in effective area of
watershed not including contaminated site, mg/kg
SLW = average unit soil loss for land area within watershed not including
contaminated site, kg/ha-yr
Aw = effective area of watershed; the area contributing sediment which
mixes with the sediment originating from As, ha
SDW = sediment delivery ratio for watershed, unitless
The solution for estimating water column sorbed phase, water column dissolved
phase, and bottom sediment concentrations then requires estimating total water body
concentration, defined as the mass of contaminant in the water column divided by the
volume of water. This is equal to:
Ctot = Cswb TSS 1CT6 (5-2)
where:
Ctot = total water column concentration, sorbed + dissolved, mg/kg
Cswb = concentration on soil entering water body, mg/kg
TSS = total suspended sediment concentration, mg/L
5-10 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
10"° = converts Cswb in mg/kg to mg/mg
Dissolved phase water column concentrations are then estimated using an approach
developed by Mills, et al. (1985) and others. The equation is given as Equation (5-3a)
below. The concentration on suspended solids can then be estimated given dissolved
phase concentrations and a partition coefficient, as in Equation (5-3b). For estimating
bottom sediment concentrations, it will be assumed that the carbon normalized
concentrations on suspended solids equals the carbon normalized concentration on bottom
sediments. This is given below as Equation (5-3c):
- 7
TSS 10
(5-3a)
csgad = Kdssed rt (5-3b)
oc
'sed = Cssed—^- (5-3C)
ut~ssed
where:
Cwat = dissolved-phase water concentration of contaminant, mg/L
Ctot = total water column concentration, sorbed + dissolved, mg/kg
Kissed = soil-water partition coefficient for contaminant in suspended
sediment, L/kg
TSS = total suspended solids in water column, mg/L
Cssed = concentration on suspended sediments, mg/kg
Csed = concentration on bottom sediments, mg/kg
= fraction organic carbon on bottom sediments, unitless
= fraction organic carbon on suspended sediments, unitless
10~6 = converts mg/kg to mg/kg
5-11 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Note that this solution is applicable to both a pond/lake or a stream/river setting. The
differences in the two water systems can be expressed in the parameters, effective
watershed area, Aw, organic carbon contents of suspended solids and bottom sediments,
OCssed and OCsed, and total suspended sediment, TSS. Guidance on each of these
parameters is now given:
• C8 and Cw: These are concentrations of dioxin-like compounds in the
contaminated site soil, Cs, and the average within the effective area of the watershed, Cw.
The contaminated site concentrations drive the concentrations assumed for most
exposures, and is a principal user input (for the on-site source category, the contaminated
site is also the site of exposure). The simplest assumption for Cw is that it is 0.0.
However, examination of soil data from around the world shows that, where researchers
have measured concentration in what they described as "background" or "rural" settings,
soil concentrations of PCDDs and PCDFs are in the non-detect to low ng/kg (ppt) range
(see Section 9.5). Example Scenarios 1 and 2 in Chapter 9 demonstrate the on-site source
category. For these example scenarios, Cw and Cs are both initialized at 10~6 mg/kg (1
ppt) for all example compounds representing low concentrations that might be possible for
basin-wide areas.
• SL8 and SLW: These are the unit soil loss, in kg/ha, from the exposure site
and the average from the effective land area draining into the surface water body. In the
simplest case, the unit losses can be considered equal, in which case the SLS and SLW
terms drop out of Equation (5-1). In the most complicated solution, the effective drainage
area can be broken up into "source areas", where each source area can be unique in terms
of the erosion potential, concentration of contaminant, and so on. The total contribution
equals the sum of contributions from each source area, as: ICj*SLj*Aj*SDj for the
numerator in Equation (5-1) and ISLj*Aj*SDj for the denominator, for j number of source
areas not including the exposure site. For direct input into Equation (5-1), the terms Cj,
SLj, Aj, and SDj, should be determined and Cw, SLW, and SDW should be estimated as
weighted averages over all source areas, Aj. The effective drainage area, Aw, would be
the sum of all source areas, Aj.
For the example scenarios in Chapter 9 demonstrating the on-site source
5-12 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
categories, SLS and SLW are assumed equal, and hence drop out of Equation (5-1).
However, soil loss estimates for modeling surface water and off-site impacts will be
required for other source categories. The following is offered as general guidance and
background for estimation of unit soil losses in this assessment.
The unit soil loss is commonly estimated using the Universal Soil Loss Equation.
This empirical equation estimates the amount of soil eroding from the edge of a field
(Wischmeier and Smith, 1965):
SL = R K LS C P (5-4)
where:
SL = average annual soil loss, Eng. tons/acre-year
R = rainfall/runoff erosivity index, t-ft/ac-yr
K = soil credibility factor, t/ac-(unit of RF)
LS = topographical factor, unitless
C = cover and management practice, unitless
P = supporting practices factor, unitless.
Several references are available to evaluate USLE factors for agricultural and
non-agricultural settings (EPA, 1977; USDA, 1974; Wischmeier, 1972; Novotny and
Chesters, 1981). For this assessment, values for four of these terms, the R, K, LS, and P
factors, were determined from a survey of over 70 sanitary landfills in the United States
(SAIC, 1986). The fifth term, the cropping factor C, is discussed independent of that
survey.
• Rainfall/erosivity index, R: The R term represents the influence of
precipitation on erosion, and is derived from data on the frequency and intensity of storms.
This value is typically derived on a storm-by-storm basis, although it has been compiled
regionally for average annual values (EPA, 1977). This value is one such annual average
value. Annual values range from < 50 for the arid western United States to > 300 for
the Southeast. The value from the landfill survey of 155 is typical of rainfall patterns seen
in much of the midwestern United States.
5-13 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
• Soil credibility, K: The soil erodibility factor reflects the influence of soil
properties on erosion, with values ranging from <0.05 for non-erodible sandy soils to
>0.50 for highly erodible silty soils. The landfill survey value of 0.23 represents
mid-range erodibility typical of, for example, sandy loam with 2% organic matter.
• Length-slope factor, LS: The topographic factor reflects the influence of
slope steepness and length of the field in the direction of the erosion. Steeper slopes and
longer lengths lead to higher LS values, with a range of 0.1 for slopes < 1.0% and lengths
< 100 ft to >2.0 for slopes generally >10%. A landfill survey value of 1.5 can
correspond to a variety of combinations, such as a 8% slope over a 100-foot length or a
4% slope over a 1,000-foot length.
• Support practice factor, P: The support practice factor reflects the use of
surface conditioning, dikes, or other methods to control runoff/erosion. P can be no
greater than 1.0. The landfill survey value of 1.0 reflects a compacted surface without
control structures.
The final term in the USLE is the cover and management practice factor, C, which
primarily reflects how vegetative cover and cropping practices, such as planting across
slope rather than up and down slope, influences erosion. C values can be no greater than
1.0, with this value appropriate for bare soils. A C value of 1.0 is an appropriate choice
for active landfills. For an inactive landfill with grass cover or any area with dense
vegetative cover, a value of 0.1 is appropriate.
The landfill survey data (SAIC, 1986) allowed for averages for the R, K, LS, and P
values, as noted above. The survey data could be used in two ways. These average
values could be multiplied together, which yields (R)(K)(LS)(P) = 53 t/ac-yr. Alternatively,
the factors for each landfill in the SAIC survey could be multiplied together and then
averaged, which yields (R)(K)(LS)(P) = 62 t/ac-yr. Since there may be some dependency
among the factors, the latter method may be more appropriate. A unit erosion rate from
bare, contaminated sites will be assumed to be 62 t/ac-yr (converted to kg/ha-yr) in this
assessment. The average erosion rate for other areas in the watershed, SLW, used in this
assessment will be 6 t/ac-yr. The factor of ten difference assumes that vegetation
5-14 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
reduces erosion by a factor of ten on the average in the watershed as compared to bare
contaminated sites.
It should be noted that more sophisticated models are available for estimating
erosion rates (i.e., CREAMS as described in Knisel, 1980), and should be considered in
actual site-specific assessments.
• As and Aw: These are the area terms, including the area of the exposure
site and the effective drainage area of the watershed, both in ha. The scenarios
demonstrated in Chapter 9 have assumed 0.4 ha (1 acre roughly) for exposure sites
described as rural residences and 4 ha (10 acres) for farms. The total area impacting a
river system has been termed a watershed. For purposes of this assessment, an
"effective" drainage area can be less than the total area of a watershed. Instead, it equals
only the portion of the watershed impacting the river system upgradient of the point where
water is extracted for drinking or fish are recreationally caught for fishing. For example,
say a contaminated site is at the top of a watershed, and water was extracted and fish
were caught also at the top of the watershed. Then sediments and hence water will
principally be impacted by that contaminated area and a small top portion of the larger
watershed; the "effective" drainage area would equal a small part of the total watershed
including the contaminated site. If, on the other hand, water were extracted for drinking
and fish for consumption at the bottom of the watershed, the "effective" drainage area
would equal the entire watershed. If water were extracted upgradient of the contaminated
site, than the sediment would not be impacted by the contaminated site. For a standing
water body such as a lake, and given an assumption of complete mixing, the effective area
equals all areas contributing sediment, including the contaminated site.
A useful data source for this term and the suspended sediment term below, for
specific sites in the United States, is Appendix F in Mills, et al. (1985). This appendix
includes a compilation of data from river and reservoir sediment deposition surveys,
including drainage area, water body volumes, and rates of sediment deposition
(mass/area-time). Water bodies in this data base are located in the 48 conterminous
states. An estimate of suspended sediment concentrations can be made using the water
volume and the sediment deposition rates from this data, and an assumption on sediment
deposition velocity. The specific weight of sediments in the water body, also supplied in
5-15 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
this appendix, can be used to estimate sediment deposition velocity.
A survey of 70 landfills describe above (for estimation of unit soil loss terms)
revealed that the landfill were within an average watershed size of 4,000 hectares
(10,000 acres, 15.6 mi2). This will be the assumed size of all watersheds for the example
Scenarios in Chapter 9. Furthermore, it will be assumed that water is extracted at the
bottom of the watershed for drinking, and this will also be where fish are recreationally
caught for consumption. With this assumption, the placement of the contaminated site in
relation to the point where water is extracted does not become relevant. The effective
drainage area for the example scenarios in Chapter 9 will be 4000 hectares.
• SD8 and SDW: These are the sediment delivery ratios applied to the exposure
site and the watershed as a whole. Such a ratio is required because not all the soil which
erodes from an area deposits into the receiving water body. The following delivery ratio
was proposed for construction sites (EPA, 1977):
SDS = (3.28 DL )"°'22 (5-5)
where:
SDS = sediment delivery ratio from site of interest, unitless
DL = distance from site to receiving water body, m
3.28 = converts m to ft (empirical equation was developed for units of ft).
Note that the sediment delivery empirical equation simplifies all land features
pertinent to erosion to a function only of length. The equation was developed to estimate
sediment loads from construction sites to nearby surface water bodies, and from distances
up to 250 m (800 ft, roughly). Without specific information on the sites from which it
was developed, it is assumed that the land area between the construction sites and the
receiving water body is "average" and this relationship can be used for applications other
than construction sites.
As noted in previous bullets, the example scenarios demonstrating the on-site
source category assumed Cs = Cw, and SLS = SLW. With these assumptions, derivation
for SDS and SDW are irrelevant since they cancel in Equation (5-1). However, sediment
5-16 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
delivery ratios are required for the other example scenarios. In those scenarios, the
impacted water body was assumed to be 150 meters away from the site of
contamination, or site of exposure. This distance translates to a delivery ratio of 0.26.
Site-specific conditions could result in a larger (steeper slope, e.g.) or smaller proportion of
the eroded soil being delivered to the water body than would be estimated with this
equation.
Figure 5.1 shows a watershed delivery ratio as a function of watershed size (figure
from Vanoni, 1975). As seen, the ratio decreases as land area increases. The total
watershed size assumed for the example scenarios in Chapter 9 was 4,000 hectares, or
40 km2. From Figure 5.1, this translates to a watershed delivery ratio, SDW, of 0.15.
10 100
DRAINAGE AREA (km2)
1000
Figure 5.1. Watershed delivery ratio, SDW, as a function of watershed size (figure from
Vanoni, 1975).
• TSS: This is the total suspended sediment in the water body. This value.
will be lower for standing water bodies such as ponds or lakes as compared to streams or
rivers. The more turbulent flow in rivers will suspend sediments to a greater degree than a
5-17
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
relatively calm lake. A complex modeling exercise evaluating the impact of 2,3,7,8-TCDD
to Lake Ontario assumed a suspended sediment concentration of 1.2 mg/L (EPA, 1990b).
For use in pond or lake settings, an assumption of a suspended sediment concentration of
1-2 mg/L is reasonable. All example scenarios in Chapter 9 assume that the 4,000 ha
watershed drains into a river suitable for supporting fish for consumption and water for
drinking purposes. General guidance offered for the potential for pollution problems in
rivers and streams as a function of suspended sediment concentration are: 10 mg/L or
less - no problem, 100 mg/L or less - potential problem, and greater than 100 mg/L -
probable problem. A cutoff concentration for protection of aquatic life is 80 mg/L (Mills,
et al., 1985). The value assumed for TSS for all example scenarios in Chapter 9 is 10
mg/L, indicating no turbidity problems and a river fully supportive of fish for consumption.
* KcUsed: Tn's adsorption partition coefficient describes the partitioning
between suspended sediment and the water column. For numerous applications for
organic contaminants, particularly for estimating the partitioning between soil and soil
water, this partition coefficient has been estimated as a function of the organic carbon
partition coefficient and the fraction organic carbon in the partitioning media:
OCssed (5-6)
where:
= partition coefficient between suspended sediment and water, L/kg
Koc = organic carbon partition coefficient for contaminants, L/kg or cm3/gm
= fraction organic carbon content of suspended sediment, unitless.
The organic carbon partition coefficient, Koc, can be a measured value or it can be
estimated. Schroy, et al. (1985) listed an organic solids/water partition coefficient of
468,000 for 2,3,7,8-TCDD. Information in Jackson, et al. (1986), imply that this is a very
low partition coefficient for 2,3,7,8-TCDD. They obtained soil samples contaminated with
2,3,7,8-TCDD from 8 sites in the Times Beach area of Missouri, and 2 from industrial sites
in New Jersey. These contaminated soils had 2,3,7,8-TCDD concentrations ranging from
5-18 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
8 to 26,000 //g/kg (ppb), and organic carbon contents ranging from 0.015 to 0.08. They
determined soil water partition coefficients, Kds/ for these soil samples, and using the
organic carbon fraction data, estimated Kocs for 2,3,7,8-TCDD. The mean Koc from these
ten samples was roughly 24,500,000. EPA (1990) evaluated the Koc for sorption of
2,3,7,8-TCDD onto Lake Ontario sediments. They concluded that log Koc was greater
than 6.3 (Koc = 2,000,000), but less than 7.3 (Koc = 20,000,000).
In the absence of measured values, the Koc can be estimated from a chemical's
octanol water partition coefficient, Kow. Empirical equations relating Kow to Koc are
listed in Lyman, et al. (1982). Of six different equations listed in that reference, the
following derived by Karickhoff, et al. (1979) is used to estimate the Koc for the example
compounds in Chapter 9:
log Koc = log (Kow) - 0.21 (5-7)
where:
Koc = organic carbon partition coefficient, L/kg
Kow = octanol water partition coefficient, unitless
This equation was empirically developed from a limited number of hydrophobic
contaminants (n = 10, R2 = 1.00). It implies that Koc is very similar to Kow for strongly
sorbed compounds such as the dioxin-like compounds. Using the log Kow of 6.64 given
in this assessment for 2,3,7,8-TCDD in Karickhoff's relationship estimates a Koc of
roughly 2,700,000.
* OCsed, OC8ged: The organic carbon content of solids and sediments of
water bodies are generally higher than organic carbon contents of the surrounding lands.
Furthermore, organic carbon contents of suspended organic materials and solids are
typically greater than those of bottom sediments. A significant sink for strongly
hydrophobic contaminants such as the dioxin-like compounds is thought to be suspended,
or non-settling, organic material. In modeling 2,3,7,8-TCDD in Lake Ontario (EPA, 1990b)
using the WASP4 model, a compartment separate from suspended solids termed "non-
settling organic matter" served as a permanent sink. For purposes of this assessment, a
5-19 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
single reservoir of suspended materials onto which incoming TCDD sorbs is principally
characterized by OCssed, and the values selected for OCsed and OCssed should reflect the
relative partitioning behavior of suspended and bottom materials. As noted above, these
water body carbon contents are also related to the organic carbon contents of surrounding
soils. The model parameter, OCs), is the soil organic carbon fraction and is required for
modeling of soil contamination by dioxin-like compounds. Foth (1978) lists the organic
nitrogen content of several soil types ranging from sand and sandy loam to clay. The
range from that list is 0.0002 - 0.0024 on a fractional basis. Assuming a carbon to
nitrogen ratio of 10 (Brady, 1984; who notes that C:N ratios vary from 8 to 1 5, with the
typical range of 10 to 12), organic carbon ranges from 0.002 to 0.024. A soil organic
carbon fraction, OCs), is assumed to be 0.01 for all example settings in Chapter 9, which
is in the middle of this range. The organic carbon content of bottom sediments, OCsed will
be higher at 0.02. Bottom sediments originate as erosion from surrounding land, but also
include decay of organic materials within water bodies. The organic carbon content of
suspended materials can approach 0.20, but OCssed will be assumed to be 0.05 for the
example settings in Chapter 9.
5.3.2. Vapor-Phase Air Concentrations
The algorithms for estimating vapor-phase concentrations of contaminants were
presented and derived in Hwang, et al. (1986). These procedures were developed for soil
surface and subsurface contamination with polychlorinated biphenyls, PCBs. The models
are based on the assumptions that: 1) PCBs move through the soil primarily by vapor
phase diffusion, i.e., leaching is not considered, 2) PCB vapor in the soil matrix reaches a
local equilibrium with pore air, 3) degradation processes for PCBs were not considered ,
and 4) the PCB contamination occurs at the surface and extends down infinitely. These
assumptions are similar to the general types of assumptions that have been made for all
the algorithms estimating exposure media concentrations in this assessment. The
procedures in that PCB assessment were also used for this assessment. Details of the
derivation are presented in Hwang, et al. (1986).
The average flux rate over an exposure duration of ED can be estimated as:
5-20 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
where:
H
ED
(2) (Eslp) (Dea) (Cs) (H) (41)
Kds [ (if) (I) (ED) ]
°'5
(5-8)
FLUX =
average volatilization flux rate of contaminant from soil, g/cm2-s
soil pore porosity, unitless
effective diffusivity of contaminant in air, cm2/s
contaminant concentration in soil, ppm or mg/kg
Henry's Constant of contaminant, atm m3/mol
soil/water partition coefficient, cm3/g
exposure duration, s
interim undefined term for calculation, cm2/s
P
ea
soil
Eslp + psoi|n-Eslp)[Kds/(41 H)]
particle bulk density of soil, g/cm3.
The effective diffusivity, Dea, is solved as a function of contaminant diffusivity in
air, and soil pore porosity:
vea = DC Esip
0.33
(5-9)
where:
:slp
= effective diffusivity of contaminant in air, cm2/s
= molecular diffusivity of contaminant in air, cm2/s
= soil pore porosity, unitless.
The soil adsorption partition coefficient, Kds, is given as:
Kds = Koc OC
si
(5-10)
5-21
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
where:
Koc = contaminant organic partition coefficient, cm3/g
OCs) = fraction organic carbon in soil, unitless.
Porosity, Es,p, is defined as the pore space in soils occupied by air and water, and
for sandy surface soils show a range of 0.35-0.50. Medium to fine-textured soils (loams,
clays, etc.) show a higher range of 0.40-0.60 (Brady, 1984). Soil porosities in the
example settings were 0.50. Particle bulk density is defined as the mass of a volume of
soil solids. This contrasts the more common parameter, bulk density, which is the mass of
a unit of dry soil, which includes both pores and solids. Particle bulk density, Psoil, has a
narrow range of 2.60 to 2.75, and for general calculation purposes, Brady (1984)
recommends a value of 2.65 for average mineral surface soils, the value used for the
example settings. Background on Koc and OCS| were given in Section 5.3.1. above.
It is noted in Hwang, et al. (1986) that this procedure would tend to overestimate
emissions and resulting exposures in situations involving small spills which would not
involve deep contamination. It is also noted that the average flux rate is inversely
proportional to the square root of the duration of exposure - the longer the duration of
exposure, the lower will be the average flux rate. Whereas this solution assumes an
unlimited reservoir of contaminant, it is an unsteady state solution (unlike other solution
strategies) and is essentially an average flux rate over an amount of time defined by the
exposure duration. Inherent in the solution was the consideration that residues dissipate
by volatilization at the surface layers, resulting in contaminants diffusing upwards from
deeper soil layers over time. With this longer path of diffusion, volatilized amounts
decrease, and hence the average flux over time also decreases.
Vapor-phase concentrations along the center (y = 0.0) of an area source can be
estimated from (Hwang, 1987):
r - (2/7T)0-5 FLUX a 1010 erf (e) -0.5 (Z/5Z)2 (5_1:L)
(_ - - - - - (^ I \ I
um
5-22 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
where:
va
FLUX
z
erf
e
b
S
10
io
vapor-phase concentration of contaminant in air, //g/m3
Average volatilization flux rate of contaminant from soil, g/cm2-s
side length parallel to the wind direction, m
mean annual wind speed, m/s
vertical dispersion coefficient in air, m
height of the exposed individual, m
error function
error function term, unitless
b/(2*(2*Sy)-5)
side length perpendicular to the wind direction, m
horizontal dispersion coefficient in air, m
converts g/cm2-m to //g/m3.
This was the model used to estimate on-site vapor-phase concentrations. The
dispersion terms, Sz and Sy can be estimated using site-specific wind rose data. In the
absence of data, these terms can be estimated assuming the most common stability class,
D, as:
= 0.1414 X
0.894
(5-12)
Sz = 0.222 X
0.725
(5-13)
where:
'v-z
X
horizontal and vertical dispersion coefficient, m
distance upwind of the contaminated site, m.
Mean annual windspeeds, Um, vary from between 2.8 and 6.3 m/s (EPA, 1985b).
An assumption of 4.0 m/s in the absence of site-specific average wind speeds might be
acceptable. Simple assumptions can be made to assign values to the length terms above:
5-23
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
a, b, z, and x. Assuming a square-shaped contaminated site, a equals b which equals the
square root of the area of the site. A common assumption for z, the height of the exposed
individual, is 2 m. The x term can be assumed equal to a side length (a or b), or can equal
the side length plus the distance to the exposed individual if the contamination is not
on-site and dispersion is modeled as "near field." For the residence and farm setting
examples in Chapter 9, where the contamination was on-site, the x term was equal to a
side length. Other assumptions and parameter values as described here were also made
for the example settings in Chapter 9.
5.3.3. Particulate-Phase Air Concentrations
The method for determining the flux of soil particles due to wind erosion for on-site
conditions was developed in EPA (1985b) based on Gillette's (1981) field measurements
of highly erodible soils. A key assumption for this solution is that the soil surface is
assumed to be exposed to the wind, uncrusted, and to consist of finely divided particles.
This creates a condition defined by EPA (1985b) as an "unlimited reservoir" and results in
maximum dust emissions due to wind only. This wind erosion flux is given as (EPA,
1985b):
Ee = 0.036 (1-V) (Um/U( )3 F(x) (5-14)
where:
2
E. = total dust flux of < 10 um particle due to wind erosion, g/m -hr
V = fraction of vegetation cover, unitless
Um = mean annual wind speed, m/s
Ut = threshold wind speed, m/s
F(x) = a function specific to this model.
EPA (1985b) provides details allowing for the application of this equation under a
variety of circumstances. The following is offered as guidance specific to on-site
conditions:
5-24 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
• Fraction of vegetation cover, V: For a "residence" or "farm", grass or crops
are likely to substantially cover the soil, and V can range from 0.5 (minimal coverage) to
0.9 (more lush coverage). For the residence example settings, V was set at 0.9 which
assumes a continual grass cover over the contaminated soil. The V for the farm settings
was instead 0.5. The area of contamination for the example farm settings was larger than
the residence setting, 10 acres to 1 acre. The land where crops were grown was also
contaminated; the 0.5 value for V assumes that the cropland is totally or partially bare at
some times - perhaps during spring land preparation and fall harvest.
• Mean annual wind speed, Um: EPA (1985b) lists the mean annual wind
speeds at 10 meter height for the 60 major cities in the U.S. These values range from 2.8
to 6.3 m/s, with an average of approximately 4 m/s. This average value was assumed for
all example settings.
• Threshold wind speed, Ut: This is the wind velocity at a height of 7 m above
the ground needed to initiate soil erosion. It depends on nature of surface crust, moisture
content, size distribution of particles, and presence of non-erodible elements. It can be
estimated on the basis of the following procedure (EPA, 1985b):
Step 1. Determine the Threshold Friction Velocity
This is the wind speed measured at the surface needed to initiate soil erosion. EPA
(1985b) shows how it can be determined as a function of soil aggregate size distribution.
However, for the "unlimited reservoir" approach for which Equation (5-14) was developed,
soil particles are assumed to be fine at 1.5 mm or less as an average. This translates to a
threshold friction velocity of 75 cm/s and less. A value of 50 cm/s might be reasonably
assumed to be representative of these types of surfaces, and was assumed for this
assessment.
Step 2. Estimate the "Roughness Height"
EPA (1985b) graphically shows the roughness height for a range of possible
conditions. Included in this range are a roughness height of 0.1 cm for natural snow, 1.0
cm for a plowed field, 2.0-4.0 cm for grassland, 4.0 cm for a wheat field or for suburban
residential dwellings, and up to 1000 cm for high rise buildings. The assumption made for
5-25 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
the residence and farm example settings was 4.0 cm, following the information given for a
wheat field or a suburban residence.
Step 3. Estimate Ratio of Threshold Wind Speed at 7 m to Friction Velocity
A chart provided by EPA (1985b) shows this ratio as a function of roughness
height. For a roughness height of 4.0 cm, this ratio is seen to be 13.
Step 4. Estimate Threshold Wind Speed
This is simply the product of the ratio given in step 3 above and the friction
velocity. Using values given above, 50 cm/sec * 13 = 6.5 m/sec.
Finally, the model-specific function, F(x), is determined by first calculating the
dimensionless ratio x, where x = 0.886 Ut/Um (where Ut is the erosion threshold wind
speed (m/s) and Um is the mean wind speed (m/s)) and finding F(x) from a chart of F(x)
versus x, as provided in EPA (1985b). For Ut = 6.5 and Um = 4.0, x = 1.44 and F(x) =
1.05.
The unit dust flux is easily converted to a total contaminant flux by multiplying by
soil concentration and area:
WE = (2.8 X 10~13 ) Cs Ee Asc (5-15)
where:
WE = contaminant wind erosion emission rate, g/s
2
Ee = total dust flux of < 10 um particle due to wind erosion, g/m -hr
Cs = contaminant concentration in soil, ppb or ng/g
Asc = area of contaminated site, m2
2.8*10"13 = converts ng/hr to g/sec.
The next step in estimate particulate-phase contaminant concentration is to
estimate the dispersion term. The model that is used is the same as the one used to
estimate on-site vapor phase dispersion given in Equation (5-11) above. The following two
5-26 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
changes obtain the correct "FLUX" term for use in Equation (5-11):
1) Instead of WE, a total flux term presented here in units (g/sec) consistent with
other particulate flux terms discussed in this chapter, an appropriate "FLUX" for Equation
(5-11) is a unit flux term: Cs*Ee (ng/m2-hr). Since the algorithm for Ee was developed for
10 /jm size particles, the multiplication of Ee by Cs assumes that the concentration of
contaminant on 10 fjm size particulates is the same as that for the soil overall.
2) Cs*Ee is still not in the right units for Equation (5-11). The conversion term of
Equation (5-11), 1010, should instead be, .00028.
Substituting Cs*Ee for FLUX, and .00028 for 1010 in Equation (5-11) will allow for
the estimate of Cpa, the particulate phase concentration of contaminant in air, in units of
//g/m3.
5.3.4. Biota Concentrations
This section summarizes the algorithms to estimate contaminant concentrations in
fish, vegetation (including vegetables for human consumption and pasture grass or fodder
grown on contaminated soil for beef cattle consumption), beef, and milk. As will be
shown, all algorithms are simple empirical equations which relate an environmental media
concentration to a biota concentration, using a "biotransfer" or "bioaccumulation" factor.
5.3.4.1. Fish Concentrations
The procedure and supportive data for the algorithm to estimate fish tissue
concentrations can be found in Cook, et al. (1991). The information in that reference
focuses on 2,3,7,8-TCDD, although there is discussion on the related compounds covered
in this assessment including other CDDs, CDFs, and PCBs. As noted in that reference,
these compounds share a high degree of hydrophobicity that increases as the degree of
chlorination increases. Cook, et al. (1991) note that this corresponds in general to an
increase in lipophilicity and an increase in ability to bind to organic carbon in sediments and
to dissolved organic matter in water. However, these tendencies are not paralleled by an
increase in bioaccumulation. Only the 2,3,7,8-chlorine-substituted congeners are
substantially bioaccumulated by fish, although large quantities of other PCDD and PCDF
congeners are found in sediments. This pattern of bioaccumulation results because of
higher rates of metabolism of PCDDs and PCDFs in fish as compared to the
5-27 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
2,3,7,8-chlorine-substituted congeners (Cook, et al., 1991, with references to Muir et al.,
1986; Gobas, 1990). While the highly chlorinated 2,3,7,8-substituted congeners are very
slowly accumulated, they have very slow elimination rates.
2,3,7,8-TCDD and other planar polyhalogenated aromatic hydrocarbons often have
not been detected in water from aquatic ecosystems even when biota are highly
contaminated. Because surface layers of bottom sediments are a good indicator of the
relative amount of chemical in the system over the time scale involved for bioaccumulation
of super-hydrophobic chemicals, a term known as the Biota to Sediment Accumulation
Factor, or BSAF, has been offered as a measure of site-specific bioaccumulation potential.
This term was recently proposed to replace equivalent terms which were known as the
Bioavailability Index, or Bl (Kuehl, et al., 1987; Cook, et al., 1991, EPA, 1990b), the
Accumulation Factor, AF (Lake, et al., 1990) and the Biota to Sediment Factor, or BSF
(Parkerton, 1992; Parkerton, 1991, Thomann, et al, 1992). The BSAF is defined as:
BSAF = . (5-16)
where:
BSAF = biota to sediment accumulation factor, unitless
Ciipid = concentration of contaminant in lipid of fish, mg/kg,
Coc = concentration of contaminant in bottom sediment organic carbon,
mg/kg
The surface water algorithms estimate concentration of contaminant in bottom
sediments (see Section 5.3.1 above). This concentration, Csed, can be converted to an
organic carbon basis as a function of OCsed:
Cnr = ^L (5-17)
'sed
'oc 6C,
5-28 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
where:
n = concentration of contaminant in bottom sediment organic carbon,
OC
mg/kg;
Csed = concentration of contaminant in bottom sediment, mg/kg;
d = fraction organic carbon in bottom sediment, unitless
The organic carbon content of bottom sediments was assumed to 0.02, and the
concentration of contaminants on bottom sediments was modeled as a function of the
concentration of contaminants on incoming erosion and a modeled concentration on
suspended materials (see Section 5.3.1).
Since the accumulation of contaminant is assumed to occur only in fish lipid, a
correction term to estimate the whole fish tissue concentrations is needed since fish
consumption in g/day refers to whole fish consumption. The correction term is simply
f|ipid, and so whole fish concentrations are simply C|ipid * fNpid.
The BSAF was developed as a measure of bioaccumulation potential rather than as
a predictor, as it is being used here. It is uncertain as to whether measured BSAFs are
generally applicable to other water bodies. Efforts are underway to evaluate the general
applicability of BSAFs (P. Cook, Duluth Environmental Research Laboratory, US EPA, 6201
Congdon Boulevard, Duluth, MN 55804, personal communication). Using the BSAF
approach as a predictive tool greatly underplays the complexity of the processes
transferring contaminants from aquatic ecosystems to aquatic organisms. Following are
some of the key issues to consider:
1) Resident vs. Migratory Species: Parkerton (1991) applied a
bioenergetics-based bioaccumulation model in an attempt to duplicate BSAFs for
2,3,7,8-TCDD found for carp and blue crabs in the Passaic River, New Jersey. He showed
nearly a ten-fold difference in 2,3,7,8-TCDD BSAF calculated from data for resident
species as compared to migratory species in the Passaic River. This would be expected
for fish which also reside part of the time in relatively clean water bodies; migration would
enable depuration of residues from fish. The possibility that migration patterns might
explain some of the results for fish concentrations of 2,3,7,8-TCDD in the Lake Ontario
bioaccumulation study was also raised (EPA, 1990b). That assessment also discussed a
5-29 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
related issue of concern - to consider lakewide average sediment concentrations or
concentrations near where sampled fish were captured in calculating the BSAF. Even
within a large lake, more sedentary populations of fish may be impacted by localized
contamination.
2) Past history of contamination: If contamination of surface water bodies
with hydrophobic compounds like the dioxin-like compounds has occurred principally in the
past, then it can be expected that most of the contamination occurs in or near the bottom
sediment layer and not within the water column. Furthermore, if inputs to water bodies
are declining or low in comparison to past loadings, then sediments would be undergoing
depuration - residue levels would be declining, and the system may not be equilibrium.
EPA (1990b) noted that very low BAF*s (defined as a fish to sediment ratio not including
the sediment organic carbon and fish lipid considerations of BSAFs) and BSAFs for
2,3,7,8-TCDD in Lake Ontario contrasts higher BAF*s for other hydrophobic compounds
such as DDE or PCBs. An explanation offered is that loadings to the Lake may be
declining, such that there is a substantial disequilibrium between sediments, water, fish,
and their prey. One parameter required in the bioenergetics model Parkerton (1991) used
(referred to in the above bullet) was a ratio of contaminant concentration in bottom
sediment to that in suspended sediment, rs/rw. He found closer agreement between
measured and predicted BSAFs with this ratio equal to 10 in contrast to 1, the only two
values tested; a ratio of ten means that much more contaminant is in the sediment
compared to the water column. BSAFs found with this ratio equal to 1 were roughly 4
times as high as measured BSAFs, and BSAFs found with rs/rw equal to 10 were twice as
high as measured. A related result of his modeling exercise was that, at best fit between
modeled and measured BSAFs where the rs/rw was 10, dietary exposures explained over
50% of the BSAFs for carp and 85% in blue crabs, in contrast to water column exposures.
He speculates that prey organisms consist of benthic animals which ingest contaminated
sediment. Finally, he notes that 2,3,7,8-TCDD contamination in Passaic river largely
occurred as a result of historical loadings. The picture that emerges from Parkerton's
modeling is as follows: sediments are serving as an internal source of contaminants due
to past historical loadings despite a reduction of contaminant levels in the water column.
As a result, the bioaccumulation of these compounds in carp and blue crabs appears to be
mediated by trophic transfer via the benthic foodweb. Finally, in both the Lake Ontario
5-30 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
and Passaic River studies, concentrations of 2,3,7,8-TCDD were higher in deeper bottom
sediments as compared to surficial bottom sediments - this implies historical loadings and
possibly depuration of surficial residues.
This issue is non-trivial for the methodology of this assessment, since an
assumption for deriving suspended and bottom sediment concentrations is that the
contamination is ongoing, and that the hypothetical water body may be closer to a state of
equilibrium as compared to situations where contamination was principally in the past.
The BSAF assumed for 2,3,7,8-TCDD was based mainly on the data from EPA (1990b) on
Lake Ontario and from Parkerton (1991) from data in Passaic River. If 2,3,7,8-TCDD
contamination in these water bodies occurred mainly in past, and not in the present, then
measured BSAFs might be lower than those that might be measured in water bodies where
contamination is ongoing.
3) Variations among fish species: Feeding habits, age, migratory patterns, and
lipid contents (including lipid content of edible vs. inedible fish tissues) are just a few of
the interacting factors which determine a site-specific BSAF as a function of fish species.
The demonstration of this approach in Chapter 9 assigns a single BSAF to each of the
three example contaminants. Although not unlike other simplifications of this assessment,
such approaches are recognized as oversimplifications.
4) Study designs to obtain BSAFs: Although there is some evidence that
BSAFs specific to a contaminant may be applicable to other aquatic settings, data to
evaluate such a hypothesis is still sparse. Even data sets that do exist need to be carefully
evaluated before deriving BSAFs. Such an evaluation should consider data associated with
the sediment as well as data associated with the fish species. Critical factors for
sediment sampling include location, number, depth of sampling, variability, availability of
organic carbon fraction information, and so on. Similar issues are pertinent for fish
sampling and analysis.
Following now are guidance for the terms required for estimating fish tissue
concentrations.
5-31 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
• BSAF: Table 5.2 summarizes available biota sediment accumulation factors
for dioxin and furan congeners. Only four sets of data with such estimates were available
in the literature. The data from the Wisconsin River (Kuehl, et al. 1987) and that from 1
lake in Sweden (Kjeller, et al. 1990) both show decreasing BSAF with increasing
chlorination. The BSAF of 2.94 for 2,3,7,8-TCDD determined from a lake in Sweden
should be questioned since it is more than an order of magnitude different than any of the
other data. Causes for this discrepancy could be manifold. Some observations from
Kjeller, et al. (1990) might shed some light on this result. Although sediment data was
from three water bodies, 8 of the 9 Pike samples (pike samples were composites of 2-5
fish from one location in the water body) were from one of the water bodies. This is why
only data from the one water body was summarized in Table 5.2. This water body. Lake
Vanern, was clearly the most contaminated of the three water bodies studied. A paper
mill was located at the northern part of this lake and the authors concluded that
discharges from this mill impacted the lake. The average of 2,3,7,8-TCDD and 2,3,7,8-
TCDF organic carbon normalized concentrations for five sediment samples from this lake
was 297 pg/g; the analogous average concentration for 10 samples taken from another
lake, Lake Vattern (5 samples), and a river, Dala (5 samples), was 65 pg/g. A similar
disparity between Lake Vanern and the other water bodies is found with the
penta-CDD/CDF concentrations: 205 pg/g vs. 108 pg/g, with similar comparisons for the
hexa-, hepta, and octa-CDD/CDF. The sediment and corresponding pike sample nearest
this mill had the highest concentrations reported - pike samples were given as 3000 and
833 pg/g lipid normalized 2,3,7,8-TCDF and 2,3,7,8-TCDF (a composite from 5 pike taken
at this sampling station), respectively, and sediment was 1800 and 244 pg/g organic
carbon normalized-for 2,3,7,8-TCDF and 2,3,7,8-TCDD. Note the BSAF for 2,3,7,8-TCDD
implied from this data point is 3.41. Another consideration for high BSAFs might be
source of contamination. Speculation from the Lake Ontario and Passaic River field data
was that contamination principally occurred in the past, whereas in the Swedish data,
contamination appears to have been ongoing at the time of sampling. This might be one
indication that BSAFs for aquatic systems where contamination is ongoing might be
greater than from systems where the contamination is primarily historical.
The Swedish data also illustrates some of the complexities of interpreting literature
data. First, the sediment data was expressed concentrations normalized to "sediment
5-32 7/31/92
-------
Table 5-2. Available Biota to Sediment Accumulation Factors, BSAF, for dioxin-like compounds.
Reference/Congener
Water
Body
# Sed. samples
# Fish samples
BSAF
Comments
Kuehl, etal., 1987
2,3,7,8-TCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDD
1,2,3,6,7,8- &
1,2,3,4,7,8-HxCdd
1,2,3,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8-HpCDF
°^ US EPA, 1990b
2,3,7,8-TCDD
Carp
Wisconsin
River
1/1
CO
CO
NJ
Lake
Ontario
Brown Trout
Lake Trout
Smallmouth Bass
White Perch
Yellow Perch
55/81
55/81
55/14
55/38
55/77
Kjeller, etal. (1990)
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
OCDD
Pike
Lake
Vanern
in Sweden
4/6
0.27
0.06
0.06
0.035
0.037
0.0048
0.0033
0.03
0.07
0.05
0.20
0.03
2.94
1.03
0.17
0.086
0.018
0.002
Laboratory flow through
experiment using Wisconsin
River sediment and Lake
Superior water; BSAFs deter-
mined from one "represen-
tative" sediment sample and
one "composited" fish
sample - no other details
provided; sed. organic
carbon and fish lipid
determined analytically
Comprehensive field study
on bioaccumulation of
2,3,7,8-TCDD in Lake
Ontario; BSAFs are lakewide
and fish averages; report
evaluates matching fish
with sediment data from
where fish were caught
Results presented at right
derived from data in Kjeller,
et. al (1990); data includes
sediment samples from four sites
in Lake Vanern and 6 composited
(2-5 fish in composite) pike
associated with the four sites;
pike concentrations were expressed
on a lipid basis; Lake Vanern is near
a paper mill
D
3)
D
O
D
C
O
O
33
O
(continued on next page)
-------
Table 5.2. (cont'd)
00
CD
NJ
Reference/Congener
Water
Body
# Sed. samples
# Fish samples
BSAF
Comments
CJ1
I
GO
Kjeller, et al. (1990) (cont'd)
2,3,7,8-TCDF
1 2348/12378-PeCDF
2,3,4,7,8-HxCDF
123479/123478-HxCDF
1,2,3.6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
1,2,3,4,6,7,8-HpCDF
OCDD
Parkerton, T.F.
2,3,7,8-TCDD
Pike
Lake
Vanern
in Sweden
4/6
Passaic
River
Resident fish
Migratory fish
Blue Crab
61M1
61/15
61/14
1.40
0.25
0.71
0.036
0.065
0.27
0.047
0.0009
0.023
0.006
0.0001
0.081
0.009
0.055
7 "resident" fish species
was most represented by
carp; "migratory" species
were eel and striped bass;
TCDD contamination attributed
to historical industrial input,
particularly a 2,4,5-T plant
in operation in 1940s to 60s
D
DO
D
O
Z
O
H
D
C
O
H
m
O
33
O
H
m
-------
DRAFT-DO NOT QUOTE OR CITE
contents of organic material" (sic). This was interpreted as organic matter normalized, not
organic carbon normalized. Parkerton (1991) assumed that organic carbon was 45% of
organic matter to derive BSAFs when organic carbon data was unavailable; following this
lead, organic matter normalized concentrations in Kjeller, et al. (1990) were divided by
0.45 to arrive at organic carbon normalized concentrations. Also, there was not an exact
match in "sites" between sediment samples and fish samples; these sites were physical
locations within the large lake where samples were taken. There were five sites where
sediment samples were taken, and five sites where composited pike samples were taken in
Lake Vanern. However, one of the sediment and one of the pike samples were from
unique sites; only four sites had both sediment and pike samples. The results in Table 5.2.
were derived using average sediment and pike concentrations from only these four sites.
Another way to have derived BSAFs would be to average all lake sediment and pike
concentrations; since there may be some relationship between sediment and pike
concentrations based on lake location, it was decided to include only the four sites with
both fish and sediment samples. Finally, there were two sets of results listed for
1,2,3,4,6,7,8-HpCDF as though there were two unique sets of analyses for the same
congener; this is why there are two entries for this congener in Table 5.2.
Excluding the Swedish data, there are nine reported BSAFs in Table 5.2 for
2,3,7,8-TCDD. These range from 0.009 to 0.27, with an average of 0.09. A BSAF of
0.09 will be assumed for 2,3,7,8-TCDD in the example scenarios in Chapter 9. Although
there is indications of declining BSAFs with increasing chlorination, there is probably not
sufficient grounds to assign a BSAF for the second example compound, 2,3,4,7,8-PCDF,
significantly different from that of 2,3,7,8-TCDD. The BSAF for this example furan will
also be 0.09.
EPA (1990b) estimates BSAFs for PCBs and other selected chemicals (DDE, HCB,
etc.) for Lake Ontario from several data sets. Parkerton (1992) summarizes BSAFs for
PCBs and other compounds from other water bodies using other data sets. A selected
summary by water body taken from these two sources for PCBs is given in Table 5.3.
Two trends are apparent. First, the BSAFs for PCBs appear to exceed those of the
dioxin and furan congeners by an order of magnitude and more. Second, and from limited
data, it would appear that BSAFs increase from dichloro- through hexa- or perhaps
hepta-chloro PCBs, and then decrease thereafter. An assignment of a BSAF for
5-35 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 5.3. Summary of Biota Sediment Accumulation Factor, BSAFs, for PCBs (summaries
developed by EPA (1990b) and Parkerton (1992)).
PCS congener
Water bodv/Soecies
BSAF
Comments
PCB
Lake Ontario
trout, salmon,
perch, bass
4.06,0.77, Compiled in EPA (1990)
0.77, 0.52, from several data sources,
0.52, 0.86, years of study, fish
0.86, 3.35, species, and so on.
3.35, 1.42 Summary in this table
0.58 includes all uniquely
derived BSAFs for PCBs
from fish species noted;
PCBs not further identified
except BSAF = 0.58
identified for perch and
Aroclor 1254.
Siskiwit Lake
lake trout,
whitefish
trichloro-PCB
tetrachloro-PCB
pentachloro-PCB
hexachloro-PCB
heptachloro-PCB
octachloro-PCB
Compiled by Parkerton
(1992) from data in
Swackhammer, et al.
0.45-2.6 (1988) and Swackhammer
0.71-1.3 and Hites (1988); Parkerton
3.4-9.4 presents data for individual
2.9-20.8 congeners - summary at
12.5 right aggregates by
2.2-12.7 chlorination and including
both fish species; only
one data point presented
for heptachloro-PCB
(continued on next page)
5-36
7/31/92
-------
Table 5.3. (cont'd)
DRAFT-DO NOT QUOTE OR CITE
PCB congener
Water bodv/Species
BSAF
Comments
dichloro-PCB
trichloro-PCB
tetrachloro-PCB
pentachloro-PCB
hexachloro-PCB
heptachloro-PCB
octachloro-PCB
nonachloro-PCB
New Bedford
Harbor
flounder,
lobster, crab
0.11-
0.26-
0.65
1.05-
1.29-
0.84
0.23
0.02-
0.59
0.65
1.02
2.08
4.00
2.74
1.17
0.38
Compiled by Parkerton
(1992) from data in BOS
(1990); summary at right
is the range of values
across noted species
Total PCB
Rio de La Plata
Argentina
Three species of
marine fish
4.40
Determined by Parkerton
(1992) from Colombo, et
al. (1990) on total PCBs;
reference also has data
on PCB IUPAC congeners
5-8, 14, 19, 28-31, 52,
101, 110, 138, 153, 180
2,3,3',4,4',5,5'-HPCB is not apparent from the data summary below. The data point from
Siskiwit for the single heptachloro-PCB, which was 2,2',3,4',5,5',6-HPCB, was estimated
by Parkerton (1991) as 12.5. The BSAF for flounder from New Bedford Harbor estimated
by Parkerton (1991) was 0.84, with BSAFs for lobster and crab as 1.29 and 2.74,
respectively. A value of 2.00 is assigned to 2,3,3',4,4',5,5'-HPCB for the example
Scenarios in Chapter 9.
Finally, it should be noted that these assignments are based on data on vertebrate
rather than invertebrate aquatic species. It is generally recognized that invertebrates do
not possess the enzymatic capability to metabolize hydrophobic compounds as effectively
5-37
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
as higher chordates. As a result, invertebrate species such as mussels, clams, oysters,
shrimp, crabs and lobsters may have BSAF values much higher than those observed for
fish. Parkerton (1991) and Parkerton, et al. (1992) reviewed the literature to estimate
BSAFs of 1 to 5 for species including grass shrimp, sandworms, deposit feeding clams,
and blue mussel for PCDD/PCDFs and PCBs.
* f|ipid: Lipid contents of edible fish species have not been compiled, although
such a compilation would clearly be useful if applying a BSAF in an assessment mode such
as is done here. BSAFs are developed on the basis of whole fish lipid content, so
estimates of whole fish concentrations should assign f,- id on a whole fish rather than on a
tissue basis. Parkerton (1992) cautions, however, that lipid contents of edible portions of
fish may be lower than lipid contents of some of the fish portions that were sampled and
used to develop BSAFs. Non-edible high lipid content portions include, for example, liver
and hepatopancreas. Parkerton (1992) develops the parameter, /?, which is defined as the
ratio of the lipid content of the edible portion and the sampled tissue. To demonstrate the
impact of this ratio, Parkerton used data from Niimi and Oliver (1989) which included PCB
and other halocarbon compound concentration in whole fish and fillets of fish taken from
the Great Lakes. The ft (defined here as the ratio of lipid in fillet to lipid of whole fish)
forthese fish, which included brown trout, lake trout, rainbow trout, and coho salmon,
ranged from 0.22 to 0.51 . The ratio of fillet to whole fish contaminant concentrations
ranged from 0.20 to 0.54. Cook, et al. (1990) assumed a lipid content of 0.07 for fish in
discussions of BSAF and related methodologies for estimating bioaccumulation of
2,3,7,8-TCDD in aquatic ecosystems. This assessment will also assume a f|ipid of 0.07.
Different lipid contents have been reported for the same fish, so generalizations are
difficult to make at this point. EPA (1990b) lists percent lipid contents for Lake Ontario
fish including brown trout: 14.3%, lake trout: 21.1%, coho salmon: 6.45%, yellow perch:
5.2%, and white perch: 17.1%. Kuehl, et al. (1987) lists a range of percent lipid for carp
taken at different days during a study of between 13.0 and 18.7%..
5.3.4.2. Vegetation Concentrations
Vegetation concentrations are required for the estimation of exposure to
homegrown fruits and vegetables, and also for the beef and dairy food chain algorithms.
5-38 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Three principal assumptions are made to estimate vegetative concentrations:
• Outer surfaces of bulky below ground vegetation are impacted by soils
which contain dioxin-like compounds. Inner portions are largely unimpacted.
• Translocation of dioxin-like compounds from roots to above ground portions
of plants are negligible compared to other mechanisms which impact above ground
portions of plants (this assumption is examined in Chapter 10, Section 10.2.11.3). As
such, translocation into above ground portions will be assumed to be zero.
• Similar to the assumption concerning transport of contaminants from outer
to inner portions of below ground vegetation, it will be assumed that outer and not inner
portions of above ground bulky vegetation are impacted.
Concentration of contaminants in below ground vegetation is only required for
vegetables (carrots, potatoes, e.g.) grown underground. The basis for the below ground
algorithm is the experiments of Briggs, et al. (1982) on uptake of contaminants into barley
roots from growth solution, and their elaboration of a Root Concentration Factor. The
below ground concentration is given by:
C. RCF VGhp
s
where:
^bgv = fresh weight concentration of below ground vegetables, mg/kg
Cs = contaminant concentration in soil, ppm or mg/kg
Kds = soil-water partition coefficient, L/kg
Koc*OCs,
Koc = contaminant organic partition coefficient, L/kg
OCS| = fraction organic carbon in soil, unitless.
RCF = root concentration factor equaling the ratio of the contaminant
concentration in roots (fresh weight basis) and the concentration in
soil water, unitless
VGbg = empirical correction factor for below ground vegetation which
5-39 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
accounts for the differences in the barley roots for which the RCF
was derived and bulky below ground vegetables, unitless
Two processes, air-borne vapor phase absorption and air-borne particle deposition,
are assumed to contribute to above ground vegetation concentrations:
Cabv = Cvpa + Ca (5-19)
where:
Cat>v = concentration in above-ground vegetation, expressed on a dry weight
basis, mg/kg or ppm
Cvpa = contribution of concentration due to vapor-phase absorption or
airborne contaminants, mg/kg or ppm
Cppa = contribution of concentration due to settling of contaminated
particulates onto plant matter, mg/kg or ppm.
The basis for a vapor-phase bioconcentration factor for various airborne
contaminants, including 1,2,3,4-TCDD, from the atmosphere to vegetation was developed
by Bacci, et al. (1990, 1992). These authors conducted laboratory growth chamber
experiments on vapor-phase transfer from air to azalea leaves and developed a model to
predict the vapor-phase bioconcentration for 14 organic compounds. The outcome of their
work is a air-to-plant bioconcentration factor, which is termed Bvpa in this assessment.
The algorithm estimating plant concentrations as a function of vapor-phase air
concentrations is:
(5-2o)
5-40 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
where:
Cvpa = contribution concentration due to vapor-phase absorption or airborne
contaminants, mg/kg or ppm
Bvpa = air-to-leaf biotransfer factor, unitless [(//g contaminant/kg plant
dry)/(//g contaminant/kg air)]
Cva = vapor-phase concentration of contaminant in air,//g/m3
VGag = empirical correction factor which reduces vegetative concentrations
considering that Bvpa was developed for transfer of air-borne
contaminants into leaves rather than into bulky above ground
vegetation
da = density of air, kg/m3, 1.19
1/1000 = converts resulting concentration from/yg/kg to mg/kg.
Several exposure efforts for 2,3,7,8-TCDD (Fries and Paustenbach, 1990; Stevens
and Gerbec, 1988; Connett and Webster, 1987; Travis and Hattermeyer-Frey, 1991), have
modeled the accumulation of residues in vegetative matter (grass, fodder, vegetables)
resulting from deposition of contaminated particulates. Key components of their approach,
as well as the one for this assessment, include:
• Vegetative concentrations result from particulate deposition onto plant
surfaces.
• Vegetative dry matter yield is the reservoir for depositing contaminants; this
reservoir varies according to crop.
• Not all particulate deposition reaches the plant, some goes directly to the
ground surface; the "interception fraction", less than 1.0, reduces the total
deposition rate.
• Weathering processes, such as wind or rainfall, remove residues and this
process is reasonably modeled as a first-order exponential loss with an
associated weathering dissipation rate. All the above references have
justified a dissipation rate derived from a half-life of 14 days (based
principally on field measurements described in Baes, et al. (1984)); this is
the value used for all dioxin-like compounds in this assessment as well.
5-41 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
• Vegetative concentrations may not reach steady state because of harvesting
or grazing, but a steady state algorithm is used.
The steady state solution for plant concentrations attributed to particle deposition
is:
xooo
where:
cPPa = Vegetative concentration due to settling of contaminated particulates
onto plant matter, mg/kg or ppm
F = unit contaminant deposition rate,//g/m2-yr
Ij = fraction of contaminant deposition intercepted by crop j, unitless
kw = first-order weathering dissipation constant, 1/yr
YJ = dry matter yield of crop j, kg/m2
1/1000 = converts//g/kg to mg/kg.
Following is brief guidance on assignment of values to the terms in Equations (5-
18) to (5-21).
• Cs and Kds: This is the soil concentration and soil/water partition
coefficient, respectively. The soil concentration is specified for the on-site source
category. For the two source categories where soil contamination is distinct from the site
of exposure, the soil concentration at the site of exposure is estimated. As discussed in
Section 5.4.1 below, two soil concentrations including one for a no-till and one for a tilled
situation, are estimated. For estimating vegetable concentration, the tilled concentration is
required. The soil partition coefficient is a function of the contaminant organic carbon
partition coefficient, Koc, and the soil organic carbon fraction, OCS|, as discussed above in
Section 5.3.1. Division of Cs by Kds results in the equilibrium soluble phase concentration
5-42 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
of the contaminant, in mg/L.
• RCF: Briggs, et al. (1982) conducted experiments measuring the uptake of
several compounds into barley roots from growth solution. He developed the following
relationship for lipophilic compounds tested (lipophilic compounds were identified as those
tested that had log Kow 2.0 and higher; n = 7, r = 0.981):
log RCF = 0.77 log (Kow) - 1.52 (5-22)
where:
RCF = root concentration factor equaling the ratio of the contaminant
concentration in roots (fresh weight basis) and the concentration in
soil water, unitless
Kow = contaminant octanol water partition coefficient, unitless
Since his experiments were conducted in growth solution, the RCF is most appropriately
applied to soil water in field settings. This is why the Cs was divided by Kd in Equation (5-
18).
• VGu : This correction factor and the one used to correct for air-to-leaf
bg
transfer of contaminants, VGag, are based on a similar hypothesis. That hypothesis for
VGb is that the uptake of lipophilic compounds into the roots of this experiments is due to
sorption onto root solids. High root concentrations were not due to translocation to within
portions of the root hairs. Direct use of the RCF for estimating concentrations in bulky
below ground vegetation would greatly overestimate concentrations since an assumption
(stated above) is that there is insignificant translocation to inner parts of below ground
bulky vegetation for the dioxin-like compounds. Concentrations in outer portions of edible
below ground vegetation would mirror concentrations found in barley roots, by this
hypothesis.
VGb can be estimated by assuming that the outer portion, or skin, of below ground
5-43 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
vegetables would contain concentrations that can be predicted directly using the RCF, but
that the inner portions would effectively be free of residue. The correction factor can be
estimated as the ratio of the mass of the outer portion to mass of the entire vegetable:
VGbg = AM" (5-23)
MASS Vegetable
where:
VGbg = below ground vegetation correction factor, unitless
MASSskin = mass of a thin (skin) layer of below ground vegetables
MASSvegetab|e = mass of the entire vegetable
Simplifying assumptions are now made to demonstrate this ratio for a carrot and a potato.
First, it will be assumed that the density of the skin and of the vegetable as a whole are
the same, so the above can become a skin to whole vegetable volume ratio. The
thickness of the skin should be assumed to be same as the thickness of the barley root for
which the RCF was developed. Without that information given in Briggs, et al. (1982),
what will instead be assumed is that the skin thickness is equal to 0.03 cm. This was the
thickness of a leaf from broad-leaved trees assumed by Riederer (1990) in modeling the
atmospheric transfer of contaminants to trees. The shape of a carrot can be assumed to
be a cone. The volume of a cone is given as (rr/3)r2\, where r is a radius of the base and I
is length. Assuming a carrot base radius of 1 cm and a length of 15 cm, the volume is 16
cm3. The curved surface area of a cone is given as: m(r2 + I2)1/2, which equals 47 cm2,
given the r and I assumptions. The volume of the cone surface area is 47 cm2 * 0.03 cm,
or 1.41 cm3. The skin to whole plant ratio for this carrot is 0.09 (1.41 /16). A similar
exercise is done for a potato, assuming a spherical shape with a radius of 3 cm. The
volume is given as 4/3/rr3, or 113 cm3. The surface area of a sphere is 4/7r2, or 113 cm2,
and the volume of this surface area is 3.39 cm3. The skin to whole plant ratio for the
potato is 0.03.
This exercise indicates upper bounds for such an empirical parameter. For exposure
assessments, other factors which reduce vegetative concentrations should also be
5-44 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
considered and will be considered in this empirical correction factor in this assessment.
Additional reductions in concentration result from peeling, cooking, or cleaning, for
example. Wipf, et al. (1982) found that 67% of unwashed carrot residues of 2,3,7,8-
TCDD came out in wash water, and 29% was in the peels. A peeled, washed carrot
correction factor might instead be, 0.09*0.04, or 0.004 (0.09 from above; 0.04 =
100% - 67% - 29%). A 96% reduction in the estimated VGbg for the potato (the potato is
cleaned and the skin is not eaten; additional reductions possibly when cooking the potato)
would equal 0.001. In a site-specific application, the type of vegetation, preparation, and
so on, should be considered. The VGbg for underground vegetables for this assessment is
assumed to be 0.01. This is less than the estimates of 0.09 and 0.03 for the carrot and
potato above, but greater than it might be if based on this discussion on cleaning,
washing, peeling, and so on.
• Cva: The vapor-phase concentration of contaminant in air used in this
equation, Cva, is estimated using procedures described in Section 5.3.2 above.
• Bvpa: The model developed by Bacci, et al. (1990, 1992) relates the vapor-
phase bioconcentration factor from air to azalea leaves, Bvpa, to the chemical octanol-
water and air-water partition coefficients, Kow and Kaw. The air-water partition
coefficient, Kaw, is a dimensionless form of Henry's Law constant, H, derived by dividing
H by the product of the ideal gas constant, R, and the temperature, T. After substituting
H/RT for Kaw, Bacci's algorithm for Bvpa can be expressed as:
= Kow1-065 R T (5_24)
VP° 101160 H
where:
^vpa = air-to-leaf biotransfer factor, unitless [(/vg contaminant/kg plant
dry)/(//g contaminant/kg air)]
Kow = contaminant octanol water partition coefficient, unitless
5-45 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
H = contaminant Henry's Constant, atm-m3/mol.
R = ideal gas constant, 8.205 x 10~5 atm-m3/mol-deg K
T = temperature, 298.1 K
101160 = empirical constant
Substituting R and T into Equation (5-24) simplifies the right hand side to: 2.42*10~6
Kow1-065/H
• VG : The same discussion for this correction factor for below ground
vegetation applies here. Fruits such as apples, pears, plums, figs, peaches, and so on, can
be approximated by spheres, and upper bound estimates of correction factors would be
less than 0.05. Cooking and cleaning further reduces residues. The VGag for unspecified
above ground fruits and vegetables in this assessment is assumed to be 0.01. Like VGbg,
this value is assigned considering that it should be less than estimated just based on
surface volume to whole fruit volume ratios.
Two other VGag values are required for this assessment. One is for pasture grass
and the other for fodder consumed by cattle. Both are required to estimate concentrations
in these vegetations consumed by cattle in order to estimate beef and milk concentrations.
A VGag value of 1.0 was used to estimate pasture grass concentrations since there
appears to be a direct analogy to exposed azalea leaves. Hay can be considered to consist
mostly of leafy vegetation, and a VGag of 1.0 might be appropriate. Corn grown for
fodder would include substantially protected grain although cattle also consume some of
the outer portions of the vegetation. Given these considerations, fodder was assigned a
VGag of 0.5 in this assessment.
• F: Section 5.3.3 above describes the algorithm estimating the air-borne
concentration of contaminant in the particulate phase, Cpa, in units of//g/m3. This term
needs to be multiplied by the deposition velocity, Vd, in units of m/yr, to arrive at a unit
deposition rate in appropriate units. The value of Vd of 31 5,000 m/yr was adopted for
this assessment from Seinfeld (1986) who listed a gravitational deposition velocity of 1
cm/sec for 10//m size particles. This size is appropriate since airborne concentrations of
contaminants in the particulate are estimated with parameter assignments corresponding
5-46 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
to inhalable, or 10//m, size particulates.
• kw: Fries and Paustenbach (1990) note that this approach may
overestimate concentrations because crops can be harvested or pastures grazed before the
plant concentrations reach steady state, and that a kw based on a weathering half-life of
14 days may be too long given experimental results of Baes, et al. (1984) which showed a
range of 2-34 days, and a median value of 10 days. Stevens and Gerbec (1987)
considered harvest intervals by including the exponential term, (1-e"kt), and assigning
values of t based on harvest intervals of different crops. This assessment uses a kw of
18.02 yr~1, which is equivalent to a half-life of 14 days.
• 1: and Y-: Interception values and crop yields were determined in the afore-
mentioned assessments based on geographic-specific crop yield data provided in Baes, et
al., (1984) and the following types of crop-specific relationships estimating interception
fraction based on yield, also presented in Baes, et al., (1984):
corn silage: 1 = 1- e-°-768Y
hay/grasses: 1 = 1- e'2-88Y
lettuce: 1 = 1- e-°-068Y
Judgments by Fries and Paustenbach (1990) on high, medium, and low yields of silage,
hay, and pasture grass, and the use of the first two interception equations above (the first
for silage, and the second for hay and grass), can give some guidance on interception
fractions and yields for these crops:
corn silage
hay
grass
Yield
(kg/m2)
0.30 (low)
0.90 (med)
1.35 (high)
0.25 (low)
0.45 (med)
1.30 (high)
0.05 (low)
0.15 (med)
0.35 (high)
Intercept
Fraction
0.20
0.50
0.64
0.51
0.73
0.98
0.13
0.35
0.64
5-47 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
This information can be used for cattle intake of vegetation, and the resulting beef and
milk concentrations. The medium values for grass and hay were used for pasture grass
and fodder assumptions in the example setting in Chapter 9. Use of the hay yields and
interceptions assumes that the cattle fodder grown on-site is hay and not corn silage.
Stevens and Gerbec (1988), using yields obtained from the Minnesota State
Agricultural Office, derived the following yield and interception estimates, respectively, for
vegetables for human consumption in their assessment: lettuce - 8.6,0.72; tomatoes -
12.0,0.55; and beans - 2.7,0.18. Average yields and interception fractions from their
exercise: 7.8 kg/m2 and 0.48, were used in the example setting in Chapter 9. These
vegetable yields are fresh weight, so they need to be converted to a dry weight basis in
order to estimate a Cppa appropriate for use in Equation (5-19). Since vegetables are
generally 80 - >90% water, a fresh to dry weight conversion factor of 0.15 was used,
resulting in an average vegetable dry matter yield of 1.17 (7.8 * 0.15). This was used in
the example settings in Chapter 9.
When calculating concentrations in below ground fruit and vegetables using
Equation {5-18), Cbgv is on a fresh weight basis since the RCF developed by Briggs, et al.
(1982) is on a fresh weight basis, and no correction for estimating exposures is necessary.
However, Cabv as estimated in Equation (5-19) is on a dry weight basis, and should be
multiplied by a dry weight to fresh weight conversion factor when applied to above ground
fruits and vegetables. A reasonable estimate for this parameter for fruits and vegetables is
0.15 (which assumes 85% water), which was used in this assessment. When using
Equation (5-19) to estimate Cabv for the beef and milk food chain algorithm, a conversion
to fresh weight is not required, however, since the algorithms were developed assuming
dry weight concentrations.
5.3.4.3. Beef and Milk Concentrations
The algorithm to estimate the concentration of contaminant in beef and/or milk was
based on methods developed by Fries and Paustenbach (1990). Their key solution
assumptions were: 1) contaminants bioconcentrate equally in the fat portions of beef and
milk, 2) bioavailability, which, as they define it, is the fraction of ingested contaminant
which is absorbed into the body, depends on the vehicle of ingestion - dioxin in corn oil
5-48 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
has a bioavailability in the range of 0.7 to 0.8, in rodent feed it has an estimate of 0.5,
while in soil it has a range of 0.3 to 0.4, 3) differing concentrations of contaminant found
in beef and milk fat are not due to differences in how the contaminant bioconcentrates in
beef or milk fat, but rather to differences in the diets of cattle raised for beef and cattle
raised for dairy products.
The concentration in the fat of cattle products is given as:
Cfat = (FDFSBSACS) + (FDFgACg) + (F DFf ACf) (5-25)
where:
Cfat = concentration in beef fat or milk fat, mg/kg
F = bioconcentration ratio of contaminant as determined from cattle
vegetative intake (pasture grass or fodder), unitless
DFS = fraction of cattle diet that is soil, unitless
Bs = bioavailability of contaminant on the soil vehicle relative to the
vegetative vehicle, unitless
ACS = average contaminant soil concentration, mg/kg
DFg = fraction of cattle diet that is pasture grass, unitless
ACg = average concentration of contaminant on pasture grass, mg/kg
DFf = fraction of cattle diet that is fodder, unitless
ACf = average concentration of contaminant in fodder, mg/kg.
The following is offered as brief guidance to these terms and also the justification for the
values selected in the example Scenarios in Chapter 9.
• Bioconcentration ratio, F: Fries and Paustenbach (1990) reviewed the
literature on steady-state concentrations of 2,3,7,8-TCDD in the fat of meat and milk from
cows ingesting contaminated feed ("feed" was not specifically defined in Fries and
Paustenbach (1990); the word used in this discussion is "fodder" which is cattle feed
grown on-site and impacted by soil contamination). They calculated an F of between 4
5-49 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
and 6, and assumed a value of 5.0 in their example applications. This value is used in the
example settings for 2,3,7,8-TCDD. It is not used for the other example compounds,
however. Fries and Paustenbach (1990) concluded that the bioconcentration for PCDDs,
PCDFs, and PCBs were comparable to TCDD based on a literature review. They found
PCB BCFs of 3.1, 4.4, and 4.8 at 20, 40, and 60 days in a particular study, and a factor
of 5.7 for 1,2,3,6,7,8-CDD at 70 days. They also note, however, that bioconcentration
ratios and PCDDs and PCDFs in body and milk fat of cattle decreased significantly as
chlorination increased. The bioconcentration ratio of 1,2,3,4,6,7,8-CDD in beef in cows at
70 days and heifers (two data points for heifers) at 160 days was 0.4 and 0.2 (and 0.3),
respectively. They note a bioconcentration ratio of OCDD in cows at 70 days and heifers
at 160 days of 0.1 and 0.05. Finally, they note a bioconcentration ratio of 0.1 for OCDF
in beef fat of heifers at 260 days. The second example compound in Chapter 9 was
2,3,4,7,8-PCDF. This should have a slightly lower bioconcentration ratio as
2,3,7,8-TCDD, since it is more chlorinated. However, there was no data found in the
literature for this compound. The F value of 2,3,4,7,8-PCDF was set at 3.0. The third
example compound was 2,3,3',4,4',5,5'-PCB, a heptachloro-PCB, might have a
bioconcentration ratio less than 1.0, consistent with the above summary. It was given a
value of 0.5.
• Soil bioavailability, Bs: This parameter reduces the bioconcentration ratio, F,
considering that soil is a less efficient vehicle of transfer compared to fodder. Fries and
Paustenbach (1990) reviewed several studies on the oral bioavailability of TCDD in soil in
the diet of rats. Most studies used corn oil as the positive control, since there is a high
absorption of TCDD, with 70-83% of the dose absorbed. Their literature review showed
that the bioavailability of TCDD in soil was between 0.4 and 0.5 that in corn oil, or 0.3 to
0.4 overall. The literature implied a range of 50 to 60% of TCDD in standard rat feed is
absorbed, and although few studies were available, a similar 50% absorption rate of TCDD
in cattle feed was noted. They concluded, therefore, that the rat data was a reasonable
surrogate for cattle. This would lead to a Bs of 0.6 to 0.7 (0.3-0.4 divided by 0.5-0.6)
implying that absorption of TCDD when soil is the vehicle is 60 to 70% of what it would
be if fodder were the vehicle. The soil bioavailability term used for all example compounds
in Chapter 9 is 0.65.
5-50 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
• Soil diet fraction, DF8: Fries and Paustenbach (1990) report that soil intake
by cattle feeding on pasture varies between 2 and 18% of total dry matter intake,
depending on whether the grazing area is lush or not. The soil diet fraction would be
lower for cattle which are barn-fed with minimal opportunity for contaminated soil intake.
Cattle raised for milk are rarely pastured, so one possible assumption for estimating milk
fat concentrations would be a DF. of 0.0. Fries and Paustenbach (1990) assumed
o
between 0 and 2% of the dry matter intake by lactating cattle was soil in various
sensitivity tests. Since cattle raised for beef are commonly pastured, a conservative
assumption would be a high DFS of 0.15 (15%), although a more reasonable assumption
which would consider grazing in lush conditions and/or a portion of diet in fodder or
supplemental feed leads to DFS less than 0.10. Fries and Paustenbach (1990) assumed
DFS of between 0 and 0.08 for beef cattle in various sensitivity tests. The example
settings in Chapter 9 assume 0.02 (2%) for lactating cattle, and 0.08 (8%) for beef cattle.
• Fodder and grass diet fractions, DFf and DFg: The sum of the three diet
fractions, DFS + DFf + DFg must equal 1.0. Setting DFS equal to 0.02 (2%) for lactating
cattle assumed that they are pastured to some extent or could be taking in residues of soil
sticking to home grown fodder. Assuming lactating cattle graze a small amount of time,
the DFg for lactating cattle will be 0.08 (8%), and therefore the DFf will be 0.90. Beef
cattle are often fattened prior to slaughter by being fed fodder, so that some of their
lifetime intake of dry matter is fodder. Assuming that most of their lifetime dry matter is
pasture grass, beef cattle DFf will be 0.08, and DFg will be 0.90.
• Average contaminant soil concentration, AC8: The simplest assumption for
ACS would be that it equals the initial level of contamination, Cs. However, this would be
too high if the cattle also graze in uncontaminated areas. Where cattle have random
access to all portions of a grazing area with contaminated and uncontaminated portions, a
ratio of the spatial average of the contaminated area to the total area should be multiplied
by Cs to estimate ACS. If cattle spend more time in certain areas, these areas should be
weighted proportionally higher. Different assumptions might also be in order when using
Equation (5-25) to estimate milk fat as compared to beef fat concentrations. Lactating
cattle, if pastured, might graze on different areas than beef cattle. After determining a
5-51 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
spatial average based on current conditions, a second consideration might be given to
temporal changes. If soil levels are expected to change over time (due to changes in
source strength or other factors) then the concentrations should be averaged over the
exposure duration as well. The example scenarios in Chapter 9 where beef and milk
exposures were estimated were termed "farms". The methodologies in this chapter were
used to estimate the average soil concentration over the entire farm property. Assuming
the cattle are raised on the farm property, than 100% of their intake of soil comes from
the farm. This means that the average soil concentration, ACS in Equation (5-25), is equal
to the level of contamination given as the initial level, or determined as average for the
farm based on fate and transport algorithms.
• Average fodder and pasture grass concentration, ACf and AC : The
concentration of contaminant in pasture grass or fodder is equal to Cabv as calculated in
Equation (5-19). As detailed in Section 5.3.4.2. above, pasture grass or fodder grown
on-site can be impacted by air-to-plant vapor phase transfer and particulate deposition.
Refinements noted above include the empirical parameter VGag equals 1.00 when applying
the air-to-leaf transfer algorithm to pasture grass and 0.50 when applied to cattle fodder.
A refinement noted here, and like ACS above, is that an assumption needs to be made
about the fraction of fodder or fraction of pasture grass that is impacted by contamination.
Part of the fodder diet could come from outside sources and not be contaminated, and part
of the grazing area could be far from a localized area of soil contamination, making it less
impacted by contaminated particulates or vapors. The simplest assumption is that the
entire vegetative diet of the cattle includes pasture grass and fodder impacted by the
contaminated soil, in which case ACf and ACg would equal Cabv. For the sake of
simplicity and consistency, the assumption made for ACS was also made for ACf and AC
in the example Scenarios in Chapter 9. That is, the grass and fodder intakes of beef and
dairy cattle originate within the farm property and concentrations in grass and fodder are a
function of the soil concentrations within the farm property; ACf and ACg are equal to
Cabv as calculated in Equation (5-19). For site-specific situations, ACf and ACg should be
estimated as Cabv reduced according to assumptions on quality of cattle fodder, and
impacts of air-borne contaminants on grazing land and cattle fodder grown at the site
where cattle are raised.
5-52 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
There is one final but critical note on solving for beef and milk concentrations given
a solution for Cfat as in Equation (5-25). Human daily ingestion amounts are typically
expressed in whole product rather than the fat portion of product. Whole milk is 4% fat,
meaning that the Cfat needs to be multiplied by 0.04 to get whole milk concentration.
Similarly, beef is generally 22% fat, meaning that the Cfat needs to be multiplied by 0.22
to get whole beef concentration. However, the ingestion rates in this assessment for beef
and milk were developed on a fat basis, so no adjustment is necessary (see Chapter 8).
5.4. ALGORITHMS FOR THE "OFF-SITE" SOURCE CATEGORY
As noted in Section 5.1, the contaminated soil is remote from the site of exposure
for the "off-site" source category. A common example is an industrial site with soil
contamination or a landfill with contaminated soil. The example settings in Chapter 9
include an industrial site with bare, contaminated soil and an active ash landfill. Since
many of the parameters in the algorithms discussed below are specific to particular off-site
soil contamination sites, guidance in this section as well as Section 5.6 will be specific to
the example settings in Chapter 9.
Table 5-1 summarized the fate and transport algorithms for this source category
and as seen, several of the algorithms estimating exposure media concentrations are the
same or very nearly the same as in the on-site source category. Following now are bullet
summaries for similarities and small refinements to these algorithms. Sections below
describe algorithms that are unique for the off-site source category.
• Surface water impacts: Equation (5-1) is applied to estimate concentrations
in suspended and bottom sediments for this source category. Section
5.3.1. describes the parameters required for these equations. In applying
this algorithm for the example Scenario demonstrating the off-site source
category in Chapter 9, Example Scenario 3, the important assumption was
made that the average concentration of contaminants in the watershed was
very low compared to the concentration on the contaminated soil - hence,
Cw was set to 0.0. The impact to the surface water body was just from the
contaminated site. The unit erosion rate from the contaminated site, SL_, is
o
assumed to be 62 t/ac-yr. The unit erosion rate from other areas of the
5-53 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
watershed, SLW, is assumed to be 6 t/ac-yr. Derivation of these terms is
given above in Section 5.3.1. The contaminated site is assumed to be 1 50
meters away from the water body, and SDS is estimated therefore as 0.26.
Water is assumed to be extracted for drinking at the bottom of the
watershed. Therefore, the effective drainage area is 4000 ha. From Figure
5.1, it is seen the sediment delivery ratio associated with this drainage area
is approximately 0.15, which is assumed for the example scenario r
Chapter 9. The contaminated site for the example scenario is 4 ha.
Vapor-phase air concentrations: The volatilization of contaminants from soil
can be estimated similarly to the way described in Section 5.3.2 for the
on-site source category with no clean cover. Section 5.4.2 below describes
an approximation which reduces this flux when there is a clean cover, such
as in a capped landfill. The example scenario in Chapter 9 does not assume
a clean cover. Also, section 5.4.2 describes a dispersion model which
transports contaminants through the air to the exposure site. The far-field
dispersion model described in Section 5.4.2 differs from the near-field
dispersion model presented in Section 5.3.2.
Particulate-phase air concentrations: The same model for particulate flux
due to wind erosion is used for the off-site source category as described in
Section 5.3.3. This flux is considered only when there is no clean cover.
However, two parameters might be different than described above (Section
5.3.3, Equation (5-14)) if the model is applied to off-site soil contamination
when the soil is bare. One is the vegetative cover, V, which might be more
appropriately assigned a 0.0 implying no ground cover for an active landfill
or an industrial site. The other is the threshold wind speed, Ut. The
different assumption would be in roughness height, assumed 4 cm for a
residence or farm setting, but perhaps more appropriately assumed to be 1.0
cm for bare soil. This value is appropriate for a tilled field (EPA, 1985b).
With this change, a Ut is calculated as 8.25 m/sec, and F(t) is calculated as
0.5. Note that V and Ut might be the same as a residential or farm setting if
the off-site soil contamination had a grass cover, and contamination
originated at the surface - i.e., no layer of clean soil. Once a flux has been
5-54 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
calculated, the far-field dispersion model used in vapor-phase transport
(Section 5.4.2. below) is used also for estimating air-borne particulate-phase
contaminant concentrations at the exposure site. It is noted that if a landfill
were to have a clean cover of uncontaminated soil, than particulate
emissions would not occur from the landfill. The example ash landfill in
Chapter 9 was "active", meaning that the soil was bare and contaminated at
the surface.
• Biota concentrations: The basic strategy for estimating biota concentrations
- as a linear function of environmental media concentrations (bottom
sediment concentrations, soil, air) and based on bioaccumulation or
biotransfer factors (along with diet fractions, etc.) - remains the same.
Section 5.4.1. describes how exposure site soil concentrations are estimated
from concentrations at the off-site source. Exposure site soil then becomes
a "source" for plant, beef, and milk contaminant concentrations. Similarly,
air-borne particle and vapor-phase contaminants originating from the off-site
become sources for pasture grass and fodder concentrations, which are
above ground vegetations. As described below, exposure site soil
concentrations are a function both of the amount of soil estimated to erode
from the off-site contamination, and of a mixing depth which is different for
"tilled" vs. "untilled" situations. The soil concentration used for cattle
ingestion of soil is assumed to be untilled. The soil concentration used to
estimate concentrations in underground vegetables is assumed to be tilled.
The algorithm to estimate fish tissue concentrations as a function of bottom
sediment concentrations remains the same for this source category.
The soil ingestion and dermal exposure pathways are still a function of exposure
site soil concentrations; i.e., no assumption of direct contact with the off-site
contamination is made. Also, both of these direct soil exposure pathways used the
untilled soil concentrations.
Section 5.4.1. discusses how exposure site soil concentrations are calculated from
off-site concentrations. Section 5.4.2. describes adjustments to the volatilization flux
algorithm and the far-field dispersion model which transports vapor and particulate-phase
5-55 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
residues to the nearby exposure site.
5.4.1. Exposure Site Soil Concentrations
The key assumptions for the solution strategy estimating exposure site soil
concentrations resulting from an off-site soil contamination source are: 1} the exposure
site soil becomes contaminated due to erosion of contaminated soil from the source to the
exposure site, 2) the amount of soil at the exposure site does not increase, which means
that soil delivered to the site via erosion is matched by an equal amount which leaves the
site, and 3) not only does soil erode off the contaminated site en route to the exposure
site, but soil between the contaminated site and the exposure site also erodes to the
exposure site.
The second and third assumption translate to:
Hi + D2 = R (5-26)
where:
DT = mass of soil delivered from off-site contaminated source, kg
D2 = mass of soil delivered from land area between contaminated source
and exposure site, kg
R = mass of soil removed from exposure site, kg.
The mass balance equation for exposure site soil concentrations can now be
qualitatively stated as (with "AC" used as shorthand for change in exposure site soil
concentration over time):
(the incremental addition to C resulting from the change in erosion of
contaminated soil) *
AC = (the incremental substraction of C resulting from removal of now
contaminated soil from the exposure site) *
(the incremental substraction of C resulting from dissipation of
residues at the exposure site)
5-56 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
This can be expressed mathematically as:
dC = Dj Co _ RC _ k (5_27)
dt M M
where:
C = the exposure site soil concentration, mg/kg
D1 = mass of soil delivered from off-site contaminated source, kg/yr
C0 = concentration of contaminant at contaminated site, mg/kg
M = mass of soil at exposure site into which contaminant mixes, kg
R = mass of soil removed from exposure site, kg/yr
k = first order dissipation rate constant, 1/yr.
Assuming that the contaminant concentration at the exposure site, C, is initially 0,
Equation (5-27) can be solved to yield:
C = Dl C° ll-e * I (5"28)
£ +
which computes C as a function of time, t (in years since k is in years). This can be
solved for various increments of time starting from a time when the exposure or
contaminated site initially became contaminated, or it can be simplistically assumed that
the contamination has existed at the contaminated site for a reasonably large amount of
time such that the exponential term approaches zero. This can be alternately stated that
the assumption is made that the system has reached steady state over a long period of
time. Either way, the exponential term drops out, and C is estimated as:
5-57 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
C = Dl°° (5-29)
R + kM
Equation (5-29) was used to estimate exposure site concentrations resulting from off-site
contamination for the example scenario in Chapter 9. Guidance for estimation of these
terms including justification for their values as selected in the example settings are:
• k: For the on-site source category, and for contaminated soil at the off-site
contaminated location, the assumption is made that residues do not degrade or dissipate
to the point of reducing the concentration of the "initial" soil levels. This was partly based
on information indicating generally low rates of biological or chemical degradation for the
dioxin-like compounds of this assessment, coupled with the assumption that on-site and
off-site contamination was sufficiently deep implying a reservoir of contaminant that would
remain available during a period of exposure. These assumptions are less likely to be valid
for residues which have migrated over the surface to deposit on the exposure site. The
deposition is likely to result in only a thin layer of contaminated soil. Though very small,
surface-related dissipation mechanisms such as photolysis, volatilization, or degradation,
might reduce surface soil contaminant concentrations. For these reasons, a "dissipation"
rate constant is assumed to apply to delivered contaminant, where the precise
mechanisms of dissipation are not specified, but could include transport (volatilization,
erosion) and degradation (photolysis, biodegradation) mechanisms. The studies on
2,3,7,8-TCDD described in Young (1983) imply a dissipation half-life of 10 years. Fries
and Paustenbach (1990) suggested the use of a half-life of at least 10 years, and used a
15 year half-life in their example scenarios on the impact of air-borne deposition of
2,3,7,8-TCDD originating from stack emissions. This assessment uses a dissipation
half-life of 10 years for all of the three example compounds in Chapter 9. This half-life
translates to a first-order dissipation rate constant of 0.0693 yr"1.
• M: The delivered contaminant mixes to a shallow depth at the exposure
site. The mixing depth depends on activities which disturb the surface, such as
construction, plowing, vehicle traffic, movement of cattle or other animals, burrowing
5-58 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
action of animals, other biological activity, normal leaching, and raindrop splash. Mixing
depths for fallout plutonium have been found to be 20 cm on cultivated land and 5 cm on
uncultivated forest and rangeland (Foster and Hakonson, 1987). Fries and Paustenbach
(1990) suggested a depth of 15 cm for agricultural tillage, but assumed values of 1 and 2
cm for various sensitivity tests. However, they did not need to make a distinction
between tilled and untilled situation because vegetation (pasture grass and forage for
estimating beef and milk fat concentration; above ground fruits and vegetables for human
consumption) was assumed to be impacted only by particulate deposition and not root
uptake. In another assessment on indirect impacts from incinerator emissions, EPA
(1990) estimated vegetation concentrations as a function of particulate depositions, root
uptake, and air-to-leaf transfer from the vapor phase. Different mixing depths for untilled
and tilled concentration estimation was required. For root uptake estimation for vegetable
and other crops, the estimated soil concentrations assuming a tillage mixing depth of 20
cm. For soil concentrations in untilled situations, they assumed a mixing depth of 1 cm.
The methodology of this assessment uses 1 cm for the untilled and 20 cm for the tilled
conditions. Soil concentrations for calculation of concentrations of underground
vegetables will be a function of a 20-cm depth. This assumption is made because tilling
gardens is assumed to distribute surface residues to the 20-cm depth. Soil concentrations
for dermal contact, soil ingestion, and pasture grass and soil intake for cattle grazing will
assume a depth that has been used for the untilled situation, that of 1 cm. These
activities are assumed to occur on soil which has not been tilled. These assumptions will
be used for both of the situations where the source of contamination is off-site and
residues get delivered to the exposure site: the off-site as well as the stack emission
source categories.
Given the area of the exposure site, the mass of soil into which the eroded
contaminant is mixed can be calculated as:
M = Aes ED d (5-30)
where:
5-59 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
M = mass of soil for contaminant mixing per unit depth, kg/m
Aes = area of exposure site, m2
BD = soil bulk density, kg/m3
d = depth of mixing, m
• D1 and D2: The first step in deriving both these amounts of soil is to use
the Universal Soil Loss Equation (USLE). This approach was described above.
Justification was given for an assumption of unit soil loss from the contaminated site of
62 t/ac-yr, based on USLE calculations and survey data from 70 landfills nationally. D.,
equals this unit loss times the area of contamination times a sediment delivery ratio. The
example scenario in Chapter 9 assumed that the exposure site was 1 50 meters from the
contaminated site, and using Equation (5-5), the sediment delivery ratio is 0.26. The unit
loss assumed for the area between the contaminated site and the exposure site is 6
t/ac-yr. Since this area is adjacent to the exposure site, there is no sediment delivery, and
D2 equals this unit loss times the area between the contaminated and exposure sites.
D2 and D2 can now be expressed as:
D} = 0.224 SL1 SD1 ALS (5-31a)
£>2 = 0.224 SL2 SD2 ABLE (5-31b)
where:
DT 2 = mass of soil delivered from off-site contaminated source, Dv and
from the land area between contaminated source and exposure site,
D2, kg/yr,
SL1 2 — average annual unit soil loss, Eng. tons/acre-year, equal to 62 t/ac-yr
for SL-| and 6 t/ac-yr for SL2
SD1 2 = sediment delivery ratio, unitless, 0.26 for SD-, (with distance = 150
meters) and 1.00 for SD2,
5-60 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
ALS/ABLE = land area of contaminated site, ALS, and of area between
contaminated site and exposure site, ABLE, m2
.224 = converts t/ac-yr to kg/m2-yr.
An adjustment is made to the sediment delivery ratio, SD1( considering the size
discrepancies between the contaminated site and the exposure site. For example, if the
contaminated site is larger than the exposure site, then the amount of eroded soil delivered
150 meters downgradient would not all mix with soil at the exposure site. On the other
hand, if the contaminated site were smaller than the exposure site, than the full amount of
eroded soil delivered 1 50 meters downgradient would be contained within the exposure
site. A simple correction factor, equaling the ratio of a side length of the exposure site
(assumed square-shaped) and a side length of the contaminated site size (also assumed
square shaped), is used to adjust the sediment delivery ratio SD1:
SDla = SDl CF (5-32)
where:
SD1a = adjusted sediment delivery ratio corresponding to SDV unitless
SD1 = sediment delivery ratio reducing the amount eroding from the
contaminated site to be delivered to the exposure site, unitless
CF = AES°-5/ALS°-5 if AES < ALS
1 if AES > ALS
AES = area of exposure site, m2
ALS = area of contaminated site, m2
Similar considerations are pertinent to the land area between the contaminated and
exposure site. Remember that the algorithm assumed that some "clean" (D2) and some
"contaminated" soil (D.,) erodes onto the exposure site, and that a similar amount of soil
entering the exposure site (R, which equals D-\ + D2) leaves the exposure site so as to
maintain a mass balance. The amount of clean soil eroding from upgradient sources
5-61 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
mixing with exposure site soil can be larger than the amount of contaminated soil if the
exposure site is larger than the contaminated site. If the exposure site is smaller than the
contaminated, and similar to the solution for SD1a above, then only the small corridor
defined by the size of the exposure site contributes clean soil. Either way (i.e., the
exposure site is larger or smaller than the contaminated site), the size of the land area
contributing clean soil is defined by the size of the exposure site. ABLE can be estimated
as the product of the distance between the exposure and contaminated site, and the side
length of the exposure site:
ABLE = DL SL (5-33)
where:
ABLE = 'anc* area between contaminated and exposure site, m2
DL = distance from landfill to exposure site, m
SL = side length of exposure site, m
= °-5
AES = area of exposure site, m2.
5.4.2. Vapor-phase Transport
In the case of bare soil contamination or soil contamination with a grass cover
where the contamination originates at the surface, the contaminant flux can be estimated
as given in Section 5.3.2, Equation (5-8). If the contaminated soil is instead covered with
contaminant-free soil, a "clean cap" situation, the solution is more complicated. A
rigorous approach is detailed in Hwang, et al. (1986). Use of this approach requires a
computer to iteratively solve a partial differential equation, expressed in terms of a Fourier
series. It can be shown, with these equations, that the vapor emission rate through such
a cover will not reach steady state for hundreds of years. For this assessment, a
computer program was written, and calculations performed for 2,3,7,8-TCDD
contamination with a thickness of contamination of 8 ft, and clean caps ranging from 10
to 25 cm. The results of this exercise suggest that the average emission rate of a 70-year
period are 1/4 to 1/5 of what they would be without the cap. Based on this exercise, a
5-62 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
simple assumption can be made that a clean cap will reduce the average emission rate
calculated without a clean cap by 80%. Obviously, this assumption should be viewed
with caution. The industrial site and the ash landfill example settings in Chapter 9 did not
have a clean cap, so this adjustment was not made.
Estimating the dispersion and resulting exposure site concentrations requires a
different solution for the off-site as compared to the on-site situation. A simplified
solution, given as a virtual point source model, can be found in Turner (1970). This model
approximates the dispersion that occurs from an area source by using an imaginary point
source. This point is located upwind of the actual source at a distance calculated to
create a lateral dispersion at the site equal to its width:
2.03 FLUX Afr FREQ 10
10
VD
U
m
(5-34)
where:
Cair =
FLUX =
Msc
FREQ
VD
10
10 _
concentration of contaminant in air, //g/m3
Average contaminant volatilization flux rate, calculated with or
without a clean cap, g/cm2-s
area of contaminated site, m2
frequency wind blows from source to receptor, unitless
virtual distance, source center to receptor, m
vertical dispersion coefficient, m
average wind speed, m/s
converts g/cm2 to //g/m2.
The term, FREQ, has been added to this equation to appropriately account for
changing wind directions, and hence, obtain a more accurate annual average air
concentration. The vertical dispersion, Sz, is estimated as an empirical function of the
distance from the source center to receptor:
5-63
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
S, = 0.222 x0-725 - 1.7 , x < 1000 m
(5-35a)
S. = 1.260 x-0'516 - 13.0 , X > 1000 m
(5-35b)
where:
Sz = vertical dispersion coefficient, m
X = actual distance from source center to receptor, m.
The virtual distance, VD, is an empirical function of the width of the contaminated
area and the actual distance from source center to receptor:
VD = 2.514 a + x
(5-36)
where:
VD
a
X
virtual distance, source center to receptor, m
width of contaminated area perpendicular to wind direction - defined
previously as side length for assumed square-shaped contaminated
area, m
actual distance from source center to receptor, m.
Prior guidance on windspeed (Section 5.3.2) indicated windspeeds ranged from 2.8
to 6.3 m/sec, and suggested a mid-range of 4.0 m/sec in the absence of better
information. Where the wind blows from all directions equally, then it will blow from one
compass sector about 1 5% of the time. On these bases, a windspeed of 4.0 m/sec and a
FREQ of 0.15 were used in the example scenarios in Chapter 9. In most places, however,
wind direction is much less variable, and the appropriate value is best determined with site
5-64
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
specific information.
5.5. ALGORITHMS FOR THE INCINERATOR STACK EMISSION SOURCE CATEGORY
Contaminants emitted from incinerator stacks are transported in air and deposit on
the exposure site and on surrounding land. Chapter 6 describes the application of the
Industrial Source Complex (ISC; EPA, 1986} air transport model to obtain vapor-phase air
concentrations and deposition rates of particles at a specified distance from an example
incinerator. These quantities are assumed to be given for purposes of discussion in this
section; further discussion of the ISC model application is given in Chapter 6.
Table 5-1 summarized the fate and transport algorithms for this source category.
Estimating soil concentrations based on particulate depositions follows a similar approach
as estimating exposure site soil concentrations resulting from erosion of contaminated soil
from off-site areas of contamination. Section 5.5.1. describes how soil concentrations are
estimated given total (wet plus dry) deposition rates. Following now are bullet summaries
for similarities and small refinements to algorithms previously discussed:
• Surface water impacts: Stack emission depositions over a watershed will
result in an average watershed soil concentration, similar to the term Cw of
Equation (5-1). The suspended and bottom sediment concentrations will be
equal to this average concentration, which is a simplification of Equation
(5-1) when all soil in the watershed is represented by one concentration.
Section 5.5.1. shows how soil concentrations are estimated given deposition
rates of contaminated particulates. Two key values are required for
estimating average watershed soil and sediment concentration given
deposition: a representative deposition rate and a representative depth of
contaminant mixing. When using the ISC model in a site-specific application
to estimate a representative deposition rate, isopleths of particulate
deposition rate should be overlain on a watershed map. An average
deposition rate can be estimated as an area-weighted average from this
overlay. The depth of mixing might consider land use within the watershed.
For a watershed predominantly undeveloped, a representative mixing depth
of 1 cm, used for the "non-tilled" assumption for estimating mixing depth,
5-65 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
might be appropriate, or if the watershed is predominantly agricultural, a
20-cm mixing depth would be appropriate. For the example scenarios in
Chapter 9 demonstrating the incinerator scenarios, the following
assumptions were made for deposition rate and mixing depth. Table 6-18
and 6-19 show deposition rates at various distances up to 5 km from the
incinerator stack for the example incinerator. Deposition rates at 5 km will
be used to estimate the average watershed sediment concentration. The
watershed is within a watershed with agricultural and non-agricultural
settings. As such, the representative mixing depth will be 10 cm.
Particle-phase air-borne concentrations: There is no air-borne particulate
phase exposure. ISC model outputs include air-borne vapor phase
concentrations (actually the hypothesis is made that what is called "vapor
phase" is actually "aerosol phase" contamination, or contamination on
particles less than 1 /mi in diameter; see Chapter 6) and deposition rate of
contaminants on particulates. Particulates are not assumed to remain
air-borne for inhalation exposures. This, of course, is different from the
assumption made for on-site and off-site soil source categories, where < 10
//m size particulates are suspended due to wind erosion and are available for
particulate inhalation exposures. That algorithm was characterized as
conservative as the soil was described as containing an unlimited reservoir
of particles available for wind erosion. Also, on-site and off-site soil
contamination is assumed to be relatively deep, which makes wind erosion a
more plausible phenomena for those source categories.
Biota concentrations: The algorithm estimating concentration in fish tissue
based on bottom sediment concentrations is the same as in previous source
categories. Underground vegetable concentrations are a function of "tilled"
soil concentrations. The soil concentration used for cattle soil ingestion is
"untilled". Untilled and tilled depths of mixing are 1 and 20 cm,
respectively, as in the off-site source category. Air-to-plant contributions are
a function of ISC modeled vapor phase concentrations. ISC outputs also
include the rate of particulate deposition for that portion of vegetation
concentrations. Beef and milk concentrations are again a function of
5-66 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
vegetative and soil concentrations, diet fractions, and bioconcentration and
bioavailability factors as described in Section 5.3.5.
5.5.1. Steady-State Soil Concentrations
Chapter 6 describes the use of the ISC Model to estimate the particulate phase
deposition rate at the exposure site. This deposition rate, F, includes both dry and wet
deposition, and is used to estimate the steady state soil concentrations. The deposition of
contaminated particulates from the air is assumed to be somewhat analogous to the
process of eroding contaminated soil from an off-site source depositing on an exposure
site. Specifically, the following assumptions are also made: 1) only a thin layer of soil
becomes contaminated, 2) this layer is either 1 cm or 20 cm deep, depending of surface
activities, and 3) surface residues are assumed to dissipate with a half-life of 10 years
corresponding to a first order decay rate of 0.0693 yr"1. Considerations of upgradient
erosion and exposure site soil removal are not made. Depositions occur over the exposure
site and surrounding land area on an on-going basis. It might be said that upgradient soil
concentrations are similar to exposure site concentrations at all times. Contributions of
upgradient erosion to exposure site soil concentrations are matched by an equal amount of
downgradient erosion removal of contaminants. The qualitative mass balance statement
(similar to the one made above in Section 5.4.1, with AC equalling change in exposure site
soil concentrations over time) can now be made as:
(the incremental addition to C resulting from the change in
deposition of stack emitted particulates) *
AC = (the incremental substraction of C resulting from
degradation of residues at the exposure site)
This is mathematically stated as:
If • |-*c (5-37)
5-67 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
where:
C = the exposure site soil concentration, mg/kg
F = deposition rate of contaminant on particles, mg/yr
M = mass of soil at exposure site into which contaminant mixes, kg
k = first order dissipation rate constant, 1/yr.
The solution to this equation is:
= -L ( 1 - e-kt ) (5-38)
which computes C as function of time, t. Similar to the assumption made above in
Section 5.4.1., the steady state solution for C is simply F/kM. The deposition rates
supplied by the ISC model are in units of g/m2-yr, so a conversion to mg/yr requires a
multiplication by the land area of the exposure site and a multiplication of 1000 mg/g.
Procedures to estimate M are given above in Section 5.4.1. The assumptions made for
the mixing zone depth, d, described in Section 5.4.1, as also appropriate for particle
deposition impact on soil for the stack emission source category.
5.6. ALGORITHMS FOR THE ASH DISPOSAL IN LANDFILLS SOURCE CATEGORY
The algorithms to estimate exposure media concentrations in this source category
are the same as those described in Section 5.4, the off-site source category, with three
exceptions. One is that the size, or potential size, of the landfill can be a function of the
amount of ash disposed. Simple considerations for estimating this landfill size are
described in Section 5.6.1. The second is that the transport, unloading, and spreading and
compacting of the ash at the landfill are three processes that contribute to the particulate
flux that gets transported by air to the exposure site. The transport process results in two
distinct emissions. One is vehicular resuspension of roadway particulates; roads and other
surfaces over which vehicles travel at a landfill can become contaminated over time by the
contaminated soil of the landfill. Since ash is being transported in trucks to the ash
5-68 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
disposal site, another potential transport related emission occurs directly off the trucks.
These fugitive ash emissions into air are modeled used "AP-42" emission factor equations
(EPA, 1988), which are described in Section 5.6.2 below. The other component of the
total paniculate flux, described in Sections 5.3.3 above, is due to wind erosion.
Discussions in Section 5.4 about application of the wind erosion algorithm to off-site soil
contamination are also relevant for this section. The third difference between this source
category and the off-site source category is in the estimation of concentrations of soil/ash
in the landfill. Chapter 6 discusses the estimation of the dioxin-like compounds on fly ash.
For the example setting, it will be assumed that the ash mixes with landfill soil and that
final concentrations are half the concentrations for the ash.
5.6.1. Landfill Size
A size of a contaminated landfill can be assumed, or a simple estimate can be made
considering the amount of ash transported to the landfill. An estimate of a landfill size
receiving ash from the 2727 TPD incinerator is made in this section.
The first consideration is the amount of ash generated. It is typical to combine
bottom and fly ash for disposal, and this is the assumption made for the example landfill in
Chapter 9. Chapter 6 justifies a total fly ash generation for a plant of this capacity
employing fabric filters with semi-dry alkaline scrubbers to be 30 metric tons/day, or
30000 kg/day, and estimates that bottom ash is ten times this quantity, or 300000
kg/day. Total daily ash generation is therefore 330000 kg/day. Next is required an
assumption of ash bulk density. Since ash is comprised of finer particles than soil in
general, a bulk density lower than that of soil is appropriate. The average bulk density of
ash varies between .9-1.3 g/cm3 (EPA, 1991 a; soil bulk densities range from 1.2-1.8
g/cm3). A reasonable assumption might therefore be an ash bulk density of 1.1 g/cm3, or
1100 kg/m3. With a daily amount of ash estimated, a bulk density, and an assumption on
the total number of days over which disposal occurs, one can estimate the total landfill
volume required. This volume can be converted to an area if an assumption on ultimate
depth of disposal is made. The example assessments in Chapter 9 assumed a depth of 3
meters.
A thirty-year landfill life, with continuous ash disposal (365 d/yr, 30 yrs) to a depth
of 3 m results in landfill sizes of 110 hectares for a 2727 TPD facility, as follows:
5-69 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
330,0001^/3) (365d/yr) (30 yrs) = 1(J9.5 hectares
(10,000 m2/ha) (3m) (1,100 kg/m3)
It will be assumed that one-fourth of that size, or 27 hectares, is active at all time during
the 30 year landfill lifetime. The other 83 hectares are either not filled or completed and
covered. Other refinements might be considered. It may be unlikely for a single landfill to
receive all the ash generated. Further, it may be unlikely that a 27 hectare area is
uncovered at all times. The concentration of the dioxin-like compounds in the ash are
described in Chapter 9 and were based on information provided in Chapter 6.
5.6.2. Fugitive Particulate Emissions at the Landfill
Four additional processes contribute to the total flux of particulates at the landfill.
These include vehicular resuspension of roadway dust, the emission from trucks
transporting the ash to the landfill, the unloading at the landfill, and the spreading and
compacting operations.
5.6.2.1. Vehicular Traffic Over Landfill Roadways
A particulate flux is introduced in situations involving vehicular traffic over
roadways at or near the soil contamination. Particulates on these roadways, paved or
unpaved, can become contaminated over time by movement of particulates from the
contaminated soil surface - by wind erosion, soil erosion, tire track-off, and so on. While
the concentration of contaminants on these particulates is considerably less than the
concentration on the contaminated soil itself (because of mixing with clean particulates,
ongoing migration away from roadways, and so on), the flux of contaminant can be
significant because of much higher particulate emission rates for vehicular traffic as
compared to a wind erosion flux.
Estimates of particulate emission rates from paved and unpaved roadways can be
estimated with AP-42 emission factor equations (EPA, 1988). The ash landfill example
setting in Chapter 9 contains roadways that are unpaved. This is assumed because more
emissions occur over unpaved roadways. Use of both emission factor equations for
5-70 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
roadway emissions is demonstrated in EPA (1991 a). The equation for unpaved roadways
is:
E =17* 1S\1VS\1 W \-/™\-- (5_3g)
*up J--/ Kunp -T2-4s2.7 4 365 '
where:
E = emission flux for unpaved surfaces, kg/VKt (VKt equals vehicle
kilometer traveled)
kunp = particle size multiplier specific to the unpaved road emission flux
equation, unitless
s = silt content of unpaved roadway, %
Vs = vehicle speed, km/hr
W = vehicle weight, kg
nw = number of wheels per vehicle, unitless
P = number of days with at least 0.254 mm (0.01 inch) precipitation per
year, unitless.
Critical information from EPA (1988) includes different values for the particle size
multiplier, kunp, depending on the paniculate size of concern, and ranges of other
parameters (number of vehicle wheels, for example) for which the equations are relevant.
Guidance for parameter selection specific to a landfill setting is provided below.
There have been a few key applications of AP-42 roadway emission factor
equations (Kellermeyer and Zeimer, 1989; MRI, 1990) to evaluate the potential impact of
suspended particulates in landfills. One in particular, conducted by EPA's Office of Solid
Waste (MRI, 1990), not only applied the equations but also took a limited number of
roadway samples to determine, among other factors: realistic measurements of the
parameter s, roadway length measurements and vehicle counts to estimate the number of
vehicle kilometers travelled, and other relevant information using a questionnaire survey
given to the landfill disposal operator. The portion of their work used in the discussions
below was conducted on 46 disposal sites accepting ash from municipal solid waste
5-71 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
(MSW) combustors. A portion of these disposal sites were industrial solid waste (ISW)
landfills. Following now is brief guidance on these factors specific to a landfill setting.
• Particle size multiplier, kunp: Values for unpaved roads were developed for
particle sizes 2.5 //m and less, kunp = 0.095, up to particle sizes 30 //m and
less, kunp = 0.80 (EPA, 1988). The value corresponding to particle sizes 10
fjm and less, kunp = 0.36, is the appropriate selection if the application is to
estimate the flux of inhalable size particulates. If the objective is to estimate
total flux for purposes such as upwind deposition onto soil, plants, or water,
then the highest kunp value, kunp = 0.80, would be more appropriate. The
value used in the example settings of Chapter 9 was 0.36, since the
objective was to estimate the flux of particulates which are then transported
downwind for inhalation exposures.
• Silt content of unpaved roadways, s: MRI (1990) took only two samples
from haul routes on unpaved roads in two landfills, and obtained values of
6.7 and 20.1%. The mean of these values, 13.4%, is used in the example
settings. EPA (1988) gives an appropriate range of 3.8-15.1% for this
parameter.
• Vehicle speed, Vs: Based on their observations and questionnaires, MRI
(1990) assumed a vehicle speed of 24 km/hr (15 mph) in their assessment,
which was also used for this assessment. EPA (1988) gives an appropriate
range for this parameter of 21-64 km/hr.
• Vehicle weight, W: A range of vehicle weights for vehicles disposing
materials in industrial solid waste landfills was given as 14-40 x 103 kg in
MRI (1990). The midpoint of this range, 27 x 103 kg, was used in
Chapter 9. EPA (1988) gives an appropriate range for this parameter as 2.7-
142x 103 kg.
• Number of wheels, nw: MRI (1990) noted a range for this parameter of
between 6 and 14 wheels. An average of 10 wheels per vehicle was used
in Chapter 9. EPA (1988) gives an appropriate range for this parameter of
4-1 3 wheels.
• Number of days with precipitation > 0.254 mm, P: EPA (1988) presents a
5-72 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
map of the United States showing the number of days with greater than this
amount of rain. Regional values range from: 1 50-170 days for the
Northeast, 110-120 days for the Southeast, 70-110 days for the Mid and
Southwest, 30-40 days for the Southern West Coast, and 90-1 50. The
example setting in Chapter 9 assumed a value for P of 121, a mid-range
value equalling 1/3 of a year.
After determining a value for Eup, the next step is to determine the total number of
vehicle kilometers traveled per time period. This involves making assumptions for the
length of impacted roadway (how much roadway has particulates that are contaminated),
and the number of "vehicle passes" over that roadway during a specified time period.
Again, based on the study of MRI (1990):
• Length of impacted roadway: The median haui route length from 46 landfills
accepting MSW ash was 402 meters, although the mean length was 614
meters with the highest noted of 4828 meters. It may be reasonable to
assume that only a portion of the haul route has contaminated particulates.
One way to estimate an appropriate length would be to multiply the total
haul route length (perhaps 400-600 m without better information) by a ratio
of the area of contaminated landfill soil to the entire area of the landfill. If
the entire landfill has contaminated soil, than the entire length of the haul
route would have contaminated particulates. Another consideration is that
higher capacity landfills have longer roadways than lower capacity landfills.
The example settings in Chapter 9 assumes an impacted haul route of 100
m, with no particular justification.
• Number of vehicle passes: MRI (1990) determined that the number of
vehicle "transactions", number of vehicles entering the 46 landfill sites in
order to conduct business per day, excluding those specifically associated
with ash disposal (it was not explained why ash transactions were excluded
from their presented data; it was noted that vehicles other than those
associated with ash comprised the large majority of transactions in industrial
solid waste landfills), on the average was 116, although the median was 26.
One observation they made was that the higher capacity landfills, the
highest having > 1600 vehicle transactions, had paved roads, while those
5-73 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
having lower capacities had unpaved roads. An assumption of 100 vehicle
transactions over unpaved haul routes might be a reasonable assumption. A
"vehicle pass" estimate could then equal vehicle transactions * 2, which
would assume that each vehicle travels over the contaminated roadway
going in and coming out. However, if one assumption is that only a portion
of the roadway is contaminated (as discussed above), then it would be
consistent to assume that not all vehicle transactions occur over the portion
which is contaminated. A more reasonable assumption, perhaps, is that half
the vehicles drive over impacted roadways; this would equate vehicle passes
to vehicle transactions. The example settings in Chapter 9 assume 100
vehicle passes/day, or 50 vehicle transactions occurring per day driving over
the impacted roadway twice.
There are two more considerations to note before summarizing the strategy for
estimating dust flux from vehicular traffic over landfill roadways. First is how many days
per year landfills operate. A continually operating landfill, 365 days/year, is not an
unreasonable assumption (MRI, 1990). Second is the consideration of a "control
efficiency factor." The AP-42 emission factor equations were developed with no controls
in place. It is common for landfills to use dust suppression measures on their roadways,
particularly for unpaved roadways where visibility can be an issue. Common means of
dust suppression in landfills are wetting and chemical dust suppression (MRI, 1990). An
assumption of a 90% reduction in total emissions possible because of dust suppression
measures is probably reasonable, given the observations made in MRI (1990). Landfills
operating 365 days/year and a control efficiency factor of 0.10 (90% efficiency) were
assumed for the example settings of Chapter 9.
In summary, the strategy for estimating roadway vehicular dust emissions is as
follows: estimate Eup from Equation (5-47), estimate total vehicular kilometers per day
considering length of impacted roadway and vehicle passes per day, multiply Eup by total
kilometers passes per day to get total dust flux per day (a good steady state assumption
given 365 operating days per year), and finally reduce this total by a control efficiency
factor. The resulting dust flux will be called Ev/ and will be in units of kg/day.
The total contaminant flux is then equal to Ev times a concentration of contaminant
on the dust, which can be assumed to be a linear function of the contaminant
5-74 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
concentration in the landfill soil:
VE = 1.2 X 10"11 Cs CDF Ev (5-40)
where:
VE = total contaminant flux due to vehicular traffic on landfill
roadways, g/sec
Cs = contaminant concentration on landfill soil, ppb or//g/kg
CDF = contaminant dilution factor, unitless
Ev = roadway dust emission rate, kg/day
1.2x10~11 = converts//g/day to g/sec.
Again very little information is available on CDF. However, MRI (1990) did take
some measurements which are applicable. Briefly, they took particulate samples from
landfill haul routes while at the same time taking samples of incinerator ash being delivered
for disposal the same day. Each paired sample (roadway particulate and ash), were
measured for four metals: As, Cd, Cr, and Pb. Several paired samples were taken on both
paved and unpaved haul routes. Ratios were then generated for roadway particulate metal
concentrations over ash metal concentrations. Results were: As - paved and unpaved
ratios were similar and consistently near 0.1 (roadside particulate concentrations of As
were 10% of ash concentrations of As), Cd - paved and unpaved ratios were similar and
ranged between 0.0 and 0.4, Cr - paved ratios ranged from 0.3 to 0.6, while unpaved had
a wide range of 0.3 to 2.0, Pb - paved and unpaved ratios were similar between 0.0 and
0.2. For analogous situations - daily deliveries of contaminated ash - one might assume a
CDF in the 0.1-0.2 range. A value of 0.1 for CDF was used in the ash landfill example
setting in Chapter 9.
5.6.2.2. Fugitive Emissions from Trucks
The following equation for estimating emissions from open storage piles has been
suggested for use in estimating fugitive emissions from trucks in transit (EPA, 1991 a; the
5-75 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
emission factor equation from EPA, 1985a):
E<"
where:
Et = particulates emitted from trucks in transit, kg/day/hectare
s = silt content material of ash, %
P = number of days with >0.25 mm precipitation per year
f = percentage of time that the unobstructed wind speed exceeds 5.4 m/s.
Brief guidance on these parameters is:
• Silt content of ash, s: EPA (1991) lists the particle size distribution of
combined ash, and lists 6.7% as less than 63 //m. The USDA describes silt-
size particles to be between .002 and .05 mm (.05 mm = 50 jjm; Brady,
1984), with clay-size particles less than .002. Therefore, the silt content of
ash will be assumed to be 6.7% for the example setting in Chapter 9.
• Number of days with precipitation > .25 mm per year, P: Background on
this parameter was provided in Section 5.6.2.1 above, and as noted there,
an assumption of 121 days was a reasonable mid-continent assumption.
• Percentage of time unobstructed wind speed exceeds 5.4 m/sec: Since this
speed equals just over 12 miles per hour, it is probably reasonable to assume
that trucks transporting the ash travel over this speed 100% of the time.
Hence, a value of 100% is assumed for this parameter.
As this equation was developed for stationary piles rather than moving trucks, a
few additional steps are required.
Step 1. Estimate daily number of truck loads transported. This requires an estimate of
truck capacity, which is then divided into the total load delivered per day. The average
volume of a truck delivering ash has been estimated as 40 yd3 (Wells, et al., 1988). The
5-76 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
example setting in Chapter 9 used a volume of 40 m3. With a density of 1100 kg/m3, the
average truck can haul 44,000 kg/trip. The 2727 TPD incinerator generating 330,000
kg/day might therefore send out 8 trucks per day.
Step 2. Estimate truck surface area. Wells, et al. (1988) assumed a depth of 1 yd for
their estimate of 40 yd3. The example setting in Chapter 9 had an assumption of 40 m2
surface area.
Step 3. Estimate total daily emission of ash. This first requires a conversion of Et units of
kg/day/hectare to more reasonable units for this application, which might be kg/min/m2;
that conversion factor is 6.94. Then, an assumption of the length of roadway over which
this emission occurs needs to be made. For simplicity. Chapter 9 assumed that this
emission only occurred within the landfill. Section 5.6.2.1 described the survey of 46
landfills accepting ash from municipal solid waste combustor facilities; the mean haul route
was 614 meters, and the setting in Chapter 9 used a length of 600 meters. If the truck
travels at 15 miles/hr, or 390 m/min, than it takes 1.5 minutes of travel within the landfill.
The emission rate in kg/day for the ash landfill can now be estimated as 8 trips/day * 40
m2 * 1.5 min * 6.94 (kg/min/m2)/(kg/day/ha) * Et kg/day/ha (Et as calculated with
Equation (5-41) above).
Fugitive emissions from trucks carrying ash can be minimized by wetting the ash or
the use of truck covers. The application of this equation to a situation where the ash is
properly wetted prior to transport would show that no fugitive emissions of dust are
expected to occur (i.e., a proper assumption would be that P = 365, hence Et = 0). Another
management control is the use of tarpaulins or other truck coverings. The above equation
was developed for wind erosion estimation without covering. Therefore, a control
efficiency is appropriate to use with this equation. The example setting in Chapter 9
assumed a 90% efficiency; i.e., multiply the final daily emission from the above exercises
by 0.10.
5.6.2.3. Emissions from Unloading
The unloading operations at the disposal site may result in the release of fugitive
dust. The following emission factor equation provides emission factors for kilograms of
5-77 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
particulate emitted per megagram (metric ton, or 1000 kg) of soil loaded and unloaded
(EPA, 1988):
Elu = 0.0016 kunl C^)1'3 (f T1-4 (5-42)
where:
Eju = emission factor for loading and unloading, kg fugitive dust/MT ash
kun| = particle size multiplier, dimensionless
Um = wind speed, m/s
M = material moisture content, %.
Guidance on these parameters is:
• Particle size multiplier, kun,: Values were supplied in EPA (1988) for sizes
ranging from <2.5/;m, kun) = 0.11, to sizes <30/vm, kun, = 0.74. The
value corresponding to particle sizes 10 //m and less, kun) = 0.35, is the
appropriate selection if the application is to estimate the flux of inhalable size
particulates, as it is for the example setting in Chapter 9.
• Wind speed, Um: A wind speed of 4 m/sec is used for the wind erosion
algorithm (Section 5.4.1), and was justified as being an average value. For
the loading and unloading emission factor, the range of wind speed for
which is assumed to be generally valid is 0.6 to 6.7 m/s (EPA, 1988).
• Moisture content, M: The moisture content of ash after exiting the quench
tank at combustor facilities can be as high as 40% (EPA, 1991 a), although
ash would not be transported with this high a moisture content.
Furthermore, fugitive emissions should not be considered with higher
moisture content ash. This emission factor equation was developed for a
range of moisture contents of 0.25-4.8% (EPA, 1988). The conservative
assumption of a moisture content of 0.25% was made for the example
setting.
5-78 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
The Elu factor still needs to be multiplied by the number of MT to obtain total
emissions in kg/day. The 2727 TPD facility produces 330 MT/day ash, so the
multiplication factor is 330.
5.6.2.4. Emissions from Spreading and Compacting
Fugitive dust emissions from spreading and compacting ash at disposal sites have
been estimated in more than one.way, although the different ways found in the literature
to estimate emissions from these processes are all based on AP-42 emission factor
equations. The differences were due to assumptions as to which processes the spreading
and compacting of ash was most analogous to, and used the AP-42 factor developed for
that process. MRI (1990) used an AP-42 emission factor developed for dozer moving of
overburden in western surface coal mines. Kellermeyer and Ziemer (1989) assumed that
the spreading and compaction of ash was analogous to vehicular transport on unpaved
surfaces, and used the emission factor for that process. A third possible assumption, and
the one used for this assessment, is that the processes of spreading and compacting are
analogous to agricultural tillage. That emission factor equation for agricultural tillage is
(EPA, 1988):
Eat = 5.38 kat s°'6 (5-43)
where:
Eat = emission factor for agricultural tillage, kg/ha
kat = particle size multiplier, dimensionless
s = silt content, %.
Guidance for these parameters is:
• Particle size multiplier, kat: Values were supplied in EPA (1988) for sizes
ranging from <2.5//m, kat = 0.10, to sizes <30/;m, kat = 0.33, with a
final size listed as "total suspended sediment" and given a kat of 1.00. The
value corresponding to particle sizes 10//m and less, kat = 0.21, is the
5-79 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
appropriate selection (and the selection made for the example setting in
Chapter 9) if the application is to estimate the flux of inhalable size
particulates.
• Silt content, s: This equation was developed for silt contents between 1 .7
and 88.0%. As noted above, the silt content for the example setting for ash
was 6.7%.
A final consideration for application of the agricultural tillage emission factor
equation is the number of hectares per day over which spreading and compacting occur.
A simple way to estimate this area is to assume a depth of spreading, and calculate area
of coverage based on total daily amount spread. The example setting in Chapter 9
assumed a .05 m (2 in) depth of spreading each day. Eight full truckloads of 40 m3 were
emptied each trip. At a depth of 0.05 m, 6400 m2 (320/.05) or .64 ha were required.
The total daily emission from spreading and compacting can now be estimated as: 1
operation/day * .64 ha * Eat kg/ha.
5.6.2.5. Total Particulate Emissions from Ash Landfills
A few details must be remembered for these emission factor equations. First, all
total particulate emissions need to eventually be in appropriate units of g/cm2-sec, which
are unit emissions over an area where these emissions occur. That area is assumed to be
the area of the ash landfill, the parameter termed ACS which earlier was shown to be
equal to 27 ha.
Emissions off trucks, Et, emissions from unloading, E|U, and emissions from
spreading and compacting, Eat, were uniformly converted to kg/day, so these results need
to be further multiplied by:
Emissions (g/CTO> day) = E (^/^Y) (1000 7 /kg) (1 day/8640 sec)
Asc (ha) (108 cm2 /ha)
where:
E = emission factors Et, E|u, Eat (kg/ha)
A0^ = area of soil contamination (ha)
5-80 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
The final emissions then need to be multiplied by a unitless (g/g) concentration of the
dioxin-like compound in the ash to get contaminant emission flux appropriate for air
transport modeling. Multiplication by a unitless concentration is necessary so that the
total contaminant emissions are still in g/cm2-sec.
A fourth contaminant flux, that due to resuspension of roadway particulates, was
described above as VE (Section 5.6.2.1) and was already converted to units of g/sec.
This was done in order to facilitate the discussion on the contaminant dilution factor
(CDF). To bring this flux in line with the others, it must be divided by the total landfill area
in cm2.
The final contribution to particulate emissions is that due to wind erosion. The
wind erosion flux is given as Ee, estimated by Equation (5-14) with refinements for bare,
off-site soil contamination as discussed in the introductory portion of Section 5.4. This
flux term is solved for in units of g/sec, so it as well must be divided by the total landfill
area in cm2.
Total flux of contaminant on particulates can now be given as Ee + VE + Et + E|U
+ Egt, all in units of g/cm2-sec. The final step is to set this total equal to flux factor,
FLUX, of Equation (5-34). That equation now will estimate far-field dispersion and
resulting exposure site air-borne particulate concentrations.
5-81 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
REFERENCES FOR CHAPTER 5
Baes, C.F.; Sharp, R.; Sjoreen, A.; Shor, R. (1984) A Review and Analysis of Parameters
for Assessing Transport of Environmentally Released Radionuclides Through
Agriculture. Oak Ridge National Laboratory, ORNL-5786.
Bacci, E.; Calamari, D.; Gaggi, C.; Vighi, M. (1990) Bioconcentration of Organic Chemical
Vapors in Plant Leaves: Experimental Measurements and Correlation. Environ. Sci.
Technol. 24: 885-889.
Bacci, E.; M.J. Cerejeira; C. Gaggi; G. Chemello; D. Calamari; M. Vighi (1992)
Chlorinated Dioxins: Volatilization from Soils and Bioconcentration in Plant Leaves.
Bulletin of Environmental Contamination and Toxicology (in press).
BOS (1990) Battelle Ocean Services. New Bedford Harbor Modeling Program Final
Report. U.S. Environmental Protection Agency, Boston, MA.
Brady, N.C. (1984) The Nature and Properties of Soils. Ninth Edition. New York, NY:
Macmillan.
Briggs, G.G., R.H. Bromilow, and A.A. Evans (1982) Relationships between lipophilicity
and root uptake and translocation of non-ionised chemicals by barley. Pesticide
Science 13: 495-504.
Carslaw, H.S.; Jaeger, J.C. (1959) Conduction of Heat in Solids. Oxford University
Press, Oxford, England.
Columbo, J.C., M.F. Khahl, M. Arnac, A.C. Horth (1990). Distribution of Chlorinated
Pesticides and Individual Polychlorinated Biphenyls in Biotic and Abiotic
Compartments of the rio de la Plata, Argentina, Environmental Science and
Technology 24: 498-505.
Connett, P.; Webster, T. (1987) An Estimation of the Relative Human Exposure to
2,3,7,8-TCDD Emissions Via Inhalation and Ingestion of Cow's Milk. Chemosphere
16 (8/9): 2079-2084.
Cook, P.M.; Duehl, D.W.; Walker, M.K.; Peterson, R.E. (1991) Bioaccumulation and
Toxicity of TCDD and Related Compounds in Aquatic Ecosystems, p. 143-168 in
Gallo, M.A., R.J. Scheuplein, and K.A. Van Der Heijden, eds., Banbury Report 35:
Biological Basis for Risk Assessment of Dioxins and Related Compounds. Cold
Spring Harbor Laboratory Press 0-87969-235-9/91.
Forrester, K.E. (1988) Comparison of Municipal Solid Waste Leachate to MSW Ash
Residual Leachate using a Risk Algorithm. Wheelabrator Environmental Systems,
Inc., Danvers, MA.
5-82 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Foster, G.R.; Hakonson, T.E. (1987) Erosion Losses of Fallout Plutonium. In: Dynamics
of Transuranics and Other Radionuclides in Natural Environments. Howard, W.A.,
and R. Fuller, eds. NTIS-87014456. Springfield, VA.
Foth, H.D. (1978) Fundamentals of Soil Science. 6th ed. New York: John Wiley and
Sons.
Gobas, F.A.P.C. (1990) Bioaccumulation of some polychlorinated dibenzo-p-dioxins and
octachlorodibenzofuran in the guppy (Poecilia reticulata). Chemosphere 20: 495.
Fries, G.F.; Paustenbach, D.J. (1990) Evaluation of Potential Transmission of 2,3,7,8-
Tetrachlorodibenzo-p-dioxin-Contaminated Incinerator Emissions to Humans Via
Foods. J. Toxicol. Environ. Health 29: 1-43.
Gillette, D.A. (1981) Production of dust that may be carried great distances. In: T.
Pewe, ed. Desert Dust: Origin, Characteristics, and Effect on Man. Geological
Society of America Special Paper 186, pp. 11-26.
Hwang, S.T. (1987) Methods for estimating on-site ambient air concentrations at
disposal sites. Nucl. Chem. Waste Management 7: 95-98.
Hwang, S.T.; Falco, J.W.; Nauman, C.H. (1986) Development of Advisory Levels for
Polychlorinated Biphenyls (PCBs) Cleanup. U.S. Environmental Protection Agency,
Exposure Assessment Group, Office of Research and Development. EPA/600/6-
86/002.
Jackson, D.R., M.H. Roulier, H.M. Grotta, S.W. Rust, and J.S. Warner (1986) Solubility
of 2,3,7,8-TCDD in Contaminated Soils, pp. 185-200 in Rappe, C., G. Choudhary,
and L.H. Keith (eds.), Chlorinated Dioxins and Dibenzofurans in Perspective. Lewis
Publishers, Inc.
Karickhoff, S.W., D.S. Brown, and T.A. Scott (1979) Soprtion of hydrophobic pollutants
on natural sediments. Water Research 13: 241-248.
Kellermeyer, D.A.; Ziemer, S.E. (1989) What are the health risks of ash monofills? Solid
Waste and Power lll(4): 40-44.
Kjeller, L.O., S.E. Kulp, S. Bergek, M. Bostrom, P.A. Bergquist, C.Rappe, B.Jonsson, D.de
Wit, B. Jansson, M. Olson (1990) Levels and Possible Sources of PCDD/PCDF in
Sediment and Pike Samples from Swedish Lakes and Rivers (Part One).
Chemosphere 20: 1489-1496.
Knisel, W.G., ed. (1980) CREAMS: A Field Scale Model for Chemicals, Runoff, and
Erosion from Agricultural Management Systems. US Department of Agriculture,
Conservation Research Report No. 26, Tuscon, Arizona.
5-83 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Kuehl, D.W.; Cook, P.M.; Batterman, A.R.; Lothenbach, D.B.; Butterworth, B.C. (1987)
Bioavailability of polychlorinated dibenzo-p-dioxins and dibenzofurans from
contaminated Wisconsin river sediment to carp. Chemosphere 16: 667.
Lake, J.L., N.I. Rubinstein, H. Lee II, C.A. Lake, J. Heltshe and S. Pavignano. 1990.
Equilibrium partitioning and bioaccumulation of sediment-associated contaminants
by infaunal organisms. Environ. Toxicol. Chem. 9: 1095-1106.
Liss, P.S.; Slater, P.G. (1974) Flux of gases across the air-sea interface. Nature 247:
181-184.
Lyman, W.J.; Reehl, W.F.; Rosenblatt, D.H. (1982) Handbook of Chemical Property
Estimation Methods. New York: Mcgraw-Hill.
McCrady, J.K., McFarlane, C.; Gander, L.K. The transport and fate of 2,3,7,8-TCDD in
soybean and corn. Chemosphere 21: 359-376.
Mills, W.B.; Porcella, D.B.; Ungs, M.J.; Gherini, S.A.; Summers, K.V.; Mok, L.; Rupp, G.L.;
Bowie, G.L.; Haith, D.A. (1985) Water Quality Assessment: A Screening
Procedure for Toxic and Conventional Pollutants in Surface and Ground Water -
Parts I and II. EPA/600/6-85/002a&b.
MRI, 1990. Special Management Standards for Municipal Waste Combustion (MWC) Ash.
Submitted by: Midwest Research Institute, 425 Volker Boulevard, Kansas City, MO
64110-2299, to US EPA, Municipal Solid Waste Program. EPA Contract No. 68-
01-7287. June 29, 1990.
Muir, D.C.G.; Yarechewski, A.L.; Webster, G.R.B. (1986) Bioconcentration for four
chlorinated dioxins by rainbow trout and fathead minnows. Environ. Toxicol.
Chem. 5: 261.
Niimi,A.J.; B.C. Oliver. (1989) Distribution of Polychlorinated Biphenyl Congeners and
Other Halocarbons in Whole Fish and Muscle from Lake Ontario Salmonids.
Environmental Science and Technology 23: 83-88.
Novotny, V.; Chesters, G. (1981) Handbook of Nonpoint Pollution, Sources and
Management. New York: Van Nostrant Reinhold.
Office of Technology Assessment. (1989) Facing America's Trash: What's Next for
Municipal Solid Waste? U.S. Congress, Office of Technology Assessment.
Washington, D.C. OTA-0-424.
Parkerton, T.F., J.P. Connolly, R.V. Thomann, C.G. Uchrin. (1992) Do Aquatic Effects or
Human Health Endpoints Govern Sediment Quality Criteria for Non-ionic Organic
Chemicals? Regulatory Implications of the Equilibrium Partitioning Paradigm. To
appear in: Environmental Toxicology and Chemistry.
5-84 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Parkerton, T.F. (1991) Development of a Generic Bioenergetics-Based Model for
Predicting the Bioaccumulation of Persistent Sediment-Associated Contaminants -
Final Report, New Jersey Department of Environmental Protection, Trenton, NJ.
Riederer, M. (1990) Estimating Partitioning and Transport of Organic Chemicals in the
Foliage/Atmosphere System: Discussion of a Fugacity-Based Model. Environmental
Science and Technology 24: 829-837.
SAIC. (1986) Science Applications International Corp. Conversion factors for use in
estimating environmental transport of contaminated soil from unregulated disposal
sites. Prepared for the U.S. Environmental Protection Agency, Office of Solid
Waste, Washington, DC, under EPA contract 68-01-7624.
Schroy, J.M.; Hileman, F.D.; Cheng, S.C. (1985) Physical/chemical properties of 2,3,7,8-
TCDD. Chemosphere 14: 877-880.
Shor, R.W.; Baes, F.F.; Sharp, R.D. (1982) Agricultural Production in the United States
by County: A Compilation of Information from the 1974 Census of Agriculture for
Use in Terrestrial Food-Chain Transport and Assessment Models. ORNL-5768, Oak
Ridge National Laboratory, Oak Ridge, Tenn.
Seinfeld, J.H. (1986) Atmospheric Chemistry and Physics of Air Pollution. John Wiley
and Sons, New York.
Stevens, J.B.; Gerbec, E.N. (1988) Dioxin in the agricultural food chain. Risk Analysis
8(3): 329-335.
Swackhammer, D.L.; R.A. Hites (1988). Occurrence and Bioaccumulation of
Organochlorines in Fishes from Siskiwit Lake, Isle Royale, Lake Superior.
Environmental Science and Technology 22: 543-548.
Swackhammer, D.L., B.D. McVeety, R.A. Hites (1988). Deposition and Evaporation of
Polychlorinated Congeners to and from Siskiwit Lake, Isle Royale, Lake Superior.
Environmental Science and Technology 22: 664-672.
Thomann, R.V., J.P. Connoly and T.F. Parkerton. 1992. An equilibrium model of organic
chemical accumulation in aquatic foodwebs with sediment interaction.
Environmental Toxicology and Chemistry. In Press.
Travis, C.C.; Arms, A.D. (1988) Bioconcentration of organics in beef, milk, and
vegetation. Environmental Science and Technology 22(3): 271-274.
Travis, C.C.; Hattermeyer-Frey, H.A. (1991) Human exposure to dioxin. Sci. Total
Environ. 104: 97-127.
Travis, C.C.; Hattermeyer-Frey, H.A. (1988) Uptake of Organics by Aerial Plant Parts: A
Call for Research. Chemosphere 17: 277-283.
5-85 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Turner, D.B. (1970) Workbook of Atmospheric dispersion estimates. PHS publication no.
999-AP-26 (NTIS PB 191482), EPA, Research Triangle Park, North Carolina.
U.S. Department of Agriculture. (1974) United States Department of Agriculture.
Universal Soil Loss Equation. Agronomy Technical note no. 32. Portland, Oregon.
U.S. Soil Conservation Service. West Technical Service Center.
U.S. Environmental Protection Agency. (1977) United States Environmental Protection
Agency. Water Quality Assessment: A Screening Method for Nondesignated 208
Areas. US EPA # EPA-600/9-77-023. J. Falco, Project Officer, US EPA, Athens
ERL, Athens, GA 30605. Authors: Zison, S.W., K.F. Have, and W.B. Mills; Tetra
Tech, Inc., Lafeyette, California.
U.S. Environmental Protection Agency. (1985a) Compilation of Air Pollutant Emission
Factors. Fourth Edition. U.S. Environmental Protection Agency, Office of Air
Quality, Planning and Standards. Research Triangle Park, NC.
U.S. Environmental Protection Agency. (1985b) Rapid Assessment of Exposure to
Paniculate Emissions from Surface Contamination Sites. Environmental Protection
Agency, Office of Health and Environmental Assessment, Office of Research and
Development. EPA/600/8-85/002, February, 1985.
U.S. Environmental Protection Agency. (1986) Industrial Source Complex (ISC)
Dispersion Model User's Guide. Second Edition Vol. 1. Office of Air Quality
Planning and Standards, Research Triangle Park, NC, EPA-450/4-86-005a. NTIS
PB86-234254.
U.S. Environmental Protection Agency. (1988) Compilation of Air Pollutant Emission
Factors. Volume 1. Stationary Point and Area Sources. Fourth Edition.
Supplement B. U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards. Research Triangle Park, NC. September, 1988.
U.S. Environmental Protection Agency. (1990a) Methodology for Assessing Health Risks
Associated with Indirect Exposure to Combustor Emissions. U.S. Environmental
Protection Agency, Office of Health and Environmental Assessment. Washington,
D.C. EPA/600/6-90/003. January, 1990.
U.S. Environmental Protection Agency. (1990b) Lake Ontario TCDD Bioaccumulation
Study Final Report. Cooperative study including US EPA, New York State
Department of Environmental Conservation, New York State Department of Health,
and Occidental Chemical Corporation. May 1990.
U.S. Environmental Protection Agency. (1991a) Methodology for Assessing
Environmental Releases of and Exposure to Municipal Solid Waste Combustor
Residuals. Exposure Assessment Group, Office of Health and Environmental
Assessment, EPA, Washington, D.C. EPA/600/8-91/031. April, 1991.
5-86 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
U.S. Environmental Protection Agency. (1991b) Methodology for Assessing Exposure and
Risk to PCBs in Soils and Sediments - A Screening Approach. Exposure Assessment
Group, Office of Health and Environmental Assessment, EPA, Washington, D.C.
(August 1991 draft report).
Vanoni, V.A. (ed.). (1975) Sedimentation Engineering. American Society of Civil
Engineers, New York, NY.
Wells, R.C.; Machaver, R.J.; Allen, W.R. (1988) Fugitive Particulate Emission Estimates
for Incinerator Ash Handling. Camp Dresser & McKee, Inc, Edison, NJ.
Wischmeier, W.H. (1972) Estimating the cover and management factor on undisturbed
areas. Proceedings of the USDA Sediment Yield Workshop. Oxford, MS: U.S.
Department of Agriculture.
Wischmeier, W.H.; Smith, D.D. (1965) Predicting Rainfall-Erosion Losses from Cropland
East of the Rocky Mountains, Agriculture Handbook 282. U.S. Department of
Agriculture, Agriculture Research Service.
Young, A.L. (1983) Long term studies on the persistence and movement of TCDD in a
national ecosystem. In: Tucker, et al., eds. Human and environmental risks of
chlorinated dioxins and related compounds. New York, NY: Plenum Publishing.
5-87 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
6. MUNICIPAL SOLID WASTE INCINERATION
6.1. INTRODUCTION
The combustion of municipal solid waste (MSW) releases potentially harmful
pollutants to the air from the smokestack of the incinerator. Of particular concern is the
potential human health and environmental effects of these emissions during the period of
facility operation. The rapid growth in this stationary combustion source category,
coupled with the de novo synthesis and stack emissions of PCDDs, PCDFs and related
compounds, provides an opportunity to evaluate multiple pathways of human exposure of
the emissions. The following chapter provides procedures for estimating the emission
rates of dioxin-like compounds from municipal waste incinerators and the associated
deposition rates and air concentrations as a function of distance from the incinerator.
Once the air concentrations and deposition rates have been established, the reader can
then use the procedures provided in other Chapters of this document to determine possible
exposure pathways and levels of exposure. These procedures are illustrated using a
hypothetical MSW incinerator arbitrarily located in Tampa, Florida. This example exposure
scenario is not intended as a regulatory exposure/risk assessment, since it involves a
theoretical incinerator, but provides an example of an exposure assessment of the
emissions of dioxin-like compounds that can be followed and applied in a similar manner to
the evaluation of actual emissions from a planned or operational facility. This Chapter
reviews the current MSW incineration technology in the U.S.; discusses the theoretical
basis for explaining why dioxin-like compounds are emitted during MSW combustion;
describes the hypothetical MSW incinerator selected for analysis of direct and indirect
exposures to dioxin-like compounds emitted from the stack; summarizes current emissions
data of specific congeners of PCDD, PCDF, and coplanar PCBs appropriate to the
hypothetical MSW incinerator: describes the air dispersion and surface deposition modeling
of the chemicals; and gives the results appropriate for the analysis of direct and indirect
human exposure of ambient air concentrations and surface deposition flux of the dioxin-
like congeners.
The per capita generation of municipal solid waste in the United States is
approximately 1.8 kilograms (kg) per day (US EPA, 1990). This equates to a 50%
increase from the 1.2 kg/person in 1960 (U.S. EPA, 1989). As a result, nearly 164 million
6-1 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
metric tons of MSW are generated each year, and the rate is continuing to increase.
Currently the breakdown of methods for disposing of the MSW waste stream in the U.S. is
as follows: 73% is landfilled; 13% is recycled; 14% is incinerated (U.S. EPA,1990). This
is in marked contrast to the breakdown of 1985: 84% MSW was landfilled; 12% was
recycled, and 4% was incinerated (U.S. EPA, 1987a).
Faced with increasing problems of land disposal, many communities are
constructing incinerators as the primary method for waste disposal. These systems are
designed for the recovery of heat from refuse combustion in the form of steam or hot
water that can be used as an energy source to generate electricity, to supplement energy
demands of industry, and for use in district heating of residential and commercial
properties. Although the combustion of MSW does not eliminate the need for landfilling,
since residual ash must be disposed of, it does reduce the volume of waste requiring
landfilling by 70 to 90 percent, and therefore extends the operational life of existing
landfills. Without a reduction in the volume of waste that ultimately is landfilled, some
urban areas will soon reach the design capacity of existing landfills and be compelled to
select some method of MSW disposal or face a possible waste disposal crisis.
Significant growth in the population of municipal incinerators has occurred in the
U.S. since 1985. In 1985 approximately 43,240 metric tons per day (MT/d) of MSW was
incinerated in 99 facilities nationwide (U.S. EPA, 1987a). In 1989, approximately 63,670
MT/d of MSW were disposed of in an estimated 200 municipal incineration facilities
(U.S.EPA, 1989). This represents a 200 % increase in the number of facilities, and a 47 %
increase in tons of MSW incinerated within a span of four years. If trends continue, then
it is estimated that by the year 2000, a total of 400 incineration facilities may be
incinerating 33% of the MSW waste stream per day (U.S. EPA, 1987a).
6.2. OVERVIEW OF PRINCIPAL MUNICIPAL INCINERATION TECHNOLOGIES IN THE
UNITED STATES
Municipal incinerators operating in the U.S. can be classified into four general
design categories: mass burn, modular, refuse-derive fuel (RDF), and fluidized-bed (U.S.
EPA, 1987a; OTA,1989; Cleverly,1991). The first type is called mass burn, because the
waste is combusted without any preprocessing other than removal of items too large to go
through the feed system. In a typical mass burn combustor, refuse is placed on a grate
6-2 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
that moves through the combustion chamber. Combustion air in excess of stoichiometric
amounts is supplied both below (underfire air) and above (overfire air) the grate. Mass
burn incinerators are usually field-erected and can burn 45 to 909 metric tons of MSW per
day (MT/d) per incineration unit. Many mass burn facilities have two or more incineration
units comprising the incineration facility with a total refuse combustion capacity ranging
from 45 to 2727 MT/d.
Modular incinerators also burn waste without preprocessing, but they are typically
factory fabricated and shipped in reassembled sections to the intended site. Modular
incinerators generally incinerate 4.5 to 91 metric tons of refuse per day per incineration
unit. One of the most common types of modular incinerators is the starved air or
controlled air type incorporating two combustion chambers. Air is supplied to the primary
chamber at sub-stoichiometric levels. The incomplete combustion products entrained in
the combustion gases from the primary combustion chamber pass into the secondary
combustion chamber where excess air is added, and combustion is completed. Another
type of modular combustor, functionally similar to larger mass burn units, uses excess air
in the primary chamber; no additional air is added in the secondary chamber. Modular
facilities are also comprised of one or more incineration units having a total combustion
capacity of 4.5 to 270 MT/d of MSW.
The third general type of MSW incinerator is designed to burn RDF. RDF is a term
describing MSW which has been processed into a physical form that theoretically
enhances combustibility. RDF is a fuel that has been prepared by shredding, sorting,
separating, and trommeling MSW to reduce the noncombustible content (Cleverly, 1991).
It may vary in physical form from shredded waste having a uniform size to a finely divided
fuel suitable for co-firing with pulverized coal. Some cases of RDF processing involves
magnetic separation of ferrous metals and reduction of other noncombustible components
of the refuse, such as aluminum and glass products. RDF facilities have a total
combustion capacity of 227 to 3636 MT/d.
The fourth type of MSW incineration technology is the fluidized-bed design. In this
design, the waste burns in a turbulent bed of non-combustible material, usually sand. The
MSW may be fed into the incinerator either as unprocessed waste, or as a form of RDF.
There are two basic design concepts to the fluidized-bed technology: a bubbling-bed
incineration unit, and a circulating-bed incineration unit (U.S. EPA, 1989). Fluidized-bed
6-3 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
MSW incinerator units are typically designed to burn from 182 to 455 MT/d per unit, and
the total incineration facility capacity by combining one or more incineration units is in the
range of 184 to 920 MT/d.
Existing MSW incinerators can be further classified as having or not having heat
recovery boilers. All RDF and fluidized bed incinerators have boilers to recover heat. Over
66% of existing mass burn incinerators are equipped with heat recovery boilers. Modular
units are usually absent heat recovery capability because of size limitations.
There is a distinct distribution of incinerator technologies (U.S. EPA, 1987a). This
distribution can be determined as a percentage of the total mass of MSW that is
incinerated annually in the U.S. On this basis, the approximate distribution by technology
is as follows: 58% mass burn; 24% RDF; 6% modular; 12% fluidized-bed
-------
DRAFT-DO NOT QUOTE OR CITE
molecules composing MSW and the formation of combustion products proceed in long
sequences of reactions, and each sequence involves a rearrangement of chemical bonds.
Heat activates the collision of molecules, and energy increases molecular collisions
disassociating molecules into fragmentary atoms that react with other molecules or
radicals yielding new intermediate combustion products. The intermediate products
collide, become dissociated into free radicals which sets up the process of creating
intermediates. The repetition of the same molecular events is termed chain reactions, and
the atoms or radicals that cause the chain reactions are termed reaction centers. It may
require a relatively small number of chain centers to lead to a large occurrence of chemical
reactions. The randomness of these molecular events makes it difficult, if not improbable,
to predict chemical reactions and the formation of chemical intermediates during MSW
combustion with any absolute certainty. It is possible, however, to identify the central
chemical events in the formation of chlorinated dioxins and dibenzofurans through
laboratory experiments under simulated incineration conditions.
Various theories have been proposed to explain the presence of PCDDs and PCDFs
in municipal solid waste incinerator emissions (Cleverly, 1991; Vogg, 1987;
Gullett,1990,1991a,b; Bruce, 1991; Commoner, 1987; Choudhry, 1983; ASME, 1981;
U.S. EPA, 1987b). These theories include one or more of the following:(1) these
compounds are present as contaminants in bleached paper or other constituents of MSW,
and a portion of them survives the thermal stress imposed by the incineration process; (2)
they result from the de novo synthesis from precursors, such as polychlorinated biphenyls
(PCBs), chlorophenols (CPs), and chlorinated benzenes (CBs); and (3) they are synthesized
from materials not chemically related to PCDDs and PCDFs such as petroleum products,
polynuclear aromatic hydrocarbons (PAH), inorganic chloride ions, and plastics.
It is postulated that PCDDs and PCDFs are created on the reactive surface of fly
ash (paniculate matter) entrained in the combustion plasma downstream of the furnace
zone in regions where the temperature of the combustion offgases have cooled to between
200° and 400° Celsius (Vogg,1987; Bruce,1991; Cleverly,1991; Gullet,1990;
Commoner,1987). Through a series of laboratory experiments, Vogg and coworkers
(1987) have postulated that inorganic chloride ions, such as copper chloride, present in the
combustion gas may act as a catalyst to promote surface reactions on particulate matter
to convert aromatic compounds to chlorinated dioxins and dibenzofurans. Isotopically
6-5 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
labeled MSW incinerator fly ash was heated in a laboratory furnace purged by an air flow
for two hours. The treated fly ash was exposed to increasing temperatures in 50° C
increments in a temperature range of 200° to 400° Celsius. Table 6-1 summarizes these
data.
Vogg (1987) concluded that formation of PCDDs and PCDFs on the surfaces of fly
ash during MSW incineration probably occurs in a temperature window defined as a range
of 200° to 350° Celsius. Vogg (1987) proposed an oxidation reaction pathway giving rise
to the formation of PCDDs and PCDFs in the post furnace regions of the incinerator:
(1) Hydrogen chloride gas (HCI) is thermolytically formed as a product from the
combustion of heterogeneous refuse containing abundant chlorinated organic chemicals
and chlorides; (2) oxidation of HCI via the Deacon process, with copper chloride (CuCI2) as
a catalyst, yields free radical chlorine, (3) Phenolic compounds (present from combustion
of lignin in the waste) entrained in the combustion plasma are substituted on the ring
structure by contact with free radical chlorine; (4) the chlorinated precursor to dioxin, e.g.
chlorophenol, is oxidized with copper chloride as a catalyst to form PCDDs and PCDFs and
chlorine. Gullett and coworkers (1990, 1991) have studied the formation mechanisms
through extensive combustion research, and have concluded the PCDDs and PCDFs can be
formed from low temperature reactions (350° C) between CI2 and phenolic precursors
forming a chlorinated precursor followed by oxidation reactions of chlorinated precursors
involving copper chloride as a catalyst as in examples (1) and (2).
(1) The initial step in the formation of dioxin is the formation of chlorine from HCI in
the presence of oxygen, as follows (Vogg, 1987):
A
2HCI + 1/2 O2 > H2O + CI2
(2) PCDD formation from dioxin precursor on fly ash surfaces catalyzed by copper
chloride (Vogg, 1987: Gullett, 1990):
CuCI2
2-chlorophenol + 1/2 02 > dioxin + CI2
6-6 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-1. Concentration of PCDD/PCDF (ng/g) on municipal incinerator fly ash at
varying temperatures
Temperature (° Celsius)
PCDD
Tetra
Penta
Hexa
Hepta
Octa
PCDF
Tetra
Penta
Hexa
Hepta
Octa
200°
15
40
65
100
90
122
129
61
48
12
250°
26
110
217
208
147
560
367
236
195
74
300°
188
517
1029
1103
483
1379
1256
944
689
171
350°
220
590
550
430
200
1185
1010
680
428
72
400°
50
135
110
60
15
530
687
260
112
12
Source: Adapted from Vogg et al., 1987.
The major direct source of chlorine available for participating in the formation of
PCDDs/PCDFs is gas-phase HCI, which is initially formed as a combustion product from
the chlorine and chlorinated organic chemicals contained in the MSW (Vogg, 1987; Bruce,
1991, Cleverly, 1984; Commoner, 1987). MSW contains approximately 0.66 percent (by
weight) chlorine (Trinklein, 1982). MSW incinerators are a major stationary combustion
source of air emissions of HCI, which averages between 400 to 600 ppmv in the
combustion gas (U.S. EPA,1987c). HCI is converted to chlorine by the Deacon process,
and the free radical chlorine directly chlorinates a dioxin precursor along the aromatic ring
structure. Oxidation of the chlorinated precursor in the presence of an alkali inorganic
chloride ion catalyst (of which copper chloride is the most active catalyst) yields PCDDs
and PCDFs as intermediate reaction products. Increasing the free radical chlorine
concentration in the combustion gas generally causes an increase in the rate of formation
of PCDDs/PCDFs. Formation kinetics are most favored at temperatures between 200° to
350° Celsius. Any reduction in chlorine formation from HCI or decreases in combustion
6-7 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
gas temperatures below the window of formation will result in decreases in the rate and
magnitude of formation of PCDDs and PCDFs (Bruce,1991, Gullett,1990;
Commoner,! 987). In the testing of a variety of industrial stationary combustion sources
during the National Dioxin Study in 1987, EPA made a series of qualitative observations on
the relationship between total chlorides present in the fuel and the magnitude of emissions
of PCDDs and PCDFs (U.S.EPA,1987d). In general, combustion units with the highest
PCDD emission concentrations contained significant quantities of chlorine in the feed, and,
conversely, sites with the lowest PCDD emission concentrations contained only trace
quantities of chlorine in the feed.
The evaluation of the mass balance of PCDD and PCDF emissions from various
MSW technologies supports the observations made from these experimental data of post-
furnace, low-temperature formation. Commoner and coworkers (1989) evaluated the test
data of a mass burn MSW incinerator for the concentration of PCDDs and PCDFs at
multiple sampling points during the combustion process: (1) at the exit to the furnace;
(2) at the entry to the heat exchanger; (3) inlet to the electrostatic precipitator (ESP);
(4) exit to the ESP; (5) exit to the smokestack. Lowest, or non-detectable concentrations
of PCDDs/ PCDFs were found at sampling point (1), and highest concentrations were
measured at sampling point (5). Commoner concluded that dioxins were not substantially
formed within the furnace region, but were formed in areas downstream of the combustion
zone where the combustion offgases had cooled to less than 400° C. This phenomena
was independently observed by Environment Canada in an earlier test of a modular MSW
incinerator (Hay,1986; Environment Canada,1985). On a mass balance basis, the
concentration of PCDDs and PCDFs measured at the smokestack were approximately two
orders of magnitude higher as compared to the inlet to the boiler just after exiting the
secondary furnace. The temperatures of the combustion gases at these two points of
measurement were 130° C and 740° C at the stack and boiler inlet, respectively
(Environment Canada, 1985). For the most part,only Octa-CDD was present in the hot
gases exiting the furnace, whereas the all the congeners were present in the stack
emissions.
In a series of tests of an RDF facility conducted jointly by U.S. EPA and
Environment Canada (U.S. EPA, 1991), approximately 5 milligrams of total PCDD and
PCDF per metric ton of MSW were measured in the refuse prior to combustion, but no
6-8 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
PCDDs nor PCDFs were detected at the exit to the furnace prior to the inlet to the
economizer (heat exchanger used to extract additional heat from the hot gases). Once
heat in the combustion gas was extracted for energy purposes, and the gases were further
cooled to within the "window" of temperatures that promote dioxin formation on fly ash
surfaces, then the total array of PCDDs and PCDFs could be detected. These series of
experiments in which the mass balance of PCDD/PCDF was estimated for the entire
combustion process, including the MSW, discounts the likelihood of the first theory of
dioxin emissions: that dioxin in the MSW accounts for dioxin emissions at the stack.
The air emission of polychlorinated biphenyls (PCBs) from MSW incinerators is less
understood. There are virtually no theories explaining the detection of these compounds in
MSW incinerator emissions. However PCBs have been measured in the raw refuse prior to
incineration (Choudhry, 1983, U.S. EPA, 1991). The mass balance of total PCB beginning
with measurement in the raw refuse and ending with measurement at the stack to an RDF
MSW incinerator (U.S. EPA, 1991) can be used to calculate the destruction rated
efficiency (ORE) of incineration of the PCB contaminated MSW. Using results from test
number 11 at the RDF facility (U.S. EPA, 1991), a computation of DRE can be made with
the following equation (Brunner,1984):
DRE = 100 I^B ^ I (6-1)
["Wjn - "out!
W-
L m J
where:
Win = mass rate of contaminant fed into the incinerator system.
Wout = mass rate of contaminant exiting the incinerator system.
In test eleven, 811 nanograms total PCBs/gram of refuse (ng/g) was measured in the MSW
fed into the incineration system, and 9.52 ng/g of total PCB was measured inlet to the
pollution control device (e.g., outside the furnace region, but preceding emission control).
From this datum a DRE of 98.8 percent can be calculated. This level of combustion
efficiency corresponds to typical MSW incineration (Brunner, 1984). Therefore it appears
that PCB contamination in the raw MSW that is fed into the incinerator may account for
the emission of PCBs from the stack of the MSW incinerator. This is unlike the
6-9 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
mechanism of formation of PCDDs and PDCFs in that the existing empirical data tends not
to support the de novo synthesis of PCBs.
PCBs can be thermolytically converted into PCDFs (Choudhrly, 1983; U.S. EPA,
1984). This process occurs at temperatures somewhat lower than typically measured
inside the firebox. Laboratory experiments conducted by EPA (U.S. EPA, 1984) indicate
that the optimum conditions for PCDF formation from PCBs are near a temperature of
675° C in the presence of 8% oxygen and a residence time of 0.8 seconds. This resulted
in a 3 - 4 % conversion efficiency of PCBs to PCDFs. Since 1- 2 % of the PCBs present in
the raw refuse may survive the thermal stress imposed in the combustion zone to the
incinerator ( U.S. EPA, 1991), then it is reasonable to presume that PCBs in the MSW may
contribute to the total mass of PCDF emissions released from the stack of the incinerator.
6.4. HYPOTHETICAL MSW INCINERATOR FOR PURPOSES OF EXPOSURE ANALYSIS
For purposes of illustrating the procedures proposed in this chapter, a specific
incinerator design and set of environmental conditions must be chosen. Accordingly a
hypothetical,but realistic, facility and location were selected. The mass burn, heat
recovery incinerator technology was selected to provide an appropriate example. This
technology dominates the current and projected population of incinerators in the U.S. The
hypothetical incinerator is assumed to have a combined daily combustion capacity of 2727
metric tons per day (3000 tons/d), which represents the upper 75th percentile of the
distribution of existing and projected mass burn, heat recovery facilities in the U.S. (US.
EPA, 1989). Table 6-2 summarizes the distribution of existing and projected mass burn
heat recovery MSW incinerators by size of facility. The projected facilities are to the year
1995 (U.S. EPA, 1989). If all MSW incinerator technologies are counted, e.g., refuse
derived fuel, fluidized bed, mass burn nonheat recovery, and modular incinerator, then EPA
estimates there will be 272 MSW incineration facilities operating in 1995 (U.S. EPA,
1989).
The EPA selected the stack height, diameter, exit velocity of the gaseous emissions
from the stack, and temperature of the exhaust gases based on actual measurements at
facilities of this size. The hypothetical facility is located, for modeling purposes, in a wet
and humid climate in a geographical area where the MSW incineration industry is expected
to grow rapidly over the next 15 years (U.S. EPA,1987). This site is Tampa, Florida. The
6-10 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-2. Capacity distribution (TDP) of existing and projected mass burn,
heat recovery MSW incinerators in the United States
Mass burn
technology
MBWW
MBREF
MBRCWW
MBRCWW
MBWW
MBWW
TOTAL
Facility
capacity3
200
500
500
1050
1080
2250
Number of
facilities'5
28
3
5
3
16
18
73
Percent of
total
38.4
4.1
6.8
4.1
22.0
24.7
MBWW = mass burn waterwall
MBREF = mass burn refractory
MBRCWW = mass burn rotary combustor, waterwall
aCapacity is total tons MSW incinerated each day.
bThe number of facilities includes only mass burn, heat recovery incinerators. This counts
existing facilities, and facilities EPA estimates will be operational by 1994. Although this
totals 73 facilities, if all MSW incinerator technologies, both heat and nonheat recovery,
are included, the sum is 273 facilities.
Source: Adapted from U.S. EPA, 1989, pp 4-5 and 4-6.
climate conditions were selected to maximize the potential of surface deposition of the
emitted dioxin-like compounds with the physical action of washout during periods of
precipitation. The site receives in excess of 122 centimeters of rainfall per year,and is
classified as a humid subtropical climate. Table 6-3 summarizes the modeling parameters
for the hypothetical MSW incinerator.
6-11
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-3. Modeling parameters for the hypothetical MSW incinerator
i—Smokestack
^H
Mass Burn
incineration Facility
•
Pollution
Control
Device
Technology: Mass burn, heat recovery MSW incinerator
Rate of MSW combustion: 2727 metric tons per day (3000 tons/d)
Location: Tampa, Florida
Latitude: 27° 57'
Longitude: 82° 27'
Stack height: 46 meters (m)
Stack diameter: 4.1 m
Exit velocity of stack gases: 11.3 m/s
Pollution control scenarios: (1) electrostatic precipitators (ESPs); (2) dry scrubbers
combined with fabric filters (DSFF)
Stack temperature: (1) 470° kelvin with ESPs; (2) 430° kelvin with DSFFs
Building configuration: Height: 35 m; width: 76 m; length: 42 m
6.5 DERIVATION OF MASS EMISSION FACTORS FOR THE HYPOTHETICAL MASS BURN,
HEAT RECOVERY MSW INCINERATOR
6.5.1. Congener-Specific Stack Gas Emission Factors for Dioxin-like Compounds
Since 1980, the U.S. EPA has been actively stack testing representative MSW
incinerators to measure the magnitude of air emissions of PCDDs and PCDFs. In 1987,
the EPA gave advanced notice of proposed rulemaking under the Clean Air Act to establish
6-12 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
numeric emission limits of organic pollutants, including PCDDs and PCDFs from new and
exiting MSW incinerators (Cleverly,1989). In support of these regulatory activities, the
U.S. EPA initiated combustion research to determine the magnitude of PCDD and PCDFs
and other organic emissions from all types of incinerator systems operational in the U.S.
and Canada, e.g., mass burn, RDF, and modular. This research was a joint effort with
Environment Canada, and included measurements of pollution control technology to
determine effectiveness in reducing and controlling PCDD and PCDF emissions to the
atmosphere. At the same time regulatory agencies in Europe, Japan, and Scandinavia
were undertaking similar activities.
Regardless of the type of MSW incinerator, all MSW incineration systems emit
PCDDs and PCDFs. Every known toxic congener of PCDDs and PCDFs, e.g. having
chlorine atoms in the lateral 2,3,7,8 positions on the molecule, have been measured in
stack gas emissions to MSW incinerators. The emission of PCDDs and PCDFs is variable
between technologies, and whether constructed and operational before or after 1987
(U.S.EPA,1987b;1989). This is because the testing of incinerators has yielded information
on design characteristics, techniques of incinerator operation and performance, and air
pollution control technology that has had the overall benefit of reducing emissions of
organic pollutants (OTS,1989). Review of existing emissions data shows that the
magnitude in stack emissions of PCDDs and PCDFs between newer and older MSW
incinerators, and between facilities equipped with electrostatic precipitators (ESPs) or dry
scrubbers in combination with fabric filters (DSFF), indicates up to a 1000-fold difference
in the range of emissions of PCDDs/PCDFs (U.S.EPA,1987b; Cleverly, 1991; Siebert,1991;
OTS,1989).
Statistical analysis of the variability in emissions (using Spearman rank-order
techniques) indicated that, as a general trend, emissions of PCDDs and PCDFs appear to
be inversely related to furnace temperature, and that low combustion efficiency may
promote post-furnace formation (U.S.EPA,1 986). However further analysis of the fly ash
emissions of PCDDs and PCDFs are not related to furnace conditions alone, and most likely
involve a number of complex variables related to function of the incineration process and
design, including post-combustion formation outside the firebox region (see section 6.3:
The Mechanism of Formation). The compounds found in MSW are complex and variable,
and thus mapping any mechanism for formation of specific toxic organic compounds is an
6-13 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
improbable task and is largely theoretical. It is difficult to predict how constituents present
in the fuel will break apart, reform new compounds, dissociate into free radicals, or
recombine under the influence of oxygen, turbulent mixing and temperature.
Emission factors for specific congeners of dioxin-like compounds are key
parameters in estimating the environmental fate and transport and ultimate pathways of
human exposure to the pollutants. In the past the EPA converted the concentration of
mixtures of PCDD and PCDF congeners into an equivalent concentration of 2,3,7,8-TCDD
(Cleverly,!991; U.S.EPA,1989b; Mukerjee,1988) when deriving an emission factor.
Therefore when emissions were modeled using standard air dispersion models, the
emission to the atmosphere at the point of the stack to the MSW incinerator was the toxic
equivalent of 2,3,7,8-TCDD. The physical and chemical properties of 2,3,7,8-TCDD were
applied to model the environmental fate of the chemical mixture. In order to increase the
level of confidence with respect to predicting pathways to human exposure, the air
dispersion modeling of emissions from the hypothetical incinerator is for the specific toxic
congeners, and the chemical and physical properties of the individual toxic congeners are
employed to estimate the magnitude and pathways of human exposure. Only at the
interface to human exposure, e.g., ingestion, inhalation, dermal adsorption, etc., are the
individual congeners recombined and converted into the toxic equivalence of 2,3,7,8-
TCDD to be factored into quantitative risk assessments.
Emission factors for the analysis of emissions were estimated from test reports of
technologies that are most similar in design and operation as the hypothetical MSW
incinerator. The inclusion of multiple test reports from multiple facilities increases the
confidence that the predicted emissions of the pollutants from the hypothetical facility
under study will be in good agreement and fairly representative of emissions from actual
facilities of this size. The variability in emissions of PCDDs and PCDFs across MSW
technologies and by age of facility indicates that it is important to establish a selection
criteria to increase the likelihood that the emission factors are fairly typical. The criteria
used here was to use congener-specific emissions data from mass burn heat recovery
MSW incinerators that were appropriately designed to today's standards, and that apply
the principles of good combustion practice (Brunner, 1984).
Simplified statistical techniques were used to derive a representative emission
factor for each dioxin-like congener. After selecting test data of emissions from similar
6-14 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
technologies, it is necessary to place all emission terms into common units of
measurement. English units were converted into metric, and the density of the mass of
pollutant per unit volume of combustion gas emitted from the stack were corrected to the
same measurement of percent carbon dioxide within the combustion gas (12% C02)-
These adjustments were necessary to combine data from a variety of tests of different
mass burn incinerators. The next step involved converting the mass emission rate of the
specific dioxin-like congener in units of nanograms per normal cubic meter of combustion
gas corrected to 12% carbon dioxide into an equivalent emission rate in units of grams per
metric ton of refuse (g/T) that is incinerated. This was done for each incinerator by the
formula:
I = D V« (6-2)
T CR
where:
D = measured mass of dioxin congener per volume of combustion gas, e.g.,
grams/dry standard cubic meter
Vcg = volume of combustion gas/unit of time measured as a release from the
stack, e.g., dry standard cubic meters per minute.
CR = MSW charging rate (metric tons) to the incinerator per unit of time
equivalent to time in Vcg, e.g., metric ton per minute.
Tables 6-4 and 6-5 are the computed mass emission factors (grams/metric ton) for
each dioxin-like congener of PCDD and PCDF, respectively, of the mass burn heat recovery
MSW incinerators selected to be most similar to the hypothetical facility in this analysis.
On the bottom of each table is the computed arithmetic average emission factor derived
from inclusion of emissions from ten mass burn facilities. These congener-specific
emission factors are used to represent emissions from the hypothetical MSW incinerator,
and will be used to derive emissions for air dispersion and deposition modeling of emission
impacts.
The estimation of the emission factors of dioxin-like polychlorinated biphenyl (PCB)
congeners is made difficult by lack of emissions data from tested MSW incinerators. PCBs
6-15 7/31/92
-------
Table 6-4. Estimation of PCDD congener-specific emission factors (grams/metric ton of MSW burned) for
representative mass burn heat recovery incinerators
en
2,3,7,8-
Facility TCDD Other TCDD
1 2.10E-06 3.16E-05
2 7.85E-06 7.80E-05
3 9.40E-06 1.67E-04
4 3.97E-07 5.94E-06
5 1.94E-06 2.68E-06
6 1.56E-05 1.10E-03
7 1.97E-06 4.72E-05
8 2.78E-07 1.38E-05
9 2.91E-06 2.69E-05
10 8.15E-07 1.69E-05
Average 4.33E-06 1.49E-04
Facility
1. Chicago NW (U.S. EPA, 1983)
2. N. Andover, MA (U.S. EPA, 1988a)
3. Saugus, MA (Knisley, 1986c)
4. Tulsa, OK (Seelinger, 1986b)
5. Marion Co., ORa (U.S. EPA, 1988c)
6. Umea, Sweden (Markland, 1985)
7. Zurich, Switzerland (Siebert, 1991)
8. Alexandria, VA (Siebert, 1991)
9. Westchester, NY (Siebert, 1991)
10. Pinellas Co., FL (Entropy, 1987)
1,2,3,7,8-
CDD
NA
6.38E-06
1.88E-05
2.40E-06
2.27E-07
2.81E-04
NA
6.63E-07
7.65E-06
1.66E-06
3.98E-05
Other PCDD
NA
1 .44E-04
1.77E-04
3.54E-05
2.34E-06
2.75E-03
1 .24E-04
1 .08E-05
2.19E-05
2.30E-05
3.65E-04
1,2,3,4,7,8-
HxCDD
NA
7.80E-06
1.05E-05
1 .90E-06
2.89E-07
4.99E-05
NA
5.14E-07
1.57E-06
2.52E-06
9.38E-06
1,2,3,6,7,8-
HxCDD
NA
1 .65E-05
1.77E-05
4.68E-06
1 .47E-07
1.15E-04
NA
1.18E-06
NA
4.76E-06
2.29E-05
aMarion County emissions are controlled by dry scrubbers combined with fabric filters. Because the hypothetical facility is
equipped with an ESP as baseline conditions, the uncontrolled emissions were used in the derivation of emission factors.
NA = Congener not analyzed.
(continued on the following page)
-------
Table 6-4. (continued)
1,2,3,7,8,9-
Facility HxCDD
1 NA
2 8.84E-06
3 NA
4 NA
5 5.68E-08
6 4.06E-05
7 NA
8 1.71E-06
9 NA
10 6.65E-06
Average 1.16E-05
Facility
1. Chicago NW (U.S. EPA, 1983)
2. N. Andover, MA (U.S. EPA, 1988a)
3. Saugus, MA (Knisley, 1986c)
4. Tulsa, OK (Seelinger, 1986b)
5. Marion Co., ORa (U.S. EPA, 1988c)
6. Umea, Sweden (Markland, 1985)
7. Zurich, Switzerland (Siebert, 1991)
8. Alexandria, VA (Siebert, 1991)
9. Westchester, NY (Siebert, 1991)
10. Pinellas Co., FL (Entropy, 1987)
Other HxCDD
8.24E-05
1 .99E-04
1.63E-04
6.45E-05
4.28E-05
7.92E-04
2.45E-04
1.18E-05
3.63E-05
3.96E-05
1 .68E-04
1,2,3,4,6,7,8-
HpCDD
NA
NA
NA
2.78E-05
1 .80E-05
NA
NA
7.06E-06
NA
3.15E-05
2.11E-05
Other HpCDD
3.83E-05
1 .88E-04
1 .66E-04
5.60E-05
1.92E-05
5.62E-04
2.96E-04
6.61 E-06
5.53E-05
3.34E-05
1 .42E-04
OctaCDD
1 .28E-05
1 .54E-04
2.06E-04
6.08E-05
3.76E-05
3.74E-04
6.04E-04
1 .43E-05
9.23E-05
8.30E-05
1 .64E-04
aMarion County emissions are controlled by dry scrubbers combined with fabric filters. Because the hypothetical facility is
equipped with an ESP as baseline conditions, the uncontrolled emissions were used in the derivation of emission factors.
NA = Congener not analyzed.
-------
Table 6-5. Estimation of PCDF congener-specific emission factors (grams/metric ton of MSW burned) for
representative mass burn heat recovery incinerators
2,3,7,8-
Facility TCDF
1 NA
2 1.05E-04
3 1.29E-04
4 1.14E-05
5 2.44E-05
6 7.80E-05
7 NA
8 1.80E-05
9 2.30E-05
10 4.05E-06
en Average 4.91E-05
m
Facility
1. Chicago NW (U.S. EPA, 1983)
2. N. Andover, MA (U.S. EPA, 1988a)
3. Saugus, MA (Knisley, 1986c)
4. Tulsa, OK (Seelinger, 1986b)
5. Marion Co., ORa (U.S. EPA, 1988c)
6. Umea, Sweden (Markland, 1985)
7. Zurich, Switzerland (Siebert, 1991)
8. Alexandria, VA (Siebert, 1991)
9. Westchester, NY (Siebert, 1991)
Other TCDF
4.53E-04
3.10E-04
8.76E-04
2.86E-05
2.20E-05
2.61E-03
2.51E-04
1.15E-04
2.91E-04
8.74E-05
5.04E-04
1,2,3,7,8-
PCDF
NA
2.66E-05
3.26E-05
1.77E-06
9.47E-08
2.81E-04
NA
3.39E-06
6.58E-05
2.60E-06
5.17E-05
2,3,4,7,8-
PCDF
NA
5.39E-05
5.64E-05
3.63E-06
1.19E-05
1 .90E-04
NA
5.38E-06
NA
6.05E-06
4.68E-05
Other PCDF
NA
1.74E-04
4.96E-04
1.31E-05
7.62E-06
2.56E-03
2.60E-04
5.20E-05
1.17E-04
4.18E-05
4.13E-04
10. Pinellas Co., FL (Entropy, 1987)
"Marion County emissions are controlled by dry scrubbers combined with fabric filters. Because the hypothetical facility is
equipped with an ESP as baseline conditions, the uncontrolled emissions were used in the derivation of emission factors.
NA = Congener not analyzed.
(continued on the following page)
-------
Table 6-5. (continued)
1 23478-
' / *- 1 *^ / ^/ *!***
Facility HxCDF
1 NA
2 3.56E-05
3 7.19E-05
4 2.14E-06
5 1 .47E-07
6 1.12E-04
7 NA
8 5.25E-06
9 2.09E-05
1 0 1 .OOE-05
Average 3.23E-05
Facility
1. Chicago NW (U.S. EPA, 1983)
2. N. Andover, MA (U.S. EPA, 1988a)
3. Saugus, MA (Knisley, 1986c)
4. Tulsa, OK (Seelinger, 1986b)
5. Marion Co., ORa (U.S. EPA, 1988c)
6. Umea, Sweden (Markland, 1985)
7. Zurich, Switzerland (Siebert, 1991)
8. Alexandria, VA (Siebert, 1991)
9. Westchester, NY (Siebert, 1991)
10. Pinellas Co., FL (Entropy, 1987)
1,2,3,6,7,8-
HxCDF
NA
1.09E-05
4.31E-05
8.52E-07
1.56E-07
1.15E-04
NA
2.42E-06
NA
5.05E-06
2.54E-05
1,2,3,7,8,9-
HxCDF
NA
1 .03E-04
NA
3.53E-07
1 .89E-07
2.50E-05
NA
1 .03E-07
NA
1 .50E-07
2.15E-05
2,3,4,6,7,8-
HxCDF
NA
NA
NA
2.29E-06
8.33E-06
8.11E-05
NA
2.88E-06
NA
8.05E-06
2.05E-05
Other HxCDF
1.31E-04
3.13E-05
2.69E-04
7.10E-06
NA
6.65E-04
1 .22E-04
3.75E-06
1 .64E-04
2.72E-05
1 .58E-04
aMarion County emissions are controlled by dry scrubbers combined with fabric filters. Because the hypothetical facility is
equipped with an ESP as baseline conditions, the uncontrolled emissions were used in the derivation of emission factors.
NA = Congener not analyzed.
(continued on the following page)
-------
Table 6-5. (continued)
o-i
I
2,3,4,6,7,8-
Facility HpCDF
1 NA
2 NA
3 NA
4 5.72E-06
5 2.84E-07
6 NA
7 NA
8 6.38E-06
9 9.08E-05
10 2.65E-05
Average 2.59E-05
Facility
1. Chicago NW (U.S. EPA, 1983)
2. N. Andover, MA (U.S. EPA, 1988a)
3. Saugus, MA (Knisley, 1986c)
4. Tulsa, OK (Seelinger, 1986b)
5. Marion Co., ORa (U.S. EPA, 1988c)
6. Umea, Sweden (Markland, 1985)
7. Zurich, Switzerland (Siebert, 1991)
8. Alexandria, VA (Siebert, 1991)
9. Westchester, NY (Siebert, 1991)
10. Pinellas Co., FL (Entropy, 1987)
2,3,4,7,8,9-
HpCDF
NA
NA
NA
6.59E-07
3.79E-07
NA
NA
7.20E-07
NA
1 .50E-06
8.14E-07
Other HpCDF
3.76E-05
2.57E-04
2.00E-04
9.21E-06
3.27E-07
1 .06E-03
2.76E-05
3.19E-06
2.09E-05
7.60E-06
1.62E-04
OctaCDF
3.03E-06
2.18E-04
9.79E-05
4.16E-07
2.98E-07
3.12E-04
1.01E-04
2.93E-06
4.18E-06
6.30E-06
7.45E-05
aMarion County emissions are controlled by dry scrubbers combined with fabric filters. Because the hypothetical facility is
equipped with an ESP as baseline conditions, the uncontrolled emissions were used in the derivation of emission factors.
NA = Congener not analyzed.
-------
DRAFT-DO NOT QUOTE OR CITE
are a class of toxic air pollutants that have not been routinely monitored in stack emissions
to MSW incinerators. The data that are available usually have reported PCBs as the sum
total of all PCBs present in the sample without further speciation of toxic congeners or
congener groups (U.S. EPA,1987c). The greatest level of detail in any test report is a
further breakdown of total PCBs into homologue groups, e.g., mono - deca-chlorobiphenyl.
For purposes of undertaking an air dispersion modeling analysis of the emission of
specific dioxin-like congeners from the stack of the hypothetical MSW incinerator, it is
necessary to derive a presumption of the distribution of specific congeners using the best
available emission data, e.g., homologue groups of PCBs in the emission. Lacking any
applicable law of thermodynamics that would permit an estimation of congener-specific
emissions based on the characteristics of the MSW fed into the system, and the
characteristics of time, temperature and turbulence in the incineration process itself, one
can estimate a distribution based on simple assumptions of the probability a specific
congener exists within the appropriate homologue group. Until EPA has congener-specific
emissions data from testing operational MSW incinerators, the simplifying assumption is
that all congeners within a homologue group shall have an equal probability of occurrence
within the emission. Therefore the specific toxic coplanar PCB congener listed in Table
6-6 shall have an equal probability of occurrence in the respective homologue group. If
equal probability of occurrence is assumed, then the estimation of the emission factor of
the specific coplanar PCB congener can be made by establishing a ratio of the congener of
concern to the total possible isomers in the chlorinated homologue group. For example,
there are 42 possible isomers in the tetrachloro homologue group. Therefore if equal
probability of occurrence of each isomer in the homologue group is assumed, then each
tetrachlorinated isomer in Table 6-6 has a chance occurrence in the stack gas emissions of
1 in 42, or a ratio of 0.024. These ratios can be multiplied by the homologue-specific
emission factor derived from incinerator tests to estimate the emission factor for the
specific congener under evaluation. Table 6-7 displays the computation of emission
factors by homologue group of PCBs, and Table 6-8 is the estimated emission factor for
specific dioxin-like congeners of coplanar PCBs.
It must be noted that the estimated emission factors for specific coplanar
congeners of PCBs listed in Table 6-8 are highly uncertain. The following contribute to
this uncertainty:
6-21 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-6. Coplanar polychlorinated biphenyl congeners having dioxin-like biological
activity; number of possible isomers; ratio of toxic isomer to total isomers in homologue
group
Congener
3,3',4,4'-TeCB
3,4,4', 5-TeCB
3,3',4,4',5-PeCB
2,3,3',4,4'-PeCB
2,3,4,4', 5-PeCB
2,3',4,4',5-PeCB
3,3',4,4',5,5'-HxCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
2,3,3',4,4',5,5'-HCB
Homologue
Tetra-chloro
Tetra-chloro
Penta-chloro
Penta-chloro
Penta-chloro
Penta-chloro
Hexa-chloro
Hexa-chloro
Hexa-chloro
Hexa-chloro
Hepta-chloro
Number of
isomers
42
42
46
46
46
46
42
42
42
42
24
Ratio
0.024
0.024
0.022
0.022
0.022
0.022
0.024
0.024
0.024
0.024
0.042
6-22
7/31/92
-------
Table 6-7. Emission factors (grams/metric ton) of homologue groups of polychlorinated biphenyls from emission tests
of MSW incinerators
Facility DiCB TriCB TetraCB PentaCB HexaCB HeptaCB OctaCB
1. Padua, Italy8 2.00E-03 1.41E-02 1.37E-02 2.46E-05 4.35E-03 3.09E-03 3.08E-05
2. Chicago, NWb 2.76E-04 2.55E-04 9.97E-05 4.16E-05 NA NA NA
3. Hartford, Conn0 7.08E-05 1.01E-04 6.47E-05 NA NA 2.95E-05 NA
Average 7.82E-04 4.81E-03 4.63E-03 3.31 E-05 4.35E-03 1.56E-03 3.08E-05
"Mass burn, heat recovery unit rated at 109 MT/d (Magagni, 1990).
bMass burn, heat recovery unit rated at 363 MT/d (U.S. EPA, 1 983).
CRDF, heat recovery unit rated at 662 MT/d (U.S. EPA, 1991).
I
to
U)
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-8. Estimation of congener-specific emission factors of coplanar PCBs
PCB congener
3,3',4,4'-TeCB
3,4,4',5-TeCB
3,3',4,4',5-PeCB
2,3,3',4,4'-PeCB
2,3,4,4', 5-PeCB
2,3',4,4',5-PeCB
3,3',4,4',5,5'-HxCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
2,3,3',4,4',5,5'-HCB
Homologue
emission
factor8
(grams/MT)
4.63 x 10'3
4.63 x 1CT3
3.31 x 10'5
3.31 x 1CT5
3.31 x 1Q-5
3.31 x 10'5
4.35 x 10'3
4.35 x 1Q-3
4.35 x 10'3
4.35 x 1CT3
1.56 x 10'3
Ratio
isomer/
homologueb
0.024
0.024
0.022
0.022
0.022
0.022
0.024
0.024
0.024
0.024
0.042
Congener
emission
factor0
(grams/MT)
1.1 x 10-4
1 . 1 x 1 0'4
7.3x 10'7
7.3x TO'7
7.3x 1Q-7
7.3 x 10'7
1.0x1 0'4
1.0x1 Q-4
1.0x TO'4
1.0x 10'4
6.6x 10'5
aHomologue PCB emission factor (grams/metric ton of MSW burned) taken from
Table 6-7.
bRatio of dioxin-like coplanar PCB isomer to number of isomers in homologue group
taken from Table 6-6.
cThe homologue group emission factor times the ratio of specific isomer within the
homologue group equals the estimated congener-specific coplanar PCB emission factor
(grams/metric ton of MSW burned).
A general lack of emission test reports characterizing the stack gas emission
of coplanar PCBs from representative MSW incineration technologies.
Because of a general lack of emission characterization studies, it was
necessary to combine emissions data of unrelated MSW incinerator
technologies, e.g., mass burn and RDF facilities, in order to derive a PCB
homologue-specific emission factor. It is not known if the magnitude of PCB
emissions varies greatly from one technology to another.
The emission of PCBs seems to be related to the incomplete combustion of
PCBs present as contaminants in the raw MSW, and not a post-combustion
formation phenomenon. PCBs may or may not be present as contaminants
in the MSW fed into the incinerator, and therefore, may or may not be
6-24 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
present in the stack gas emissions.
• Congener-specific emission factors of coplanar PCBs were estimated from
homologue emission data using simplifying assumptions. No congener-
specific emissions from MSW incinerators could be found in the technical
literature.
6.5.2. Estimation of Emission of Dioxin-Like Compounds in MSW Incineration Ash
The previous section estimated the mass rate of emission of specific congeners of
PCDDs, PCDFs and PCBs from the stack of the hypothetical MSW incinerator. This
section is an estimation of the mass of ash that is produced as a residue from the
combustion of MSW on the grate within the incinerator (bottom ash), and the ash that is
collected by the paniculate matter control device preceding the stack (fly ash). This
estimation is appropriate for the analysis of human exposures to dioxin-like compounds
after storing, transporting, and disposing of the fly ash residues from the hypothetical
mass burn incinerator into a sanitary landfill or ash monofill.
In general there are many factors that may influence the formation of paniculate
matter known as bottom ash and fly ash from the incineration of MSW. Among these
factors are: the heating value of the MSW (Btu/kg), the percent moisture in the refuse, the
furnace temperature and combustion efficiency, and the efficiency of paniculate matter
capture by the pollution control device (Brunner,1984; OTS,1989).
For purposes of exposure analyses of the hypothetical MSW incinerator, the
assumptions of Cook (1991) are used. Cook (1991) extensively evaluated ash generated
from modern mass burn heat recovery incinerators and created the following generalized
mass balance from the data:
• Bottom ash from the combustion grate is produced at a rate of 10 percent of
the mass of MSW incinerated per day.
• Fly ash is generated from the paniculate matter pollution control device (ESP
or Fabric Filter) at a rate of 1 percent of the mass of MSW incinerated per
day.
• Fly ash from the addition of a semi-dry acid gas control device is produced
at a rate of 0.1 percent of the mass of MSW incinerated per day. Applying
these generalized assumptions, the daily mass of ash generated at the
6-25 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
hypothetical MSW incinerator is as follows:
With Electrostatic Precipitators (ESPs) for paniculate matter control at
99% control efficiency:
1. Bottom ash: 2727 metric tons/day incinerated X 10 %.
Bottom ash: 272.7 metric tons/day.
2. Fly ash: 2727 metric tons/day incinerated X 1 %.
Fly ash: 27.3 metric tons/day.
With Fabric Filters (99% control efficiency) combined with semi-dry
alkaline scrubbers:
1. Bottom ash: 272.7 metric tons/day.
2. Fly ash: a. 27.3 metric tons/day from fabric filters.
b. 2.7 metric tons/day from dry scrubber.
c. Sum (a) + (b) = 30 metric tons/day.
The PCDDs, PCDFs, and coplanar PCBs are primarily associated with the fly ash
that is prevented from escaping into the atmosphere by the particulate matter control
device (Chrostowski, 1991; Denison, 1991; OTS,1989). Negligible amounts of
PCDDs/PCDFs have been detected on bottom ash from the combustion grate because
these chemicals are synthesized outside of the furnace region (OTS, 1989). No data could
be found on PCB contamination of MSW bottom ash. It is customary for MSW
incineration facilities to combine fly ash with the bottom ash prior to disposal in a sanitary
landfill (Chrostowski, 1991). This will have the overall effect of diluting the initial
contaminant concentration, because the mass of bottom ash exceeds fly ash by a ratio of
10/1. For purposes of the exposure analysis of ash disposed in landfills, this Chapter will
focus only on the fly ash generated at the hypothetical MSW incinerator, and will
characterize the distribution of concentration of dioxin-like congeners without dilution
through mixing with bottom ash.
The descriptive air dispersion modeling parameters for the hypothetical mass burn
MSW incinerator are given in Table 6-3. There are basically two pollution control
scenarios to the hypothetical incinerator which will affect the estimated generation of fly
ash. The first scenario uses only electrostatic precipitators for particulate matter control,
and the second scenario uses dry-alkaline scrubbers combined with fabric filters. Both the
electrostatic precipitators and the fabric filters are assumed to have the same level of
6-26 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
efficiency for paniculate matter control, e.g. 99%. However it is assumed that ESPs will
reduce emissions of PCDDs, PCDFs, and coplanar PCBs by only 20%, which is consistent
with reports of testing ESP units (Cleverly, 1 991; U.S. EPA, 1987b). Dry alkaline
scrubbers combined with fabric filters are assumed to reduce and control these pollutants
by 95%, which is consistent with EPA tests of the effectiveness of such control devices
(U.S EPA, 1987b). The mechanism involved is the dramatic temperature reduction
achieved with the dry alkaline scrubbers (e.g., to less than 200° C.), which promotes
condensation of vapor phase pollutants onto fly ash surfaces before passing through the
fabric filter (U.S. EPA, 1987b). From this it can be seen that the mass emission rate of
dioxin-like compounds from the stack to the atmosphere will be lower with the application
of dry-alkaline scrubbers combined with fabric filters than with ESPs alone, but the fly ash
generation rate and the concentration of dioxin-like chemicals in the fly ash will be greater.
The concentration of dioxin-like compounds in the fly ash can be estimated by the
difference in capture and control efficiency between the two pollution control scenarios.
What is not emitted from the stack to the atmosphere will be contained in the captured fly
ash to the particulate matter control device. Tables 6-4, 6-5, and 6-8 are the estimated
congener-specific stack air emission factors for dioxin, dibenzofurans, and coplanar PCBs,
respectively. These emission factors were derived from mass burn heat recovery MSW
incinerators equipped with ESPs for particulate matter control. To estimate stack emission
factors resulting from the application of dry alkaline scrubbers in combination with fabric
filters, the emission factors from ESP facilities can be adjusted as follows:
= EFesp 1^ (6-3)
where:
EFC = emission factor for the additional control with dry alkaline scrubbers
combined with fabric filters
EFesP = emission factor for the ESP control scenario from Tables 6-4, 6-5, and
6-8
CEC = control efficiency (95%) assumed for dry-alkaline scrubbers combined
with fabric filters
CEesp = assumed control efficiency (20%) for ESPs.
6-27 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
From this relationship, it can be seen that the stack gas emission factor for dioxin-
like congeners is 16 times greater for ESP control scenario than for dry-scrubber coupled
with fabric filter (DSFF) control scenario. However, the concentration of dioxin-like
congeners in the collected fly ash is 4.75 times greater in the dry-scrubber, fabric filter
scenario than in the ESP control scenario. The concentration in the collected fly ash is
about 4 times greater in the collected fly ash than what is emitted from the stack under
the ESP control scenario, and about 19 times greater in the collected fly ash than in stack
emissions under the DSFF scenario. Tables 6-9 and 6-10 are the estimated concentration
of chlorinated dioxin, dibenzofuran, and coplanar PCB congeners in the fly ash in the ESP
control scenario, and Tables 6-11 and 6-12 are the fly ash concentrations of specific
congeners in the DSFF control scenario to the hypothetical MSW incinerator facility.
6.6. AIR DISPERSION MODELING OF THE STACK GAS EMISSIONS OF DIOXIN-LIKE
COMPOUNDS FROM THE HYPOTHETICAL MSW INCINERATOR
It has been customary for EPA to use air dispersion models to estimate ambient air
concentrations of specific pollutants attributable to smokestack emissions from an
industrial source. In the analysis of potential ground-level exposures to dioxin-like
compounds in the vicinity of the hypothetical MSW incinerator, the Industrial Source
Complex (ISC) air dispersion model was used.
Air dispersion models are mathematical constructs that approximate the physical
and chemical processes occurring in the atmosphere that directly influence the dispersion
of gaseous and particulate emissions from smokestacks of stationary combustion sources.
These models are computer programs encompassing a series of partial differential and
algebraic equations to calculate and approximate the dispersion and surface deposition of
the emissions to estimate the spacial and temporal ground-level concentrations of
pollutants. Concentration isopleths of the pollutants discharged from the stack are
computed along the polar azimuth at specified distances downwind of the smokestack.
The estimated ground-level concentrations are used to estimate the magnitude of potential
exposures to the human receptor.
Kapahi (1991) has reviewed the physical principles involved in most air dispersion
models used by EPA. Emphasis is placed on the fact that these models are abstractions of
reality that permit the prediction of hypothetical future states of a system that may not
6-28 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-9. Emission factors (grams/metric ton) of PCDDs/PCDFs for collected fly ash in
the ESP control scenario
Congener
Emission factor
Congener
Emission factor
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OctaCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
1.7 x 1CT5
1.6x 10'4
3.8 x 10'5
9.2x 10'5
4.6 x 10'5
8.4 x 1CT5
6.6x 10'4
2.0 x 10'4
2.0 x 10'4
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OctaCDF
2.0 x 10'4
1.3x 10'4
1.0x10'4
8.6 x ID'5
8.2x 10'5
1.0x10-4
3.3 x 10'6
3.0 x 10"4
Table 6-10. Emission factors (grams/metric ton of MSW incinerated) of coplanar PCBs for
collected fly ash in the ESP control scenario
Congener
Emission factor
3,3',4,4'-TeCB
3,4,4', 5-TeCB
3,3',4,4',5-PeCB
2,3,3',4,4'-PeCB
2,3,4,4', 5-PeCB
2,3',4,4',5-PeCB
3,3',4,4',5,5'-HxCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5-HxCB
2,3,3',4,4',5,5'-HCB
4.4 x TO'4
4.4 x 10'4
3.0 x 10'6
3.0 x 10'6
3.0 x 10'6
3.0 x 10'6
4.0 x 10'4
4.0 x 10'4
4.0 x 10'4
4.0 x 1C'4
2.6 x 10'4
6-29
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-11. Emission factors (grams/metric ton of MSW incinerated) of PCDDs/PCDFs for
collected fly ash in the dry-scrubber, fabric filter control scenario
Congener
Emission factor
Congener
Emission factor
2,3,7,8-TCDD 8.0 x 1CT5
1,2,3,7,8-PeCDD 7.6 x10'4
1,2,3,4,7,8-HxCDD 1.8 x10'4
1,2,3,6,7,8-HxCDD 4.4 x 1CT4
1,2,3,7,8,9-HxCDD 2.2 x 10'4
1,2,3,4,6,7,8-HpCDD 4.0 x 10'4
1,2,3,4,6,7,8,9-OctaCDD 3.1 x 10'3
2,3,7,8-TCDF 9.5 x 10'4
1,2,3,7,8-PeCDF 9.5 x 10'4
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
1,2,3,4,6,7,8,9-OctaCDF
9.5 x 1CT4
6.2 x 10'4
4.8 x 1CT4
4.1 x 10'4
4.0 x 10'4
4.8x TO'4
1.6x 10'5
1.4x10'4
Table 6-12. Emission factors (grams/metric ton of MSW incinerated) of coplanar PCBs for
collected fly ash in the dry-scrubber, fabric filter control scenario
Congener
Emission factor
3,3',4,4'-TeCB
3,4,4', 5-TeCB
3,3',4,4',5-PeCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
3,3',4,4',5,5'-HxCB
2,3,3',4,4',5'-HxCB
2,3,3',4,4',5-HxCB
2,3',4,4',5,5-HxCB
2,3,3',4,4',5,5'-HCB
2.1 x 10'3
2.1 x 1(T3
1.4x 1Q-5
1.4x1 CT5
1.4x 10'5
1 .4 x 1 0"5
2.0 x 1CT3
2.0 x 10'3
2.0 x 10'3
2.0 x 10'3
2.6x 10'4
6-30
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
exist. The primary utility is to avoid unwanted adverse occurrences through a thorough
simulation and analysis of possible impacts over a specified time interval.
In simple terms the ISC air dispersion model uses the basic physical processes of
advection, turbulent diffusion, and removal to estimate the atmospheric transport and
settling of particles comprising the gaseous and particulate emissions from the stack.
These physical principles are illustrated in Figure 6-1. Advection describes the physical
movement of the air pollutants by the horizontal movement of wind. The horizontal wind
direction and speed are key to estimating the geographical direction and distance the
emission plume will disperse, and to approximate the rate of dilution of pollutant
concentrations with downwind distance from the emission source. Wind speed and
direction can vary seasonally and diurnally, or as a function of passing frontal systems
(Kapahi,1991). Wind roses are developed through systematic measurement of hourly wind
speed and direction at 10 meters above the surface of the earth that reflect monthly wind
direction and intensity along the polar azimuth. These records are crucial to air dispersion
analysis.
TURBULENT
DIFFUSION
WIND SPEED + DIRECTION
Figure 6-1. Principal physical processes used by the ISC dispersion model to estimate the
atmospheric transport and deposition of pollutants discharged from the smokestack.
6-31
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Turbulent diffusion is simulated by the model to approximate the "spreading" of the
emissions plume with distance from the stack (Kapahi, 1991). Turbulence is caused by
the randomized variation of wind speed and direction over time, and the stratified
temperatures at various altitudes in the lower 500 meters in the atmosphere. Random
fluctuations in wind velocities result in rates of momentum, heat, and mass transfer in
turbulence that are greater than that of simple molecular transport (Seinfeld, 1986).
In the evaluation of the emissions of the hypothetical incinerator, the Industrial
Source Complex (ISC) model was used to characterize plume dispersion (U.S. EPA, 1986).
The ISC uses the generalized Briggs (1975,1977) equation to estimate plume-rise and
downwind dispersion as a function of turbulent diffusion. A wind-profile exponent law is
used to adjust the observed mean wind speed from the measurement height to the
emission height for the plume rise and pollutant concentration calculations. The Pasquill-
Gifford curves (Turner, 1970) are used to calculate lateral and vertical plume spread. The
ISC contains rural and urban options to account for the effects of terrain, and the
aerodynamic wakes and eddies of building and tall structures on turbulent diffusion.
Removal processes that are simulated by the ISC model are dry and wet deposition
of particulate matter to the surface of the earth. Dry deposition technically refers to the
removal of particulate-bound contaminants by gravitational settling and turbulent diffusion.
The rate at which the atmospheric particulate is removed by the physical forces of gravity
and atmospheric turbulence is termed the deposition flux (Kapahi, 1991), and is
mathematically represented by Fd. The relationship between Fd, the concentration of the
chemical contaminant, C0, and the settling velocity of the contaminated particles can be
defined by:
I'd = vd C0 (6-4)
where Vd is the settling velocity of the particulates (Kapahi, 1991). Particles greater than
30 micrometers (pm) in diameter will be removed from the atmosphere primarily by the
force of gravity, whereas particles less than 30 /ym will be removed primarily by
atmospheric turbulence. The deposition flux for the smaller particles is influenced by many
factors, including: the distribution of particles by diameter and density; assumptions of
atmospheric turbulence; the friction of the terrain; the height of the stack release of
6-32 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
emissions; and the concentration of the pollutant by particle size. The ISC Short-Term
(ISCST) version was used to estimate deposition flux based on the model developed by
Dumbauld and coworkers (1976). This model assumes that a fraction of the paniculate
comes into contact with the ground surface by the combined processes of gravitational
settling and atmospheric turbulence. The ISCST is designed to calculate deposition flux
for time periods of from one to 24 hours. If used with a year of sequential hourly
meteorological data, ISCST can also calculate annual concentration or deposition values.
A more detailed description of the default settling velocities by particle size is discussed in
the user's guide (U.S. EPA, 1986).
Wet deposition occurs by precipitation (rain,hail,snow) physically washing out the
chemically contaminated paniculate from the atmosphere. This is termed "scavenging",
and the wet deposition flux relates fraction of the time precipitation occurs, the fraction of
material removed by precipitation per unit of time by particle size, the strength of the
source emission (in grams), the wind speed at the release point to the smokestack, and
turbulent diffusion of the emission plume from the source (U.S. EPA,1987b). Based on
these relationships, scavenging coefficients were developed for varying types and
intensities of precipitation relative to different particle diameters by incorporating the
observations of Radke et al. (1980) in a study of scavenging of aerosol particles by
precipitation.
The principal assumptions made in computing wet deposition flux are: (1) the
intensity of precipitation is constant over the entire path between the source and the
receptor; (2) the precipitation originates at a level above the top of the emission plume so
that the hydrometers pass vertically through the entire plume; (3) the time duration of the
precipitation over the entire path between the source and receptor point is such that
exactly f, a fraction between zero and one, of the hourly emission is subject to a constant
intensity for the entire travel time required to traverse the distance between the source
and receptor, and the remaining fraction (1-f) is subject only to deposition processes.
Thus, no dry deposition occurs during hours of steady precipitation, and dry deposition
occurs between the periods of precipitation.
The ISC model is commonly referred to as a Gaussian plume model. In the
estimation of turbulent diffusion, the spread of the emission plume is viewed as a random
process (Kapahi,1991), with key measurement parameters following an approximately
6-33 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
normal distribution according to the Central Limit Theorem of statistics. The Central Limit
Theorem states that if random samples of n observations are drawn from a population
with finite mean, /j, and a standard deviation, 3, then when n is large, the sample mean
will be approximately normally distributed with mean equal to // and standard deviation
d/^/n. The approximation will become more and more accurate as n becomes large.
6.6.1. Estimation of Distribution in the Stack Emissions to the Hypothetical
MSW Incinerator
Certain inferences must be made concerning the distribution of particulate
differentiated on the basis of particle diameter (//m) before the ISCST program can predict
deposition of the dioxin-like congeners. Unfortunately there exists very few studies of the
distribution of particulate matter entrained in the emissions from MSW mass burn
incinerators broken down and fractionated by particle diameter. The distribution of
particulate matter by particle diameter will differ from one incineration process to another,
and is greatly dependent on the percent efficiency of particulate captor by particle
diameter from various pollution control devices. Table 6-13, for example, compares the
particle diameter distribution between emissions controlled by and ESP verses emissions
controlled by a fabric filter.
In comparison between ESP and the fabric filter controlled incinerator, it is apparent
that the distribution by particle diameter differs greatly from one control device to another.
A major uncertainty exists with respect to particulate distributions resulting from
increasing the collection plates and electrostatic fields in advanced ESP units. These
systems are capable of controlling the mass of particulate emissions to the same level of
performance as fabric filters, e.g., approximately 9.0 milligrams per dry standard cubic
meter at 1 2 percent C02 (U.S. EPA,1987c). Although specific mass fractionalization
studies are lacking, this implies that modern multi-field ESPs can be designed to result in a
similar distribution of particulate as fabric filters in the stack emissions.
For purposes of deposition analysis of particulate contaminated with dioxin-like
chemicals that are emitted from the hypothetical MSW incinerator, three particle size
categories are generalized from the data: > 10 fjm, 2 to 10 /vm, and < 2//m. This
particle size distribution will be kept constant in the analysis of dioxin-like compound
emissions resulting from the application of the two distinct control scenarios: ESPs, and
6-34 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-13. Comparison of particle-size distribution in particulate emissions resulting from
ESP controls with particulate emissions from fabric filter controls
ESP Controlled MSW Incinerator8
Fabric Filter Controlled
MSW Incinerator13
Mean diameter (//m)c
> 15.0
12.5
8.1
5.5
3.6
2.0
1.1
0.7
< 0.7
% of
particulated
12.8
10.5
10.4
7.3
10.3
10.5
8.2
7.6
22.4
Mean diameter (jjm)
> 12
7.5 - 12.0
5.1 - 7.5
3.5 - 5.1
2.3- 3.5
1.1 - 2.3
2.3 -0.7
< 0.47
% of
particulate
37.8
0.0
1.0
1.5
3.6
2.0
1.0
52.6
aESP data from U.S. EPA, 1980.
bFabric filter data adapted from Hahn and Hagenmaier, 1986.
GMean particle diameter is geometric mean.
dPercent of particulate is percent of total particulate mass.
dry-alkaline scrubbers combined with fabric filters. Not enough particle mass
fractionalization data exists to generalize a distribution pattern for each control device
based on the available limited measurements. Therefore the distribution from ESPs in
Table 6-13 will be applied to the calculation of mass emission rate of particulate-bound
congeners by particle size category.
For purposes of estimating deposition of the dioxin-like compounds emitted from
the stack of the hypothetical MSW incinerator, the relationship between the pollutant
concentration and particle diameter were assumed. The mass emission rate of the dioxin-
like congeners is distributed to the particulate matter distribution by calculating proportion
and available surface area for a given particle size (assuming that the particles are perfect
spheres, and particle density remains constant)(Hart,1985). The particle weight is
proportional to volume if density is held constant. Therefore, the ratio of the surface area
to volume is proportional to the ratio of the surface area to weight for a particle with a
6-35
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
given radius (Hart,1985). Multiplying this proportion times the weight fraction of particles
of a specific diameter (/ym) yields a number that approximates the amount of surface area
available for chemical adsorption. The logic is as follows:
(a) Assume aerodynamic spherical particles.
(b) Specific surface area of a spherical particle with radius,r:
S = 47r r2 (6-5)
(c) Volume of spherical particle with radius, r:
V = I .1 ) n r3 (6-6)
(d) The ratio of surface area to volume is:
S _ 4 n r2
V
If particle density is held constant, then it is assumed that particle weight is
proportional to particle volume. Hart (1986) further postulates that the ratio of surface
area to volume is proportional to the ratio of surface area to weight for a particle with a
given radius. Therefore, the ratio of surface area to volume represents the potential
relationship between the surface area and the weight of the particle.
When the calculations are summed for all particle size categories, total surface area
is assumed for total particle emissions. Dividing the surface area for each particle
category by the total available surface area for all particles gives an estimation of the
fraction of total area on any size particle. Multiplication of the emission rate of the dioxin-
like congener times the fraction of available surface area will estimate the emission rate of
the pollutant per particle size. The fraction of total surface area was computed for the
three particle size categories. The fraction of total surface areas for the ranges of particle
diameters are summed with each particle size category to represent a single fraction of
total surface area for the given particle size category, e.g., 0.03 for > 10 jjm; 0.095 for 2
6-36 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
to 10 /vm and 0.875 for < 2 //m. Thus 87.5 % of the emission rate of the dioxin-like
congener is calculated to be associated with particles less than 2 /vm in diameter.
In modeling the wet and dry deposition flux, and the ambient air concentrations of
the emitted dioxin-like congeners, it is necessary to assume the physical state of the
pollutants at the point of emission from the stack. This assumption is derived by the ratio
of vapor-phase to particle-bound emissions of these congeners. While the available
literature is weak in this area, inferences can be made from field validation studies of the
stack sampling procedure for dioxins (Method 23) using 13C12 labeled PCDD/PDCF
congeners as field spikes (MRI,1989). Sampling \n situ in the stack of the Chicago NW
mass burn heat recovery MSW incineratora while using a dynamic spiking system,
demonstrated that most of the isotope was recovered in the filter trap and front half of the
sampling train designed to capture paniculate, and a lower amount was recovered in the
XAD resin designed to capture vaporous compounds. In the particular tests in which the
overall percent recovery of the dynamic spike were found to be acceptable, the XAD resin
contained only about 10% of the isotope. The remainder, 90%, was associated with
paniculate. Other field validation studies involving recovery of labeled compounds while
sampling in the stack of an operating MSW incinerator have shown wide variation in the
distribution of spiked congeners (Hagenmaier,1989), depending on the configuration of the
sampling apparatus, and the temperature of the sampling probe. Hagenmaier's analysis of
a sampling train analogous to EPA Method 23 produced a distribution of dioxin congeners
of 15 % to 30 % in vapor form and 70 % to 85 % associated with paniculate matter
(Hagenmaier,1989). Thus for purposes of assuming vapor/particulate partitioning of the
dioxin-like compounds at the point of emission from the stack of the hypothetical MSW
incinerator, a vapor to particle ratio of 20 % vapor and 80 % particles is assumed. Given
the low vapor pressures of the dioxin-like chemicals, the analysis of condensate and XAD-
2 resin from EPA Method 23 could actually be aerosol particles and not true vapors.
Primary aerosols, defined as originating from the stack of a combustion source, usually are
characterized as mixtures of particles having diameters less than 1.0 jum (Seinfeld, 1984),
and within these dimensions such panicles may remain in suspension in the lower
atmosphere from a few days to several weeks. Wet deposition may be the principal
removal process of aerosol particles, however dry deposition will occur by increasing mass
through collisions and adhesion. The settling velocities of aerosol particles have been
6-37 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
estimated in the range of from 10~3 to 10"5 centimeter per second (Seingeld,1984). At
these extremely low settling velocities the aerosol particles will be transported by winds to
areas beyond microscale dispersion analysis of ISC. To simplify the analysis EPA has
assumed that aerosols of less than 1.0 //m will behave like gases, and are modeled
accordingly. Therefore aerosols will be equivalent to vapor for purposes of air dispersion
modeling, and will result in an ambient air concentration of the pollutants that will be
available for direct inhalation exposure by the human receptor.
Multiplication of the vapor/particulate ratio by the emission factor of each dioxin like
congener yields the portion of emission in vapor state and that which is associated with
particulate matter emissions. Tables 6-4, 6-5, and 6-7 are the average emission factors
(grams/Metric ton) of congeners of PCDDs, PCDFs, and PCBs, respectively. The ISCST
model accepts emission factors in units of grams per second of time. The hypothetical
MSW incinerator burns 0.03 metric tons per second (MT/sec.). Multiplication of the
average emission factors derived for specific congeners by 0.03 MT/sec. gives an emission
factor in units of grams per second. Table 6-14 displays the congener-specific emission
factors for the ESP and the dry scrubber-fabric filter control scenarios.
To estimate the wet and dry deposition flux of the emitted dioxin-like congener, it is
necessary to partition the rate of emission by the generalized particle size categories
derived for the particulate matter emissions from the hypothetical MSW incinerator. Three
general particle size categories are assumed: > 10 //m; 2 to 10 jjm; < 2//m. For the
three general particle size categories, the fraction of total surface area available for
chemical adsorption is calculated to be: 0.03 for > 10 //m; 0.095 for 2 to 10 fjm; and
0.875 for less than 2 //m. Multiplication of the emission rate of the specific congener
associated with particulate for each pollution control scenario (Table 6-14) by the fraction
of available surface area gives and estimation of the emission rate of the congener per
particle size. The emission rate (grams/sec.) of the specific congeners by particle size is
shown in Table 6-15 for both the ESP and the dry-scrubber-fabric filter control scenarios of
the hypothetical MSW incinerator. These emission rates by particle size are inputs to the
ISCST model for calculations of deposition flux.
6-38 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-14. Congener-specific emission factors (grams/sec) as input to the ISC air
dispersion model for both ESP and DSFF control scenarios
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1, 2,3,4,7, 8-HxCDD
1,2,3, 6,7, 8-HxCDD
1,2,3,7,8,9-HxCDD
1, 2,3,4,6, 7,8-HpCDD
Octa-CDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1, 2,3,4,7, 8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2, 3,4,6, 7,8-HpCDF
1, 2,3,4,7, 8,9-HpCDF
OctaCDF
3,3',4,4'-TeCB
3,4,4', 5-TeCB
3,3',4,4',5-PeCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
3,3',4,4',5,5'- HxCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5,5'-HpCB
ESP
Vapor
2.62E-08
2.39E-07
5.63E-08
1.37E-07
6.96E-08
1.27E-07
9.84E-07
2.95E-07
3.10E-07
2.81E-07
1 .94E-07
1.52E-07
1.29E-07
1.23E-07
1.55E-07
4.88E-09
4.52E-07
6.60E-07
6.60E-07
6.60E-10
6.60E-10
6.60E-10
6.60E-10
6.00E-07
6.00E-07
6.00E-07
6.00E-07
9.36E-08
control
Particulate
1.05E-07
9.55E-07
2.25E-07
5.50E-07
2.78E-07
5.06E-07
3.94E-06
1.18E-06
1.24E-06
1.12E-06
7.75E-07
6.10E-07
5.16E-07
4.92E-07
6.22E-07
1.95E-08
1.81E-06
2.64E-06
2.64E-06
2.64E-09
2.64E-09
2.64E-09
2.64E-09
2.40E-06
2.40E-06
2.40E-06
2.40E-06
3.74E-07
DSFF
Vapor
1.64E-09
1.49E-08
3.52E-09
8.59E-09
4.35E-09
7.91E-09
6.15E-08
1 .84E-08
1.94E-08
1.76E-08
1.21E-08
9.53E-09
8.06E-09
7.69E-09
9.71E-09
3.05E-10
2.83E-08
4.13E-08
4.13E-08
4.13E-11
4.13E-11
4.13E-11
4.13E-11
3.75E-08
3.75E-08
3.75E-08
3.75E-08
5.85E-09
control
Particulate
6.54E-09
5.97E-08
1.41E-08
3.44E-08
1 .74E-08
3.17E-08
2.46E-07
7.37E-08
7.76E-08
7.02E-08
4.85E-08
3.81E-08
3.23E-08
3.08E-08
3.89E-08
1.22E-09
1.13E-07
1.65E-07
1.65E-07
1.65E-10
1.65E-10
1.65E-10
1.65E-10
1.50E-07
1.50E-07
1.50E-07
1.50E-07
2.34E-08
6-39
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-15. Congener-specific emission factors by particle size (grams/sec) for both the
ESP and dry scrubber-fabric filter (DSFF) control scenarios to the hypothetical MSW
incinerator
ESP scenario DSFF scenario
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OctaCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OctaCDF
3,3',4,4'-TeCB
3,4,4', 5-TeCB
3,3',4,4',5-PeCB
2,3,3',4,4'-PeCB
2,3,4,4', 5-PeCB
2,3',4,4',5-PeCB
3,3',4,4',5,5'-HxCB
2,3,3',4,4',5'-HxCB
2,3,3', 4,4', 5'-HxCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5,5'-HpCB
< 2//m
9.16E-08
8.36E-07
1.97E-07
4.81E-07
2.44E-07
4.43E-07
3.44E-06
1 .03E-06
1 .09E-06
9.83E-07
6.78E-07
5.33E-07
4.52E-07
4.31E-07
5.44E-07
1.71E-08
1.58E-06
2.31E-06
2.31E-06
2.31E-09
2.31E-09
2.31E-09
2.31E-09
2.10E-06
2.10E-06
2.10E-06
2.10E-06
3.28E-07
2 - 1 0 fjm
9.94E-09
9.07E-08
2.14E-08
5.22E-08
2.64E-08
4.81E-08
3.74E-07
1.12E-07
1.18E-07
1.07E-07
7.36E-08
5.79E-08
4.90E-08
4.67E-08
5.91E-08
1.86E-09
1 .72E-07
2.51E-07
2.51E-07
2.51E-10
2.51E-10
2.51E-10
2.51E-10
2.28E-07
2.28E-07
2.28E-07
2.28E-07
3.56E-08
> 10/vm
3.14E-09
2.87E-08
6.75E-09
1.65E-08
8.35E-09
1.52E-08
1.18E-07
3.54E-08
3.72E-08
3.37E-08
2.33E-08
1.83E-08
1.55E-08
1 .48E-08
1 .86E-08
5.86E-10
5.43E-08
7.92E-08
7.92E-08
7.92E-11
7.92E-11
7.92E-11
7.92E-11
7.20E-08
7.20E-08
7.20E-08
7.20E-08
1.12E-08
< 2/ym
5.72E-09
5.22E-08
1.23E-08
3.01 E-08
1.52E-08
2.77E-08
2.15E-07
6.44E-08
6.79E-08
6.14E-08
4.24E-08
3.33E-08
2.82E-08
2.69E-08
3.40E-08
1.07E-09
9.90E-08
1.44E-07
1 .44E-07
1.44E-10
1.44E-10
1.44E-10
1.44E-10
1.31E-07
1.31E-07
1.31E-07
1.31E-07
2.05E-08
2 - 1 0 fjm
6.21E-10
5.67E-09
1 .34E-09
3.26E-09
1.65E-09
3.01E-09
2.34E-08
7.00E-09
7.37E-09
6.67E-09
4.60E-09
3.62E-09
3.06E-09
2.92E-09
3.69E-09
1.16E-10
1 .07E-08
1.57E-08
1.57E-08
1.57E-11
1.57E-11
1.57E-11
1.57E-11
1 .43E-08
1.43E-08
1.43E-08
1 .43E-08
2.22E-09
> 10//m
1.96E-10
1.79E-09
4.22E-10
1 .03E-09
5.22E-10
9.50E-10
7.38E-09
2.21E-09
2.33E-09
2.11E-09
1 .45E-09
1.14E-09
9.68E-10
9.23E-10
1.17E-09
3.66E-11
3.39E-09
4.95E-09
4.95E-09
4.95E-12
4.95E-12
4.95E-12
4.95E-12
4.50E-09
4.50E-09
4.50E-09
4.50E-09
7.02E-10
6-40 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
6.7. RESULTS OF THE AIR DISPERSION MODELING OF CONGENER-SPECIFIC EMISSIONS
FROM THE HYPOTHETICAL MSW INCINERATOR IN TAMPA, FLORIDA
The dispersion and deposition computations performed by the ISCST model require
data on wind speed, wind direction, and the intensity of atmospheric turbulence. The
turbulence intensity is represented by the atmospheric stability class. These data are
provided by a permanent data base in the ISCST model known as the Stability Array or
STAR data set. Martin and Tikvart (1968) developed the STAR program from routinely
collected meteorological data to generate frequencies and percentage frequencies of wind
direction by speed classes for each stability category. The specifications of stability
categories depending on wind speed and sky cover were established by Pasquill (1961),
and later modified by Turner (1964). The program was adopted for use at the National
Climatic Center (NCC), where archived records of all national reporting weather stations
are kept. The most current version of the STAR data from all STAR stations in the
country were used to produce the matrices of STAR frequencies used in the ISCST
(Martin,1968). The format of the STAR data includes sixteen wind directions, six wind
speed classes, and seven stability categories with categories A,B,C, and Dday in the
daytime, and categories Dnight, E, and F in the nighttime. There are data for 312 stations
geographically distributed in the U.S.
Meteorological data recorded at the Tampa Airport STAR station was used in the
dispersion modeling of stack gas and particulate emissions of dioxin-like congeners from
the hypothetical MSW incinerator. Inputs for the model included hourly meteorological
data, characteristics and receptor features. Hourly meteorological data requirements are
the mean wind speed, the direction towards which the wind is blowing, the wind-profile
exponent, the ambient air temperature, the Pasquill stability category, the vertical potential
temperature gradient with height, the mixing layer height, and the frequency distribution of
hourly precipitation. Source input data requirements included the congener-specific mass
emission rate partitioned by vapor and particulate; the physical stack measurements, e.g.,
diameter, base elevation of the stack, and exit velocity and temperature of the stack gas;
dimensions of the incinerator building, and settling parameters for particulate matter for
both dry and wet deposition.
The output of the ISCLT model for both surface deposition and ambient air impacts
is a concentration array for 160 ground-level receptors around the incinerator, e.g., 10
6-41 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
receptor points along each of the 16 wind directions every 22.5° on the polar azimuth.
For the assumed vaporous state of the congener emissions the ambient air, ground-level
concentrations are in units of grams per cubic meter of air (g/m3), and for the assumed
portion of emissions associated with particulate matter the concentration estimates are in
units of grams per square meter of surface area (g/m2). Because these concentrations are
expressed as annual averages using annual average meteorological data, the concentration
units are g/m3-yr., and g/m2-yr. Ground-level concentrations for both ambient air and
surface deposition were estimated at concentric radial distances from the incinerator of
0.2, 0.5, 0.8, 1.0, 2.0, 5.0, 10, 20, 30, and 40 kilometers. Ambient air and surface
deposition modeling of emissions for the hypothetical incinerator located in Tampa, Florida
were estimated for both pollution control scenarios: (1) ESP; (2) dry scrubber combined
with fabric filter (DSFF).
The air dispersion modeling for both the ESP and DSFF control scenario shows that
the maximum ambient air and surface deposition receptor concentration is 200 meters
east of the incinerator, corresponding to a westerly wind direction. If this point of impact
is referenced as 1.0, then the concentration gradient at each concentric radial distance in
the easterly direction is a fraction of 1.0. Figures 6-2, and 6-3 display the ratio of
concentration beyond the 200 meters maximum ground-level concentration resulting from
the ESP and DSFF control scenarios, respectively. For example, at the radial distance of
0.8 km in the ESP scenario, approximately 0.5 (50 %) of the maximum ground-level
concentration at 0.2 km is estimated by the ISCST model.
Ambient air concentrations of dioxin-like congeners at ground-level are expressed in
units of grams per cubic meter of air (g/m3), and are annual average concentrations. The
ambient air concentration results from the emission of specific congeners in the vapor
state, and it is assumed that the congener will remain in the vapor state, or be associated
with aerosol particles less than 1.0 //m in diameter. The annual average ambient air
concentration is directly available for human inhalation exposure. The maximum ground-
level ambient air concentration of all modeled congeners is estimated to occur 200 meters
in a westerly wind direction from the center of the stack. As Figures 6-2 and 6-3 indicate,
most of the ground-level impact associated with the stack emissions from the hypothetical
MSW incinerator will occur within 5 kilometers from the facility for both control scenarios.
Tables 6-16 and 6-17 display the annual average ambient air concentrations of specific
6-42 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
.0
•a
ro
t_
o
c
o
10
(_
0)
a
tn
0.6-
0.6-
0.4
0.2 H
0.0
0 2
11 i ' ' i
6 8
Kilometers
!0 12
Figure 6-2. Dispersion gradient of the ground-level concentration with distance from the
incinerator-ESP control scenario.
c
ro
•a
ro
L_
O
c
o
to
t_
0)
a
(O
1 0
08-
06-
04-
02
00
0
nr
4
T
8
Kilometers
Figure 6-3. Dispersion gradient of the ground-level concentration with distance from the
incinerator-DSFF control scenario.
6-43
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-16. Predicted annual average ambient air concentrations (grams/m3) of specific
dioxin-like congeners at specified distances from the hypothetical MSW incinerator--ESP
controls
Ambient air concentrations (grams/m3)-ESP controls
Congeners
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3, 4,7, 8-HxCDD
1,2,3, 6,7, 8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OctaCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3, 4,7, 8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OctaCDF
3,3',4,4'-TeCB
3,4,4', 5-TeCB
3,3',4,4',5-PeCB
2,3,3',4,4'-PeCB
2,3,4,4', 5-PeCB
2,3',4,4',5-PeCB
3,3',4,4',5,5'-HxCB
2,3,3',4,4',5-HxCB
2,3',4,4',5,5-HxCB
2,3',4,4',5,5'-HxCB
2,3,3',4,4',5,5'-HCB
<
1
1
3
8
4
8
6
1
1
1
1
9
8
7
9
3
2
4
4
4
4
4
4
3
3
3
3
5
D.2 km
.65E-15
.51E-14
.55E-15
.68E-15
.40E-15
.OOE-15
.21E-14
.86E-14
.96E-14
.77E-14
.22E-14
.63E-15
.15E-15
.77E-15
.81E-15
.08E-16
.86E-14
.17E-14
.17E-14
.17E-17
.17E-17
.17E-17
.17E-17
.79E-14
.79E-14
.79E-14
.79E-14
.91E-15
i
9
9
2
5
2
4
3
1
1
1
7
5
4
4
5
1
1
2
2
2
2
2
2
2
2
2
2
3
0.5 km
.91E-16
.05E-15
.13E-15
.21E-15
.64E-15
.80E-15
.73E-14
.12E-14
.18E-14
.06E-14
.34E-15
.78E-15
.89E-15
.66E-15
.89E-15
.85E-16
.71E-14
.50E-14
.50E-14
.50E-17
.50E-17
.50E-17
.50E-17
.27E-14
.27E-14
.27E-14
.27E-14
.55E-15
0.8km
8.26E-16
7.54E-15
1.78E-15
4.34E-15
2.20E-15
4.00E-15
3.11E-14
9.30E-15
9.80E-15
8.87E-15
6.12E-15
4.81E-15
4.07E-15
3.88E-15
4.91E-15
1.54E-16
1.43E-14
2.08E-14
2.08E-14
2.08E-17
2.08E-17
2.08E-17
2.08E-17
1.89E-14
1.89E-14
1.89E-14
1.89E-14
2.96E-15
7
6
1
3
2
3
2
8
9
8
5
4
3
3
4
1
1
1
1
1
1
1
1
1
1
1
1
2
1 km
.60E-16
.94E-15
.64E-15
.99E-15
.02E-15
.68E-15
.86E-14
.56E-15
.01E-15
.16E-15
.63E-15
.43E-15
.75E-15
.57E-15
.51E-15
.42E-16
.31E-14
.92E-14
.92E-14
.92E-17
.92E-17
.92E-17
.92E-17
.74E-14
.74E-14
.74E-14
.74E-14
.72E-15
6
6
1
3
1
3
2
7
7
7
4
3
3
3
3
1
1
1
1
1
1
1
1
1
1
1
1
2
2 km
.61E-16
.03E-15
.42E-15
.47E-15
.76E-15
.20E-15
.49E-14
.44E-15
.84E-15
.09E-15
.90E-15
.85E-15
.26E-15
.11E-15
.93E-15
.23E-16
.14E-14
.67E-14
.67E-14
.67E-17
.67E-17
.67E-17
.67E-17
.52E-14
.52E-14
.52E-14
.52E-14
.36E-15
3
3
7
1
8
1
1
3
3
3
2
1
1
1
1
6
5
8
8
8
8
8
8
7
7
7
7
1
5 km
.30E-16
.02E-15
.11E-16
.74E-15
.79E-16
.60E-15
.24E-14
.72E-15
.92E-15
.55E-15
.45E-15
.93E-15
.63E-15
.55E-15
.96E-15
.17E-17
.71E-15
.34E-15
.34E-15
.34E-18
.34E-18
.34E-18
.34E-18
.58E-15
.58E-15
.58E-15
.58E-15
.18E-15
6-44 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-17. Predicted annual average ambient air concentrations (grams/m3) of specific
dioxin-like congeners at specified distances from the hypothetical MSW incinerator-DSFF
controls
Ambient air concentrations (grams/m3)-DSFF controls
Congeners
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OctaCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OctaCDF
3,3',4,4'-TeCB
3,4,4', 5-TeCB
3,3',4,4',5-PeCB
2,3,3',4,4'-PeCB
2,3,4,4', 5-PeCB
2,3',4,4',5-PeCB
3,3',4,4',5,5'-HxCB
2,3,3',4,4',5-HxCB
2,3',4,4',5,5-HxCB
2,3',4,4',5,5'-HxCB
2,3,3',4,4',5,5'-HCB
0.2 km
1.03E-16
9.43E-16
2.22E-16
5.42E-16
2.75E-16
5.00E-16
3.88E-15
1.16E-15
1.22E-15
1.11E-15
7.65E-16
6.02E-16
5.09E-16
4.86E-16
6.13E-16
1.93E-17
1.79E-15
2.61E-15
2.61E-15
2.61E-18
2.61E-18
2.61E-18
2.61E-18
2.37E-15
2.37E-15
2.37E-15
2.37E-15
3.69E-16
0.5 km
6.20E-17
5.66E-16
1.33E-16
3.25E-16
1.65E-16
3.00E-16
2.33E-15
6.98E-16
7.35E-16
6.65E-16
4.59E-16
3.61E-16
3.06E-16
2.91E-16
3.68E-16
1.16E-17
1.07E-15
1.56E-15
1.56E-15
1.56E-18
1.56E-18
1.56E-18
1.56E-18
1.42E-15
1.42E-15
1.42E-15
1.42E-15
2.22E-16
0.8km
5.16E-17
4.71E-16
1.11E-16
2.71E-16
1.37E-16
2.50E-16
1.94E-15
5.81E-16
6.12E-16
5.54E-16
3.83E-16
3.01 E-1 6
2.55E-16
2.43E-16
3.07E-16
9.64E-18
8.93E-16
1.30E-15
1.30E-15
1.30E-18
1.30E-18
1.30E-18
1.30E-18
1.18E-15
1.18E-15
1.18E-15
1.18E-15
1.85E-16
1 km
4.75E-17
4.34E-16
1.02E-16
2.49E-16
1.26E-16
2.30E-16
1.79E-15
5.35E-16
5.63E-16
5.10E-16
3.52E-16
2.77E-16
2.34E-16
2.23E-16
2.82E-16
8.87E-18
8.21E-16
1.20E-15
1.20E-15
1.20E-18
1.20E-18
1.20E-18
1.20E-18
1.09E-15
1.09E-15
1.09E-15
1.09E-15
1.70E-16
2 km
4.13E-17
3.77E-16
8.89E-17
2.17E-16
1.10E-16
2.00E-16
1.55E-15
4.65E-16
4.90E-16
4.43E-16
3.06E-16
2.41E-16
2.04E-16
1.94E-16
2.45E-16
7.71E-18
7.14E-16
1.04E-15
1.04E-15
1.04E-18
1.04E-18
1.04E-18
1.04E-18
9.47E-16
9.47E-16
9.47E-16
9.47E-16
1.48E-16
5 km
2.07E-17
1.89E-16
4.44E-17
1.08E-16
5.49E-17
9.99E-17
7.77E-16
2.33E-16
2.45E-16
2.22E-16
1.53E-16
1.20E-16
1.02E-16
9.71E-17
1.23E-16
3.86E-18
3.57E-16
5.21E-16
5.21E-16
5.21E-19
5.21E-19
5.21E-19
5.21E-19
4.74E-16
4.74E-16
4.74E-16
4.74E-16
7.39E-17
6-45 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
dioxin-like congeners emitted from the stack under the ESP and DSFF control scenarios.
Surface deposition flux is expressed in units of grams per square meter per year.
The deposition flux is the summation of both dry and wet removal processes, and is an
annual average concentration. Tables 6-18 and 6-19 display the spacial deposition flux in
the vicinity of the incinerator in the ESP and DSFF control scenarios.
6-46 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-18. Wet + dry surface deposition (grams/m2.yr) of specific congeners at specified
distances in the vicinity of the hypothetical MSW incinerator--ESP controls
Surface deposition (grams/m2'yr)--ESP control scenario
Congeners
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OctaCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1 ,2,3,4,7, 8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1, 2,3,4,7, 8,9-HpCDF
OctaCDF
3,3',4,4'-TeCB
3,4,4', 5-TeCB
3,3',4,4',5-PeCB
2,3,3',4,4'-PeCB
2,3,4,4', 5-PeCB
2,3',4,4',5-PeCB
3,3',4,4',5,5'-HxCB
2,3,3',4,4',5-HxCB
2,3',4,4',5,5-HxCB
2,3',4,4',5,5'-HxCB
2,3,3',4,4',5,5'-HpCB
0.2 km
3.52E-08
3.21E-07
7.57E-08
1 .85E-07
9.36E-08
1.70E-07
1.32E-06
3.96E-07
4.17E-07
3.78E-07
2.61E-07
2.05E-07
1.74E-07
1.65E-07
2.09E-07
6.57E-09
6.09E-07
8.88E-07
8.88E-07
8.88E-10
8.88E-10
8.88E-10
8.88E-10
8.07E-07
8.07E-07
8.07E-07
8.07E-07
1.26E-07
0.5 km
2.11E-08
1 .93E-07
4.54E-08
1.11E-07
5.62E-08
1.02E-07
7.94E-07
2.38E-07
2.50E-07
2.27E-07
1.56E-07
1.23E-07
1 .04E-07
9.93E-08
1 .25E-07
3.94E-09
3.65E-07
5.33E-07
5.33E-07
5.33E-10
5.33E-10
5.33E-10
5.33E-10
4.84E-07
4.84E-07
4.84E-07
4.84E-07
7.56E-08
0.8 km
1.76E-08
1.61E-07
3.79E-08
9.24E-08
4.68E-08
8.52E-08
6.62E-07
1.98E-07
2.09E-07
1.89E-07
1 .30E-07
1 .03E-07
8.68E-08
8.27E-08
1.05E-07
3.29E-09
3.04E-07
4.44E-07
4.44E-07
4.44E-10
4.44E-10
4.44E-10
4.44E-10
4.04E-07
4.04E-07
4.04E-07
4.04E-07
6.30E-08
1 km
1.62E-08
1 .48E-07
3.48E-08
8.50E-08
4.31E-08
7.84E-08
6.09E-07
1.82E-07
1.92E-07
1 .74E-07
1 .20E-07
9.43E-08
7.98E-08
7.61E-08
9.62E-08
3.02E-09
2.80E-07
4.08E-07
4.08E-07
4.08E-10
4.08E-10
4.08E-10
4.08E-10
3.71E-07
3.71E-07
3.71E-07
3.71E-07
5.79E-08
2 km
1.41E-08
1.29E-07
3.03E-08
7.39E-08
3.75E-08
6.81 E-08
5.30E-07
1 .59E-07
1 .67E-07
1.51E-07
1.04E-07
8.20E-08
6.94E-08
6.62E-08
8.36E-08
2.63E-09
2.43E-07
3.55E-07
3.55E-07
3.55E-10
3.55E-10
3.55E-10
3.55E-10
3.23E-07
3.23E-07
3.23E-07
3.23E-07
5.04E-08
5 km
7.04E-09
6.43E-08
1.51 E-08
3.70E-08
1.87E-08
3.41 E-08
2.65E-07
7.93E-08
8.35E-08
7.56E-08
5. 21 E-08
4.10E-08
3.47E-08
3. 31 E-08
4.18E-08
1.31E-09
1.22E-07
1.78E-07
1.78E-07
1.78E-10
1.78E-10
1.78E-10
1.78E-10
1.61E-07
1.61E-07
1.61E-07
1.61E-07
2.52E-08
6-47 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 6-19. Wet + dry surface deposition (grams/m2.yr) of specific congeners at specified
distances in the vicinity of the hypothetical MSW incinerator-DSFF controls
Surface deposition (grams/m2.yr)-DSFF control scenario
Congeners
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OctaCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1, 2,3,7, 8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3, 4,7, 8,9-HpCDF
OctaCDF
3,3',4,4'-TeCB
3,4,4', 5-TeCB
3,3',4,4',5-PeCB
2,3,3',4,4'-PeCB
2,3,4,4', 5-PeCB
2,3',4,4',5-PeCB
3,3',4,4',5,5'-HxCB
2,3,3',4,4',5-HxCB
2,3',4,4',5,5-HxCB
2,3',4,4',5,5'-HxCB
2,3,3',4,4',5,5'-HCB
0.2 km
2.20E-09
2.01E-08
4.73E-09
1.16E-08
5.85E-09
1 .06E-08
8.27E-08
2.48E-08
2.61E-08
2.36E-08
1.63E-08
1.28E-08
1.08E-08
1.03E-08
1.31E-08
4.11E-10
3.80E-08
5.55E-08
5.55E-08
5.55E-11
5.55E-11
5.55E-11
5.55E-11
5.05E-08
5.05E-08
5.05E-08
5.05E-08
7.87E-09
0.5 km
1.32E-09
1.20E-08
2.84E-09
6.93E-09
3.51E-09
6.39E-09
4.96E-08
1.49E-08
1.57E-08
1 .42E-08
9.78E-09
7.69E-09
6.51E-09
6.21E-09
7.84E-09
2.46E-10
2.28E-08
3.33E-08
3.33E-08
3.33E-11
3.33E-11
3.33E-11
3.33E-11
3.03E-08
3.03E-08
3.03E-08
3.03E-08
4.72E-09
0.8km
1.10E-09
1 .OOE-08
2.37E-09
5.78E-09
2.93E-09
5.32E-09
4.14E-08
1.24E-08
1.30E-08
1.18E-08
8.15E-09
6.41 E-09
5.42E-09
5.17E-09
6.53E-09
2.05E-10
1 .90E-08
2.78E-08
2.78E-08
2.78E-11
2.78E-11
2.78E-11
2.78E-11
2.52E-08
2.52E-08
2.52E-08
2.52E-08
3.94E-09
1 km
1.01 E-09
9.24E-09
2.18E-09
5. 31 E-09
2.69E-09
4.90E-09
3.81E-08
1.14E-08
1 .20E-08
1 .09E-08
7.50E-09
5.90E-09
4.99E-09
4.76E-09
6. 01 E-09
1.89E-10
1.75E-08
2.55E-08
2.55E-08
2.55E-11
2.55E-11
2.55E-11
2.55E-11
2.32E-08
2.32E-08
2.32E-08
2.32E-08
3.62E-09
2 km
8.80E-10
8.03E-09
1.89E-09
4.62E-09
2.34E-09
4.26E-09
3.31E-08
9. 91 E-09
1.04E-08
9.45E-09
6.52E-09
5.13E-09
4.34E-09
4.14E-09
5.23E-09
1.64E-10
1.52E-08
2.22E-08
2.22E-08
2.22E-11
2.22E-11
2.22E-11
2.22E-11
2.02E-08
2.02E-08
2.02E-08
2.02E-08
3.15E-09
5 km
4.40E-10
4.02E-09
9.47E-10
2.31 E-09
1.17E-09
2.13E-09
1 .65E-08
4.95E-09
5.22E-09
4.72E-09
3.26E-09
2.56E-09
2.17E-09
2.07E-09
2. 61 E-09
8.21E-11
7. 61 E-09
1.11E-08
1.11E-08
1.11E-11
1.11E-11
1.11E-11
1.11E-11
1.01E-08
1.01E-08
1.01E-08
1 .01 E-08
1.57E-09
6-48 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
REFERENCES FOR CHAPTER 6
ASME. (1981) Study on state-of-the art of dioxin from combustion sources, Arthur D.
Little, Inc. for the American Society of Mechanical Engineers, New York, NY.
Briggs, G.A. (1979) Plume rise. USAEC Critical Review Series. NTIS publication no. TID-
25075.
Briggs, G.A. (1975) Plume rise predictions. In: Lectures on air pollution and environmental
impact analyses, American Meteorology Society.
Bruce, K.R.; Beach, L.O.; Gullet, B.K. (1991) The role of gas-phase CI2 in the formation of
PCDD/PCDF during waste combustion. Waste Management 11: 97-102.
Brunner, C. (1984) Incineration systems, selection and design. New York, NY: Van
Nostrand Reinhold Co.
Choudhry,G.G.; Hutzinger, 0. (1983) Mechanistic aspects of the thermal formation of
halogenated organic compounds including polychlorinated dibenzo-p-dioxins. New
York, NY: Gordon and Breach.
Chrostowski, P.C.; Sager, S.L. (1991) Management of ash from municipal solid waste
combustion.In: Hattermer-Frey, H.A.; Travis,C., eds.. Health effects of municipal
waste incineration. Boca Raton,FL: CRC Press, pp. 249-264.
Cleverly, D.H. (1984) Chlorinated dibenzo-p-dioxins and furans in incineration of
municipal solid waste. Proceedings of workshop on energy from municipal waste
research: A technical review of thermochemical systems, February 22-24,1984,
Kissimmee, FL. Argonne National Lab, U.S. Department of Energy, pp. 295-319.
Cleverly, D.H.; Morrison, R.M.; Riddle, B.L.; Kellam, R.G. (1989) Regulatory analysis of
pollutant emissions, including polychlorinated dibenzo-p-dioxins (CDDs) and
dibenzofurans (CDFs), from the stacks of municipal waste combustors.
Chemosphere 18: 1143-1153.
Cleverly, D.H.; Morrison, R.M.; Riddle, B.L.; Kellam, R.G. (1991) Regulatory analysis of
pollutant emissions, including polychlorinated dibenzo-p-dioxins (CDDs) and
dibenzofurans (CDFs), from municipal waste combustors. In: Hattermer-Frey, H.A.;
Travis,C., eds.. Health effects of municipal waste incineration. Boca Raton, FL:
CRC Press, pp. 47-65.
Commoner, B.; Shapiro, K.; Webster, T. (1987) The origin and health risks of PCDD and
PCDF. Waste Management Res. 5: 327-346.
Dumbauld, R.K.; Rafferty, J.E.; Cramer, H.E. (1976) Dispersion deposition from aerial
spray releases. Preprint volume for the Third Symposium on Atmospheric Diffusion
and Air Quality. American Meteorolical Society.
6-49 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Environment Canada. (1985) The national incineration testing and evaluation program:
Two-stage combustion (Prince Edward Island) Environment Canada, Ottawa,
Ontario, Canada, Report EPS 3/UP/1, September, 1985.
Entropy. (1987) Stationary source sampling report. Signal Resco Pinnellas County
resource recovery facility, St. Petersburg, FL. EEI, REF. #5286-6, Entropy
Environmental Inc., Research Triangle Park, NC, September, 1987.
Gardiner, W. (1982) The chemistry of flames. Sci. Am. February: 112.
Gullet, B.K.; Bruce, K.R.; Beach., L.O. (1990a) Formation of chlorinated organics during
solid waste combustion. Waste Management Res. 8: 203-214.
Gullett, B.K.; Bruce, K.R., Beach, L.O. (1990b) The effect of metal catalysts on the
formation of polychlorinated dibenzo-p-dioxin and polychlorinated dibenzofuran
precursors. Chemosphere 20: 1945-1952.
Gullet, B.K.; Bruce, K.R.; Beach. L.O. (1991a) The effect of sulfur compounds on the
formation mechanism of PCDD and PCDF in municipal waste combustors. In:
Conference papers from the second international conference on municipal waste
combustion, April 1 5-19, 1991, Tampa, FL. Air and Waste Management
Association, Pittsburgh, PA. pp. 16-34.
Gullet, B.K.; Bruce, K.R.; Beach. L.O.; Drago,A. (1991b) Mechanistic steps in the
production of PCDD and PCDF during waste combustion. In: Abstracts of the 11th
international symposium on chlorinated dioxins and related compounds, September
23-27, 1991, Research Triangle Park, NC, University of North Carolina, Chapel Hill,
NC, p. 298.
Hagemaier, H.; Brunner, H.; Haag, R.; Kraft, M.; Lutzke, K. (1987) Problems associated
with the measurement of PCDD and PCDF emissions from waste incineration
plants. Waste Management Res. 5: 239-250.
Hahn, J.L. (1986) Air emissions testing at the Wurzburg, West Germany waste-to-energy
facility. Presented at the annual meeting of the Air Pollution Control Association;
June.
Hart, F.C. (1984) Assessment of potential public health impacts associated with predicted
emissions of polychlorinated dibenzodioxins and polychlorinated dibenzofurans from
the Brooklyn Navy Yard resource recovery facility. Prepared for the New York City
Department of Sanitation, Fred C. Hart, Associates, New York, NY.
Hay, D.J.; Finkelstein, A.; Klicius, R. (1986) The national incineration testing and
evaluation program: An assessment of two-stage incineration and pilot scale
emission control. Paper presented to U.S. EPA Science Advisory Board,
Washington,DC. Environment Canada, Ottawa, Ontario, Canada. June, 1986.
6-50 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Kapahi, R. (1991) Modeling the dispersion of toxic air pollutants emitted from municipal
solid waste incinerators. In: Hattermer-Frey, H.A.; Travis,C., eds.. Health effects of
municipal waste incineration. Boca Raton, FL: CRC Press, pp. 67-82.
Knisley, D.R.; Jamgochian, C.G.; Gergen, W.P.; Holder, D.J. (1986) Draft emissions test
report dioxin/furans and total organic chlorides emission testing. Saugus resource
recovery facility, Saugus, MA. Prepared by Radian Corporation for Office of Air
Quality Planning and Standards, U.S. EPA, Research Triangle Park, NC, DCN No.
86-223,015-05. October 2, 1986.
Magagni, A.; Boschi, G.; Cocheo, V. (1990) Emissions of a MSW incinerator equipped
with a post-combustion chamber, dry scrubber and ESP. Chemosphere 20: 1882-
1890.
Marklund, S.; Kjeller, L.O.; Hansson, M.; Tysklind, M.; Rappe, C. (1985) Determination of
PCDDs and PCDFs in incineration samples and pyrolytic products. Presented at
American Chemical Society National meeting, Chlorinated dioxins and dibenzofurans
in the total environment, Miami, Fl,April, 1985.
Martin, D.O.; Tikvart, J.A. (1968) A general atmospheric diffusion model for estimating
the effects on air quality of one or more sources. Presented at the 61st annual air
pollution control association meeting, St. Paul, MN. June, 1968.
MRI. (1989) Validation of emission test method for PCDDs and PCDFs. Prepared by
Midwest Research Institute, Kansas City, MO., for Invironmental Monitoring
Systems Laboratory, U.S. EPA, Research Triangle Park, NC. EPA contract # 68-02-
4395. February 24, 1989.
Mukerjee, D.; Cleverly, D.H. (1987) Risk from exposure to polychlorinated dibenzo-p-
dioxins and dibenzofurans emitted from municipal incinerators. Waste Management
Res. 5:269-283.
OTA. (1989) Facing America's trash. What next for municipal solid waste? Office of
Technology Assessment,U.S.Congress, Washington, DC.
Pasquill, F. (1961) The estimation of the dispersion of windborne material. Meteorology
90: 33-49.
Radke, L.F.; Hobbs, P.V.; Eltgroth, M.W. (1980) Scavenging of aerosol particles by
precipitation. J. Applied Meteorol. 19: 715-722.
Seelinger, R.; Hahn, J.; VonDemfange, H.P.; Zurlinden, R.A. (1986) Environmental test
report, Ogden Martin Systems of Tulsa, Inc., Compliance with Tulsa City/County
health department and U.S. EPA permits. September 9, 1986.
Seinfeld, J.H. (1986) Atmospheric chemistry and physics of air pollution. New York, NY:
John Wiley and Sons.
6-51 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Siebert, P.C.; Alston-Gulden, D.; Jones, K.H. (1991) An analysis of worldwide resource
recovery emissions data and the implications for risk assessment. In: Hattermer-
Frey, H.A.; Travis,C., eds., Health effects of municipal waste incineration. Boca
Raton, FL:, CRC Press, pp. 2-46.
Turner, D.B. (1964) A diffusion model for an urban area. Atmospheric dispersion
estimates. 6th printing. Washington, DC: U.S. Government Printing Office,
Publication no. AP-26.
U.S. Environmental Protection Agency. (1980) Environmental assessment of a waste-to-
energy process. Braintree municipal incinerator. Office of Research and
Development, Washington, DC., EPA-600/7-80-149.
U.S. Environmental Protection Agency. (1983) Comprehensive assessment of the specific
compounds present in combustion processes. Volume 1: Pilot study of combustion
emission variability. Office of Toxic Substances, Washington, DC, EPA-560/5-83-
004, June, 1983.
U.S. Environmental Protection Agency. (1984) Thermal degradation products from
dielectric fluids. Office of Toxic Substances, Washington,DC.,EPA-560/5-84-
009,September, 1984.
U.S. Environmental Protection Agency. (1986) Industrial source complex (ISC) dispersion
model user's guide-second edition. Office of Air Quality Planning and
Standards,Research Triangle Park, NC, EPA-450/4-86-005a.
U.S. Environmental Protection Agency. (1987a) Characterization of the municipal waste
combustion industry. Office of Solid Waste and Emergency Response,
Washington,DC: EPA-530-SW-87-021h. June, 1987.
U.S. Environmental Protection Agency. (1987b) Assessment of health risks associated
with municipal waste combustion emissions. Office of Solid Waste and Emergency
Response, Office of Air and Radiation, Office of Research and Development,
Washington,DC., EPA/530-SW-87-021g, September, 1987.
U.S. Environmental Protection Agency. (1987c) Emission data base for municipal waste
combustors.Office of Solid Waste and Emergency Response, Office of Air and
Radiation, Office of Research and Development, Washington,DC., EPA/530-SW-87-
021b, June, 1987.
U.S. Environmental Protection Agency. (1987d) National dioxin study tier 4-combustion
sources. Engineering analysis report. Office of Air Quality Planning and Standards,
Research Triangle Park, NC, EPA-450/4-84-014h, September, 1987.
U.S. Environmental Protection Agency. (1988a) Municipal waste combustion
multipollutant study. Summary report. Signal Environmental Systems, Inc., North
Andover RESCO, North Andover,MA. Office of Air Quality Planning and Standards,
Research Triangle Park, NC, EMB Report No. 86-MIN-02A, March, 1988.
6-52 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
U.S. Environmental Protection Agency. (1988b) Municipal waste combustion
multipollutant study. Summary report.Marion County solid waste-to-energy facility,
Brooks, OR. Office of Air Quality Planning and Standards, Research Triangle Park,
NC, EMB Report No. 86-MIN-03A,September, 1988.
U.S. Environmental Protection Agency. (1989) Background paper, municipal waste
combustors air emission standards. Office of Air Quality Planning and Standards,
Research Triangle Park, NC. April 26, 1989.
U.S. Environmental Protection Agency. (1990) The solid waste dilemma: Solutions for the
90's. Preliminary Draft for focus group comment. Office of Solid Waste Municipal
Solid Waste Program, Washington, DC. August, 1990.
U.S. Environmental Protection Agency; Environment Canada. (1991) The environmental
characterization of RDF combustion technology. Mid-Connecticut facility,
Hartford,CT. Test program and results, volume II., June 1991.
Vogg, H.: Metzger, M.; Steiglitz, L. (1987) Recent findings on the formation and
decomposition of PCDD/PCDF in municipal solid waste incineration. Waste
Management Res. 5: 285-294.
6-53 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
7. EXPOSURE SCENARIO DEVELOPMENT
7.1. INTRODUCTION
EPA (1991 a) states, "In exposure scenario evaluation, the assessor attempts to
determine the concentrations of chemicals in a medium or location and link this information
with the time that individuals or populations contact the chemical. The set of assumptions
about how this contact takes place is an exposure scenario." These assumptions can be
made many different ways producing a wide variety of scenarios and associated exposure
levels. The number of people exposed at different levels form a distribution of exposures.
Ideally assessors would develop this entire distribution to fully describe the exposed
population. The necessary information is rarely available however, and instead EPA
(1991 a) recommends developing a central and high end scenario to provide some idea of
the possible range of exposure levels. The purpose of this chapter is to illustrate this
procedure as applied to the dioxin-like compounds. In addition this Chapter identifies the
exposure pathways which are relevant to these compounds, and provides background and
justification for the exposure parameters which were selected for the demonstrations in
Chapter 9.
7.2. STRATEGY FOR DEVISING EXPOSURE SCENARIOS
For any physical setting, a wide variety of exposure scenarios are possible. The
range of exposure levels results from a number of different factors including individual
behavior patterns, proximity of individuals to the source of contamination, the fate
characteristics of the contaminant, and others. In order to illustrate the possible range, the
assessor should try to characterize a central and high end scenario. The general strategy
recommended here for defining these scenarios is to first identify and quantify the source
of contamination. Next, the assessor should determine the geographic area that is
impacted by this source. The contaminant levels are likely to vary widely over this area.
Select locations of interest within this area such as the location of the nearest exposed
individual or most heavily populated area. For each of these locations, identify behavior
patterns which characterize central and high end exposure patterns. Central scenarios
correspond to average or median levels and high end scenarios are defined as levels above
the 90th percentile but within the actual range of exposure levels. Statistical data are
7-1 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
rarely available to precisely define such scenarios. Instead judgement is usually required to
identify behavior patterns meeting these criteria. For example, most rural areas probably
include both farming and nonfarming residents. Farmers who grow or raise much of their
own food could be selected to represent the high end scenario and those living in typical
residential areas could represent the central scenario. Alternatively, if more detail is
desired, central and high end scenarios could be defined for both segments of the
population, i.e., farmers and residents. For each scenario, determine relevant exposure
pathways and assign values for exposure parameters such as contact rate, exposure
duration, and so on, which represent average values for the type of receptor. Finally,
compute the associated exposure level. The resulting range of exposure levels for each
location can be used to illustrate the possible range of exposures.
Reference will be made in this chapter to the example scenarios found in Chapter 9
Demonstration of Methodology. Four "source categories", categories of contamination
sources described in Chapter 5, are demonstrated in Chapter 9. The on-site soil and stack
emission sources are assumed to expose a relatively large population in a rural area
containing residences and a few farms. For these sources, both central and high end
scenarios are defined in the manner outlined above. Specifically, a central scenario is
based on typical behavior at a residence and the high end is based on a farm family that
raises a portion of its own food. For the other two sources, off-site soil and ash landfill,
the principally exposed population is assumed to include individuals residing at one or two
farms located near the site. Other individuals residing further from these source categories
might be exposed via consumption of fish or water from a nearby impacted water body,
trespassing, and so on. For the demonstration scenarios in Chapter 9, only a high end
scenario was defined for the off-site and ash landfill source categories.
The methodologies of this document are site-specific; assessors need to make the
kinds of assumptions discussed here for their own source and populations of concern.
Defining a high end scenario as a farm and a central scenario as a residence was done for
demonstration purposes only. On the other hand, the example scenarios in Chapter 9
were carefully crafted to be plausible and meaningful, considering key factors such as
source strength, fate and transport parameterization, exposure parameters, and selection
of exposure pathways. For example, a beef and milk exposure pathway is demonstrated
only for the high end scenarios. Farmers raise cattle for beef and dairy products and are
7-2 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
assumed to obtain a portion of their intake of these products from their own farm,
whereas non-farming residents are assumed not to be exposed to contaminated farm
products. Key source strength terms, including stack emission rates, concentrations of
dioxin-like compounds in ash for the ash landfill source category, and plausible soil
concentrations all came an examination of numerous data sources. Introductory sections
of Chapter 9 provide a more complete description of the example scenarios.
7.3. EXPOSURE PATHWAYS
As discussed in Chapter 3, the dioxin-like compounds have been found primarily in
air, soil, sediment and biota and to a lesser extent in water. Thus, the most likely
exposure pathways are:
• Ingestion of soil, water, beef, dairy products, fish, fruit, and vegetables
• Dermal contact with soil
• Inhalation of particulates and vapors.
Exposure is calculated as the potential dose normalized over body weight and
lifetime. This value can be computed as the lifetime average daily dose (LADD) for all
exposure pathways:
LADD = (exposure media concentration x contact rate
x contact fraction x exposure duration ) / (body weight x lifetime). (7-1)
A principal reference for most of the exposure parameters in this equation (not including
the exposure media concentration) is the Exposure Factors Handbook (EPA, 1989).
Parameters relating to dermal contact are more completely described in EPA (1992). Both
references contain comprehensive literature reviews and recommendations for approaches
which consider typical and high end exposures. Their recommendations are followed for
this assessment. Details beyond the summaries provided here can be found in those
references.
Exposure factors are best determined on a site specific basis. However, generic
default values can be used as a starting point for developing site specific values or as
defaults when the data needed to further refine these estimates are not available. The
default values are expressed as a range from a central to high end value. Each term in this
7-3 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
equation is discussed briefly below.
• Lifetime: This refers to the expected lifetime of the exposed individuals and
is used to normalize the exposure estimate such that it is consistent with the
cancer slope factor. EPA (1989) recommends using the approximate United
States average of 70 years for all pathways. This value was also adopted
for the example scenarios in Chapter 9.
• Body weight: This refers to the average body weight of the exposed
individuals over the exposure period and is also used to normalize the
exposure estimate such that it is expressed in a manner consistent with the
cancer slope factor. EPA (1989) recommends using the approximate United
States average of 70 kg for all pathways involving adults and 1 6 kg for
children ingesting soil. These values were also adopted for the example
scenarios in Chapter 9.
• Exposure media concentration: This is the concentration of the contaminant
in the media of interest and should represent a temporal average over the
time of exposure. Estimation of this term was covered in Chapter 5.
• Exposure duration: This is the overall time period that individuals spend in
situations that expose them to a contaminated media. Based on EPA
(1989), a range of 9 to 30 years is recommended as default values for all
residential exposure pathways. This range corresponds to the 50th and
90th percentiles of time a person spends at one residence. In some
situations other times may be appropriate. For example farmers may have a
tendency to remain in the same house for more years than the non-farmer
resident. For the example scenarios presented in Chapter 9, and as noted in
Section 7.1. above, 9 years was adopted for residents and 20 years as the
estimate for farmers. They are characterized as central and high end
estimates of exposure duration.
• Contact rate: This is the total rate of contact with the exposure media via
ingestion, inhalation, or dermal contact. The values for this quantity vary by
pathway as discussed below.
• Contact fraction: This is the portion of contacted material that is
contaminated. It reduces the total contact rate to reflect only contact with
7-4 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
contaminated media. The contact fractions for the exposure pathways of air
inhalation and water ingestion are related to the time individuals spend at
home. Other pathways such as fish ingestion or ingestion of home grown
foods are not related to time at home. Similarly, contact fractions for
individuals exposed at work places relate largely to time spent at the work
place. EPA (1989) discusses several time use studies which can be used to
make assumptions about time spent at home (and outdoors at the home
environment) versus time spent away from home. Generally, these time use
studies asked participants to keep 24 hour diaries of all activities. Studies
summarized were national in scope, involved large numbers of individuals,
cross-sections of populations in terms of age and other factors, and up to 87
categories of activities. Results from different studies consistently indicate
that the average adult spends between 68 to 73% of time at the home
environment. The values selected for this factor for use in the example
scenarios are described below.
The parameter values for contact rate and contact fraction applicable to each of
exposure pathways are presented in Sections 7.2.1 to 7.2.7. The default values for all
exposure parameters are summarized in Table 7-1 at the end of this chapter.
7.3.1. Soil Ingestion
Soil ingestion occurs commonly among children during activities such as mouthing
of toys and other objects, nonsanitary eating habits, and inadvertent hand-to-mouth
transfers. In addition to normal soil ingestion activities, some individuals exhibit behavior
known as pica which involves intentional soil ingestion. Soil ingestion rates associated
with pica are probably much higher. No measured values for pica patterns have been
reported in the literature, though EPA (1989) reports that other assessments have
assumed values such as 5 and 10 g/day. This document considers only normal soil
ingestion among children.
To a lesser extent, soil ingestion also occurs among adults from activities such as
hand-to-mouth transfer when eating sandwiches or smoking. However, the data to
estimate the adult rate of soil ingestion is essentially unavailable, so adult soil ingestion is
not demonstrated in this assessment.
7-5 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Based on the review of literature, particularly the studies of Binder et al. (1986) and
Clausing et al. (1986), the following values for soil ingestion were suggested in EPA
(1989): average soil ingestion in the population of young normal children (under the age of
7) is estimated at approximately 0.1 to 0.2 g/d. An upper-range ingestion estimate among
children with a higher tendency to ingest soil materials, although not a pica pattern, could
be as high as 1 g/d. However, a value of 0.8 g/day is recommended for high end exposure
estimates. The values of 0.2 g/d and 0.8 g/d were the values adopted for the central and
high end exposure scenarios in Chapter 9.
EPA generally assumes a contact fraction of 1 for soil ingestion in residential
settings. This is based on the assumption all soil ingestion by children occurs at home,
and, of course, that the soil at the home is contaminated. In situations where the
contaminated area is located remote from where children live, and children have some
access to these areas (if the areas are parks or playgrounds, e.g.), lower fractions would
be appropriate. A value of 1 has been adopted in the example scenarios presented in
Chapter 9.
7.3.2. Soil Dermal Contact
The total annual dermal contact, expressed in mg/yr, is the product of three terms:
the contact rate per soil contact event, the surface area of contact, and the number of
dermal contact events per year. EPA (1992) recommends the following ranges for these
terms:
• Contact rate: 0.2 to 1.0 mg/cm2-event
• Adult surface area: 5000 to 5800 cm2
• Event frequency: 40 to 350 events/year.
An event frequency and contact rate near the upper end of these ranges may be
appropriate for an high end exposure activity pattern such as farming, where the individual
more often comes in contact with soil and may be exposed to fugitive dust emissions
while they work. An event frequency of 40 and 350, and a contact rate of 0.2 and 1.0
mg/cm2-event, are assumed for the central and high end exposure scenarios in Chapter 9.
However, the exposed surface area of 5000 cm2 may be reduced for farmers. This area
corresponds to 25% of the total body area and apparel such as short sleeves, shoes,
socks and short pants. Farmers working in the field are likely to wear long pants at least, if
7-6 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
not also long sleeves. Although EPA (1992) indicates that clothing is not always effective
in preventing dermal contact, it seems reasonable that a value of 1000 cm2 (5% of total
body area) representing hands, neck, and face might be more appropriate in a farming
scenario. Values of 5000 and 1000 cm2 were selected for the central and high end
scenarios in Chapter 9.
The considerations for contact fraction are similar to those for soil ingestion; i.e.,
that all contact occurs with contaminated soil at the residence or farm site. Accordingly a
value of 1 was selected for the example scenarios presented in Chapter 9.
One further adjustment was made for this exposure pathway. The contact as
estimated above is the amount of soil which contacts the body. EPA (1992) indicates that
only a small percent of strongly hydrophobic organic compounds such as 2,3,7,8-TCDD
are absorbed into the body from soil dermal contact. The "dermal absorption fraction"
recommended for 2,3,7,8-TCDD in EPA (1992) was 0.001 (0.1%) to 0.03 (3%). EPA
(1992) recommends using the upper end of this range for application to other dioxin-like
compounds as a conservative assumption until these compounds have been tested. An
absorption fraction of 0.03 was used for the three compounds demonstrated in Chapter 9.
The dermal contact exposure pathway was the only one in which such an absorption
fraction was used.
7.3.3. Vapor and Dust Inhalation
EPA (1989) describes derivation of the commonly used ventilation rates of 20 and
23 m3/day. As noted in that reference, these values assume 16 hours of light activity and
8 hours of resting. Other recommendations in that reference are a rate of 30 m3/day for
high end exposures, and to derive specific ventilation rates (the reference gives information
to do so) for specific activity patterns. The example scenarios of this assessment all use
20 m3/day.
An additional assumption needs to be made for the vapor and dust inhalation
pathways. This pertains to an assumption concerning the differences in air quality
between indoor and outdoor conditions. Algorithms for both paniculate and vapor-phase
air-borne concentrations of contaminants are specific to outdoor air. Hawley (1985)
assumed, based on several other studies in which measurements were made, that the
concentration of suspended particulate matter in indoor air is equal to 75% of that outside.
7-7 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Also, his report stated that most household dust is outdoor dust that is transported into
the house, and that only a small percentage is developed from sources within. He then
concluded that 80% of the indoor dust is identical in contaminant content to outdoor soil.
Refinements to the concentration of contaminants on indoor versus outdoor dust should
have a minor effect on exposure estimates. A similar trend is assumed for air-borne vapor
phase concentrations. For this reason, differences between indoor and outdoor
concentrations are not specifically considered, or equivalently, no distinctions are made for
outdoor and indoor air quality.
The contact fraction for this pathway is equal to the fraction of total inhaled air
which is contaminated. Thus it relates largely to percent of time spent in the
contaminated area. For the example exposure scenarios presented in Chapter 9, the
contact fraction corresponds to percent time at home. EPA (1989) suggests a range of
0.75 to 1.0; the lower value will be adopted as the central value used in the residence
setting. The value selected for high end scenarios will be 0.90 instead of 1.00,
recognizing that 1.00 is more likely a bounding rather than a high end estimate.
7.3.4. Water Ingestion
The water ingestion rate of 2 L/day is traditionally assumed for exposure through
drinking water. However, EPA (1989), after review of several literature sources,
concludes that 2.0 L/day may be more appropriately described as a 90% value, or a value
for high end exposure estimates. That document recommends a rate of 1.4 L/day as
representative of average adult drinking water consumption. This is the rate used for
central and high end settings in Chapter 9. Like the vapor and dust inhalation pathways,
the difference in central and high end tendencies is modeled using the contact fraction.
Again, this fraction is based on the time spent at home. The value of 0.75 is used to
model the central estimate, for the residence setting, and the value of 0.90 is used to
model the high end estimate, for the farm setting.
7.3.5. Beef and Dairy Product Ingestion
If contaminated beef or dairy products from one source are marketed along with
uncontaminated products from many sources, only a small percent of the product
consumed by an individual may be contaminated. The potential effects of such "market
7-8 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
dilution" of beef and dairy products on human exposure are discussed briefly in EPA
(1984a), at more length by Fries (1986), and at much greater length in EPA (1985) for the
particular case of cattle production in Missouri. Aspects of the beef industry in this region
specifically noted in EPA (1985) as important to estimating exposure were type of activity
within the industry (e.g. cow-calf production, "backgrounding" - preparing calves for
feedlots, feeding for slaughter), replacement rates as a function of activity, fractions of
cattle fed to maturity outside contaminated areas before slaughter, and slaughter
categories and rates relative to national figures. EPA (1984a and 1985) concluded that
dilution will vary widely between different marketing areas. EPA (1984a and 1985) and
Fries (1986) noted that the subpopulations most likely to receive high exposures are beef
producers, dairy farmers, and their direct consumers. The residents in the central
exposure scenario in Chapter 9 are not assumed to be producers or direct consumers of
farm products, and for this reason, the central estimate scenarios do not include a beef
and milk pathway. The high end farming scenarios do have these pathways, meaning that
the farmers home slaughter for beef and also obtain a portion of their milk from their
lactating cattle.
Average consumption rates and fat content data for beef and dairy products are
described in EPA (1989). Summary information presented in that reference comes
principally from a U.S. Department of Agriculture Nationwide Food Consumption Survey
(NFCS) conducted in 1977-1978 (described in EPA, 1984b), with additional information
added by Fries (1986). The NFCS covered intake of 3,735 possible food items by 30,770
individuals characterized by age, sex, geographic location, and season of the year. The
average beef fat consumption rates listed in EPA (1989) ranges from 14.9 to 26.0 g per
70-kg person/day, with a single high consumption estimate of 30.6 g per 70-kg
person/day. Based on this information, EPA (1989) recommends using an average beef fat
consumption rate of 22 g/day (this assumes 100 g/day whole beef and 22% fat content).
This is an average rate, which may underestimate the amounts eaten by households who
home slaughter. Milk fat consumption from all dairy products ranges from 18.8 to 43 g
per 70-kg person/day. Considering fresh milk only, milk fat consumption is reported to
average 8.9 to 10.7 g per 70-kg person/day, with a single high consumption estimate of
35 g per 70-kg person/day. An average milk fat consumption rate of 10.5 g/day is given
in EPA (1989) (this assumes 300 g/day whole milk and 3.5% fat). This may also
7-9 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
underestimate the consumption rate of farming families who consume milk supplied by
their own cattle. The rate of beef and milk fat consumption assumed for the high end
farming scenarios in Chapter 9 are 22 and 10.5 g/day, respectively.
Consumption rates of beef and milk are expressed in terms of fat ingested per day
because dioxin-like compounds tend to partition strongly toward lipids. It is assumed that
virtually all of such compounds will be found in the fat portion of milk or beef. Further,
the algorithms to estimate concentrations in these food products estimated fat
concentrations and not whole product concentrations (see Section 5.3.4.3).
EPA (1989) also reports on another survey of 900 rural farm households (USDA,
1966), including some where the farm's beef and dairy cattle supply a portion of the
household's beef and milk. In these situations, the average percent of homegrown beef
and milk (dairy products) is 44% and 40%, respectively. Contact fractions of 0.44 and
0.40 were used in this assessment for the high end farming scenarios. Lacking better
information, EPA (1989) recommends a contact fraction for beef and dairy of 75% if the
intent is to estimate high end estimates for a farmer who uses a portion of his farm's
products.
7.3.6. Fish Ingestion
EPA (1989) concludes that consumption rate data from two studies, that of Puffer
(1981) and Pierce, et al. (1981) are most appropriate for estimating consumption rates for
recreational fishing from large water bodies. The recommended 50th percentile
consumption rate, or typical rate, for this subpopulation is 30 g/day, and the 90th
percentile rate is 140 g/day. For smaller water bodies, EPA (1989) recommends that site-
specific information be obtained via surveys of local fisherman to obtain the most
appropriate fish consumption information for site-specific assessments. Alternately, EPA
(1989) recommends using judgement regarding how many fish meals per year an individual
could obtain from the contaminated waters and assuming meal sizes of 100 to 200 g.
Consumption of commercial fish (at restaurants or from markets) raises market dilution
issues analogous to those described earlier for beef and milk. For this reason, exposed
individuals in both the central and high end scenarios in Chapter 9 are assumed to obtain
their contaminated fish intake from a nearby contaminated stream or pond; other fish they
may consume is assumed to not be contaminated.
7-10 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
The examples used in this assessment assume that the contaminated waters are
small lakes or streams which are occasionally fished on a recreational basis. Further it is
assumed that an individual could eat 3 to 10 meals per year from the contaminated
waters. Assuming an average meal size of 150 g, this translates to 450 to 1 500 g/year or
an average of 1.2 to 4.1 g/day. The central estimate for the example scenarios in Chapter
9 will therefore be 1.2 g/day, and the high end will be 4.1 g/day. Since these fish
ingestion rates are rates of ingestion of contaminated fish, the contact fraction would
be 1.
7.3.7. Fruits and Vegetables
EPA (1989) estimated ingestion rates for individuals who have home gardens and
hence grow a portion of their fruit and vegetable intake. Their approach was to review the
literature and derive average intake rates for all individuals, whether or not they have a
home garden, and considering a variety of different fruits and vegetables. A typical and
high end exposed individual had the same total ingestion rates. Their exposure was
distinguished by the contact fractions; high end exposed individuals grew a larger
proportion of their intake in their home gardens.
The average amounts of fruit and vegetable consumption are 200 and 140 g/day,
respectively. These total ingestion rates are further refined considering two factors
pertinent to estimation of concentration of dioxin-like compounds: whether the vegetation
is grown below (carrots, e.g.) or above ground (tomatoes), and whether the edible portion
is protected (citrus) or unprotected (apples). Chapter 5 discusses distinct procedures for
estimating vegetative concentrations for below and above ground vegetation. Also, both
algorithms assume that inner portions of vegetation are largely unimpacted, whereas outer
portions of both above and below ground vegetation are impacted (see Chapter 5 for
further detail on these algorithms and assumptions). Therefore, for fruits or vegetables
which are protected, it can be assumed that there will be no exposure since the outer
portions are not eaten. Results from a food consumption survey, such as that from Pao,
et al. (1982), can be used to determine percent of total fruit/vegetable intake which is
below/above ground and which is protected/unprotected. Such an exercise was
undertaken using data from Pao, et al. (1982) summarized in EPA (1989) to arrive at the
following percentages:
7-11 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
I. Fruits II. Vegetables
Above Below Above Below
Protected 56% 0 25% 1%
Unprotected 44% 0 54% 20%
As seen, it was found that there are no fruit grown underground, and there was a fairly
similar proportion of protected and unprotected fruit. Fruits considered protected for this
exercise included oranges, grapefruits, and cantaloupe; unprotected fruits included apples,
peaches, pears, and strawberries. It is noted that this is clearly not a complete inventory,
but only those fruits from the survey of Pao as summarized in EPA (1989). Similarly,
these percentages are not being recommended as general values for other site-specific
assessments. For this assessment, it will be assumed that a total ingestion rate of
unprotected above ground fruit is 88 g/day (0.44*200 g/day), and that there is no
ingestion of unprotected under ground fruit. Vegetables above ground and unprotected
include: cabbage, cucumbers (including cucumbers as pickles), lettuce, tomatoes, broccoli,
spinach, string beans, and squash. Above ground protected vegetables include: corn, lima
beans, and peas (several kinds). Below ground unprotected vegetables included potatoes
and carrots; mature onions were considered below ground and protected. Assumed for
this assessment are ingestion rates of unprotected above ground vegetables of 76 g/day
(0.54*140 g/day) and unprotected below ground vegetables of 28 g/day (0.20*140
g/day).
These ingestion rates are defined as total ingestion rates of unprotected
above/below ground fruits/vegetables. Only a portion of these are homegrown. Data
summarized in EPA (1989) shows that the fraction of vegetables consumed that are
homegrown ranges from 0.04 to 0.75, depending on type. The overall average of the data
is 0.25, which is recommended as a contact fraction for the average home gardener. The
recommendation for the high end exposure was 0.40. These contact fractions were
adopted as the central and high fractions for the example scenarios in Chapter 9. Similar
data for fruits show a homegrown range of 0.09 to 0.33, with an average of 0.20, which
is the central estimate used in Chapter 9. EPA (1989) recommends a high end value of
0.30, which is the value used for the high end exposure scenarios.
7-12 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 7-1. Summary of exposure pathway parameters
Pathway
description
Contact
rates
Contact
fractions
Comments
Soil Ingestion
Central
High End
200 mg/d
800 mg/d
1.0
1.0
Only pathway specific to an
age-range; 2- to 6-year-old
children
Soil Dermal Contact
Central
High End
0.2 mg/cm2-event
5000 cm2
40 events/yr
1.0 mg/cm2-event
1000cm2
350 events/yr
Dermal absorption fraction: 0.03
Vapor/Dust Inhalation
Central
High End
20 m3/day
20 m3/day
Water Ingestion
Central
High End
1.4 L/day
1.4 L/day
1.0
1.0
0.75
0.90
0.75
0.90
Unlike other pathways, daily
contact is not assumed;
approach instead estimates
contact in terms of contact/
event * events/yr; both soil
pathways distinguish typical and
high end contact by contact
rate rather than contact
fraction; high end based on
behavior of farmers; typical
pattern based on non-farming
adults; a "dermal absorption
fraction" reduces the amount
contacting the body to an
amount absorbed by the body.
Indoor/outdoor air quality
assumed equal; contact fraction
can be used to reflect a
different assumption; central
contact fraction is an average
time-at-home based on time use
surveys
2.0 L/day evaluated as 90th
percentile value in EPA (1989),
who instead recommend 1.4
L/day as an average
consumption rate; like inhalation
pathways, 0.75 contact fraction
is an average time-at-home
estimate
(continued on the following page)
7-13
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 7-1. (continued)
Pathway
description
Contact
rates
Contact
fractions
Comments
Beef Fat Ingestion
Central
High End
Milk Fat Ingestion
Central
High End
NA
22 g/day
NA
10.5 g/day
NA Central scenario includes adults
0.44 who do not home produce their
beef supply; 22 g/day beef fat
assumes 100 g/day average
beef consumption and 22% fat
NA Like beef ingestion, central
0.40 scenarios do not include home
milk production; 10.5 g/day milk
fat assumes 300 g/day whole
milk and 3.5% fat
Fish Ingestion
Central
High End
1.2 g/day
4.1 g/day
Fruit Ingestion
Central
Above ground unprotected 88 g/day
Below ground unprotected 0 g/day
High End
Above ground unprotected 88 g/day
Below ground unprotected 0 g/day
1.00 Unlike other exposure
1.00 pathways, contact rate is rate
of contaminated fish - hence
contact fraction is 1.00; 1.2
g/day based on 3 recreationally
caught fish meals/yr; 4.1 g/day
based on 10 recreationally
caught fish meals/yr
0.20 200 g/day is average total fruit
consumption rate; 44%
assumed above ground and
0.30 unprotected - no below ground
unprotected fruit; contact
fraction of 0.20 based on a
range of rates of 0.09 to 0.33
for wide variety of fruits (EPA,
1989)
(continued on the following page)
7-14
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 7-1. (continued)
Pathway
description
Contact
rates
Contact
fractions
Comments
Vegetable Ingestion
Typical
Above ground unprotected
Below ground unprotected
High End
Above ground unprotected
Below ground unprotected
0.25 140 g/day is average total
76 g/day vegetable consumption rate;
28 g/day 54% assumed above ground
0.40 and unprotected, 20% assumed
76 g/day below ground and unprotected;
28 g/day central contact fraction of 0.25
based on a range of rates of
0.04 to 0.75 rates for wide
variety of vegetables (EPA,
1989)
Exposure Duration: A duration of 9 years was assumed for the "residence" setting based on
mobility data showing the average time in one residence was 9 years (EPA, 1989). A duration of
30 years was evaluated as the 90th percentile of time in one residence, and was selected for the
"farm" settings, which assumes that farming families live in a given residence longer than a non-
farming family. These values apply to all pathways with no variation, except for soil ingestion,
which was demonstrated only for children and was at 5 years.
Body Weight/Lifetime: The standard assumptions of a 70 kg adult and 70 years lifetime were
assumed for all pathways, except that of soil ingestion. In that case, a body weight of 17 kg was
used.
7-15
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
REFERENCES FOR CHAPTER 7
Binder, S.; Sokal, D.; Maughn, D. (1986) The use of tracer elements in estimating the
amount of soil ingested by young children. Arch. Environ. Health 41: 341-345.
Clausing, P.; Brunekreff, B.; Van Wijen, J.H. (1987) A method for estimating soil
ingestion by children. Int. Arch. Occupational Environ. Health 59: 73-82.
Fries, G.F. (1986) Assessment of potential residues in foods derived from animals
exposed to TCDD-contaminated soil. Presented at 6th international symposium on
chlorinated dioxins and related compounds; September; Fukuoka, Japan.
Hawley, J.K. (1985) Assessment of health risk from exposure to contaminated soil. Risk
Analysis 5(4): 289-302.
Pao, E.M., K.H. Fleming, P.M. Guenther, et al. 1982. Foods commonly eaten by
individuals: amount per day and per eating occasion. U.S. Department of of
Agriculture. Home Economics Report No. 44.
Pierce, R.S.; Noviello, D.T.; Rogers, S.H. (1981) Commencement Bay seafood
consumption report. Preliminary report. Tacoma, WA: Tacoma-Pierce County
Health Department.
Puffer, H. (1981) Consumption rates of potentially hazardous marine fish caught in the
metropolitan Los Angeles area. EPA Grant #R807 120010.
U.S. Department of Agriculture. (1966) Household food consumption survey 1965-1966.
Report 12. Food Consumption of households in the U.S., Seasons and years 1965-
1966. United States Department of Agriculture, Washington, D.C. U.S.
Government Printing Office.
U.S. Environmental Protection Agency. (1991 a) Guidelines For Exposure Assessment
SAB Draft Final. OHEA-E-451.
U.S. Environmental Protection Agency. (1992) Dermal Exposure Assessment: Principals
and Applications. Exposure Assessment Group, Office of Health and Environmental
Assessment, Office of Research and Development, U.S. Environmental Protection
Agency. EPA/600/8-91/011B.
U.S. Environmental Protection Agency. (1989) Exposure Factors Handbook. Exposure
Assessment Group, Office of Health and Environmental Assessment, Office of
Research and Development, U.S. Environmental Protection Agency. EPA/600/8-
89/043. July, 1989.
7-16 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
U.S. Environmental Protection Agency. (1985) Dioxin Transport From Contaminated Sites
to Exposure Locations: A Methodology for Calculating Conversion Factors.
Exposure Assessment Group, Office of Health and Environmental Assessment,
Office of Research and Development, U.S. Environmental Protection Agency. EPA-
600/8-85-012. NTIS PB85-214310.
U.S. Environmental Protection Agency. (1984a) Risk Analysis of TCDD Contaminated
Soil. Exposure Assessment Group, Office of Health and Environmental
Assessment, Office of Research and Development, U.S. Environmental Protection
Agency. EPA-600/8-84-031.
U.S. Environmental Protection Agency. (1984b) Stochastic processes applied to risk
analysis of TCDD contaminated: a case study. Internal report dated May 31, 1984;
Exposure Assessment Group, Office of Health and Environmental Assessment,
Office of Research and Development, U.S. Environmental Protection Agency.
7-17 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
8. PHARMACOKINETICS
8.1. INTRODUCTION
The pharmacokinetic profiles of CDDs and the CDFs are quite complex. A thorough
analysis and understanding of these pharmacokinetic data would be very helpful in
ensuring that exposure assessments for these compounds are reliable. In addition, such
information would be useful in providing enhanced knowledge and understanding for the
purposes of risk assessment.
Exposure to 2,3,7,8-TCDD and related compounds results in numerous species and
tissue specific toxic and biological responses. Many, if not, all of these responses are
mediated by a soluble intracellular protein, the aryl hydrocarbon (Ah) receptor, to which
2,3,7,8-TCDD binds with high affinity. After 2,3,7,8-TCDD and related compounds bind
to this Ah receptor the complex undergoes a transformation process involving dissociation
of hsp90. The transformed receptor complex is then able to bind with high affinity to a
specific DNA sequence referred to as a dioxin responsive enhancer (DRE). The conserved
nature of the DRE and Ah receptor is also indicated by the ability of transformed 2,3,7,8-
TCDD: Ah receptor complexes from a wide variety of species to bind to the DRE. Studies
also indicate a similarity in DNA recognition by Ah receptor from a variety of species
suggestive of a functional role of this sequence in 2,3,7,8-TCDD responsiveness (Denison
et al., 1991; Gasiewicz and Henry, 1991; Perdew and Hollenbeck, 1990; Andersen and
Greenlee, 1991). Thus the definition of "disposition" may have to be extended to include
suborgan or subcellular sites in order to more fully describe the congener, species, and
train specific pharmacokinetics (dosimetry) of these compounds.
Pharmacokinetic analysis may be used in several ways to aid in the exposure and
dose assessment of foreign chemicals. They may, for example, allow for predicting the
time and profile of elimination of chemicals from the body. The redistribution of CDDs
among the various tissues and organs, which may occur during elimination, can be
accounted for and tracked. Effects on disposition which may result from altered
physiology, such as sudden weight loss or from lactation, can be incorporated and thus
adequately considered in exposure and risk assessments. Lactation is known to be an
efficient route for the transfer of many of these chemicals from mother to offspring (Nau
et al., 1986; Bowman et al., 1989).
8-1 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Pharmacokinetic analyses can be used to estimate background exposure levels from
body burden data. They can also be used to estimate uptake rates from various food
sources, elimination rates and times from the body, and to estimate tissues levels from
blood and adipose tissue monitoring. In addition, with the appropriate data on several
congeners, estimates can be made for other congeners about which less data are available.
The remainder of this chapter will cover areas of pharmacokinetics pertinent to
exposure assessment. Background levels and daily uptake of 2,3,7,8-TCDD will be
reviewed and discussed; a method for the calculation of uptake of other congeners from
food will be outlined; use of a compartmental model to estimate daily uptake will be
demonstrated; a method will be outlined and reviewed for determining internal tissue
concentrations from monitored blood and/or adipose tissue; pertinent data on
bioavailability will be summarized.
8.2 DAILY BACKGROUND LEVELS
8.2.1 Basis for Calculation
Physiologically based pharmacokinetic (PBPK) models are convenient and useful
methods for describing and predicting disposition of foreign chemicals in the body. These
models take into account physiologic and biochemical processes such as blood flows,
metabolism, and renal clearance, and describe the body according to its normal anatomy.
PBPK models can, given adequate data, predict disposition from one exposure scenario to
another and even from species to species. One such model was developed for 2,3,7,8-
TCDF (King et al., 1983) and is used here with some modifications. The anatomic regions
depicted in the King model are the blood, liver, fat, skin, and muscle. The remaining
organs of the body are lumped together as the "carcass." Input may be by a variety of
routes, but for the purposes of this discussion is considered to occur through the
gastrointestinal system by continuous chronic dosing. This is consistent with the common
belief that most of these compounds enter the body through the gastrointestinal track as a
result of the consumption of products containing animal fat (Schlatter, 1991). The
pertinent equations follow.
For the liver:
8-2 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
= Qr(CB - - KL + D (8-1)
*IV
dt RL RL
Where:
VL = Volume of the liver
CB = Concentration of toxin in blood
CL = Concentration of toxin in liver
t = Time
QL = Blood flow to liver
RL = Equilibrium concentration ratio between liver and blood
KL = Clearance term (L/tirne)
D = Input or dosing function
The clearance in the liver is considered to be by metabolic processes.
For the fat:
(8-2)
Where: All terms are analogous as those in equation (8-1).
Note that there is no metabolic elimination assumed. The only disappearance of material
from the fat is assumed to be diffusion driven and is accounted for in the above mass
balance equation. Equations for the skin, carcass, and muscle are analogous to that for
fat.
The equation for the blood:
(8-3)
at RL
Where:
Q = Blood Flows
C = Concentrations
R = Equilibrium Concentration Ratios between tissue and blood
Subscripts:
B, L, F, M, S, C refer to blood, liver, fat, muscle, skin and carcass.
Some assumptions may be made to simplify the model for use to estimate daily
background doses. If steady state is assumed then the equations for the individual organs
can be summed. The resultant equation for the liver is then:
8-3 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
(KL)CL
D = ' L (8-4)
t<7
and at steady state:
^ = CB (8-5)
and
CB = ^ (8-6)
and hence
(KL) CF oo
D = "^ (8-7)
When clearance is expressed in days, D is the daily intake. Clearance can be approximated
from the half life information according to the following equations:
KL = kevd (8-8)
Where:
ke = first order elimination rate constant
Vd = Volume of distribution
and
ke = 4^- (8~9)
cl/2
Where:
t1/2 = biological half life of compound in body
8-4 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
The volume of distribution may be estimated as in King et al., (1983) according to:
Vd,i = viRi <
Where:
Vd j = Volume of distribution of organ, i
Vj = Actual volume of organ, i
RI = Equilibrium concentration ratio between organ i and blood
With substitution equation (8-7) becomes:
D = (.)(Vi)(Ci ) (8-11)
Cl/2
Where:
Vj = Volume of the organ in which toxin is measured
Cj ss = Steady State concentration of toxin in organ
It is important to note that two major assumptions are in effect when the above
formula is used to calculate average daily uptake. First steady state conditions are
assumed. Given that the half life of some of these compounds (e.g. 2,3,7,8-TCDD) are at
least 5 years it would take well over 15 years to reach 90% of steady state, and over 30
years to reach 99% of steady state levels. Thus, the assumption of steady state is only
reasonable if background environmental concentrations are relatively similar and constant
throughout the nation. Under such conditions even the normal movement from one
geographical location to another would result in relatively constant exposures. Also
implicit in this assumption is that bioavailability is relatively constant through the nation.
Given that the source of this background exposure is believed to be animal fat (Schlatter et
al., 1991) the assumption of steady state for adults might be considered a reasonable
assumption. Exceptions are those individuals who may for a portion of their adult lives be
consuming foods with unusually high levels of these compounds. It would be expected
however that those individuals would have higher than average body burdens, and hence
would not be considered to have only average background exposure, but would rather be
considered part of a source specific exposure group. Conditions such as sudden weight
loss and lactation would also alter the steady state condition. However, for purposes of
8-5 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
calculating daily background exposure levels the sampling of tissues for body burden must
be so designed to account for such deviations in the average. In summary, for adults
(over 25 years of age) not in a source specific exposure group, the steady state
assumption is a reasonable approximation. It should be remembered however, that the
longer the biological half life the longer it would take to reach steady state. A compound
with a half life of 10 years, for example, would take over 50 years to reach the 90% of
steady state value.
The second major assumption is that these compounds are eliminated from the
body by monophasic kinetics. Biphasic elimination is very possible for many of these
compounds. Data gathered to calculate elimination rates or half lives would only reveal
biphasic elimination profiles if gathered several years after the last exposure. Using only
the short term half life would result in an underestimated value for half life and an
overestimate of daily intake. This is particularly problematic for those compounds with
extremely long half lives and for which few data exist. In a later section an approach to
calculating half lives for some of these compounds will be presented. Also, the elimination
kinetics are assumed to be constant over the entire life of the individual. Sudden weight
loss and lactation would, for example, be conditions which violate that assumption.
Again, it would be assumed that for calculation of daily intake due to background exposure
the body burden data from such individuals would be identified and calculations handled
accordingly.
8.2.2 Daily Intakes
Figure 8-1 shows a sample calculation for 2,3,7,8-TCDD using the above
procedure. A fat volume of 14 L was chosen, representing 20% of the body weight.
Also, for the purposes of this example, 1 ml of tissue was assumed to be equivalent to
1 gm. Table 8-1 shows the estimated daily intake of 2,3,7,8-TCDD at several conditions.
The range of daily intakes calculated are in agreement with those reported elsewhere
(FCirst et al., 1991; US EPA, 1992).
In order to perform similar calculations for other congeners three pieces of
information are necessary. First, concentrations in the adipose tissues must be known.
Second, the half lives of the compounds within the body must be known. Third, some
understanding of the kinetics and exposure conditions to assure that steady state
8-6 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
D =
In2 \
14L 1000
6.72
5.8years/ \~ ~ L/\ ml) \70KGJ \365daysj
D = 0.44 pg I kg / day
Figure 8-1: Sample Calculation of Daily Intake for 2,3,7,8-
TCDD
Table 8-1 Calculated Daily Intakes for 2,3,7,8-TCDD.
Half life
(yrs)
5.8
7.0
5.8
7.0
5.8
7.0
5.8
7.0
Fat Vol. (L)
14.0
14.0
14.0
14.0
7.0
7.0
7.0
7.0
Fat Cone.
(PPt)
6.72
6.72
5.00
5.00
6.72
6.72
5.00
5.00
Calculated
Daily
Intake
(PG/KG/Day)
0.44
0.37
0.33
0.27
0.22
0.18
0.16
0.14
conditions were achieved at the time of monitoring.
Concentrations of various congeners in adipose tissues can be found in several
sources (Stanley et al., 1986; Scheter, 1991). Values range from around 2 ppt for
2,3,7,8-TCDF to several hundred ppt for 1,2,3,4,6,7,8,9-OCDD.
Half lives could be determined from elimination data, if available. Methods have
8-7
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
been suggested to determine the half life of such compounds from uptake data relative to
2,3,7,8-TCDD. Schlatter (1991) has proposed one such method. The following has been
adapted from that proposed method.
Manipulation of equation (8-11) results in:
C
TCDD
DTCDD tl/2,TCDD V (8-12)
Where:
CTCDD = Concentration of TCDD in body
DTCDD = Dai|V '"take of TCDD
t1/2TCDD= half life of TCDD in body
V = Volume of body compartment
For some other congener x:
= Dx t1/2>> V
x In2
Where:
Symbols are same as for equation (8-12) and subscript x applies to compound x.
Thus the ratio of concentrations of TCDD to x can be described by:
TCDD _ \DTCDD t\!2,TCDD V] f In2
~
C* ln2 V
(8-14)
Which with algebraic manipulation and simplification becomes:
_ [DTCDD tl/2,TCDD] (Cx]
=
i/9 r
i/L,x
- —
< I n
TCDD \Dx
Assuming intake D, to be mostly from the food, especially animal fat products, they can
be related to absorption from these foods according to:
DTCDD = (ka,TCDD) (ATCDD) (8-16)
8-8 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Where:
ka TCDD = Absorption rate constant for TCDD
ATCDD = Concentration of TCDD in animal fat (diet)
and
Dx = (kayX) (Ax) (8-17)
Where:
ka x = Absorption rate constant for x
A^ = Concentration of x in animal fat (diet)
As a result the half life for compound x can be described by:
f 3 "\ f /^ 1 r IT "^
. _ . ATCDD\ \^x\ \Ka,TCDD \ fa-ia\
tl/2,x ~ tl/2,TCDD -f, K- —£ (B XB'
[ CTCDDJ [Ax\ [ ka,x \
When the absorption rate constants for each are equal or when the difference between
them is small compared to differences in other parameters (concentration, half lives)
equation (8-18) can be further simplified to:
IAT/rtn 1 fCy.1
1 <~.L)L) \ I X I / o _ 1 Q \
~n ~a~ ( '
CTCDD\ [Ax\
Before using the above approach to calculate half lives for some of the other
substances of interest it is well to briefly highlight one of the assumptions in this
approach. The relationship of half life to elimination as described in equation (8-9) only
applies to simple single compartment kinetics. These compounds would not necessarily be
expected to behave in such a manner. However, the error introduced by such an
assumption is not great if the one phase predominates over the other, or if it is
remembered that the calculation applies to one phase only. In fact, as will be discussed in
a subsequent section, it is believed that for these chemicals the relationship between half
life and elimination as described here is a reasonable approximation.
It should be noted that for some of these substances exposure is expected from
other than food sources. For such cases equation (8-18) would be modified to include
8-9 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
these other sources as follows.
Where:
-
l/2,x
«,,-,*
ka,i,TCDD = Absorption rate constants for TCDD from each of the i media
Ai,TCDD = Concentration of TCDD in each of the i media
ka j x = Absorption rate constants for x from each of the i media
Aj x = Concentration of x in each of the i media
Other symbols: As previously defined
Again, if the differences between the absorption rate constants for TCDD and x are judged
to be small then the following variation of equation (8-19) can be used:
4- - - fc , , Ai,TCDD
l/2,TCDD \2^ ( (~r -
CTCDD
Table 8-2 shows the results of some half lives calculated in this manner.
The half lives calculated using equation (8-19) for the first three compounds in
Table 8-2 agree with those calculated by Schlatter (1991). The large difference in the two
calculations for OCDD is due to significant differences in absorption rates between the
TCDD and the OCDD. Schlatter notes in his paper that for some compounds, including the
OCDD, corrections were made of differences in absorption. No explanation was offered
on how this was done. In US EPA (1992) results are summarized that indicate a possible
several fold greater oral absorption of TCDD over OCDD. This adjustment would result in
a calculated half life closer to that calculated by Schlatter.
In summary, this illustrates a method for calculating the half lives of similar
behaving compounds. Several pieces of information are necessary: 1) the concentration
in the body of 2,3,7,8-TCDD; 2) the half life of 2,3,7,8-TCDD; 3) the concentration of the
substance of interest in the body; 4) the concentration of the substance of interest in the
media from which exposure occurs; 5) the differential absorption rates for TCDD and the
8-10 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 8-2 Half life calculations.
Chem.
2378-TCDD
2378-TCDF
12378-PeCDD
23478-PeCDF
OCDD
Food-ppta
0.23
0.84
0.7
1.4
19.2
Body-pptb
6.72
3.9
21.5
36.8
653.0
+• c
rl/2
years
7.0
1.11
7.35
7.94
8.14
4- d
rl/2
years
6.0
1.3
5.0
6.3
50.0
Concentrations in food for TCDD, TCDF, and OCDD were
obtained as discussed in Chapter 3; those for PCDD, PCDF
were taken from Schlatter, 1991
bConcentrations in body (adipose tissue) were all taken
from Schecter, 1991
cCalculated using equation 8-19, except for TCDD
Calculated by Schlatter (1991), except for TCDD
substance of interest.
From the half life information, average daily intakes may be calculated when steady
state can be assumed, by using equation (8-12). Caution should be exercised when
calculating daily intakes for those compounds with very long half lives such as OCDD in
the above example.
8.3 COMPARTMENTAL MODELING
As previously discussed and also discussed elsewhere (US EPA, 1992) PBPK
models are very useful for describing and predicting the disposition of chemicals in the
body. They are generally designed for predicting some measure of dose at a target site.
They are also used to extrapolate from one species to another, between different doses,
and between different routes of exposure. Several PBPK models have been published
specifically for 2,3,7,8-TCDD (Leung et al., 1990; Leung et.al., 1988; Kissel and Robarge,
1988). Also, as previously discussed, a model for 2,3,7,8-TCDF (King et al., 1983) is also
8-11
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
available and from which were derived the equations for estimating daily intake from body
concentrations at steady state conditions. Most of these models have been developed to
describe in great detail the metabolism and binding of TCDD within the body, for the
purpose of estimating target tissue dose.
8.3.1 Pharmacokinetic Model
Assuming a linear relationship between the concentrations in the fat and the body
at low exposure concentrations and the near linear elimination profile, a simpler model
relating exposure, whole body elimination, and whole body concentration is developed
here. As described earlier, after a prolonged exposure most of the body's organs can be
assumed to have very similar kinetic profiles and can thus be lumped together. The fat,
while sharing such a similar profile, is kept separate because of its important role in storing
most of the body burden and because it has been the most typically monitored tissue. The
three compartments in this model are blood, fat, and a "body" representing all other
tissues. The model is a flow or perfusion limited model with the assumption that the toxin
is well stirred or uniformly distributed within each compartment. The pertinent equations
for the model follow:
dt V
bo
Where:
dCbo/dt = Rate of change of concentration of toxin in body (bo)
Qbo = Blood flow (Volume/time) to body
CB = Concentration of toxin in blood (B)
Rbo = Equilibrium concentration ratio of toxin between body and blood
K = Clearance (Vol/time) of toxin from body
Vbo = Volume of body
D = Intake of toxin
The integral of the differential equation over time is the actual concentration.
8-12 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
dCF "* r % (8-23)
dt V
F
Where:
dCF/dt = Rate of change of concentration of toxin in fat (F)
OP = Blood flow (Volume/time) to fat
RF = Equilibrium concentration ratio of toxin between fat and blood
VF = Volume of fat
~dt
It should be noted that the description of intake is somewhat different than what is
typically found in most PBPK models. D here is actually a dose rate. That is, D is in terms
of pg/kg/day coming into the body as a dose, not just a concentration in the food, drinking
water, or air. The usual description for gastrointestinal absorption of toxin from food
would be
D = kaFc (8-25)
Where:
ka = absorption rate constant
Fc = Concentration of toxin in food
In this case because of inadequate knowledge regarding the absorption rate constants for
many of the congeners a dose rate is used instead. This does not allow for estimation of
body burden directly from environmental concentrations (e.g. Fc in equation (8-25)). As
more data are collected, more accurate values of the parameters and descriptions of
absorption functions can be input into equation (8-22). For the present other approaches
can be used to relate concentration in the environmental media to daily intake (see Section
8.4.3).
8-13 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
8.3.2 Model Utilization
Most of the model's parameters are known or can be estimated from experimental
data for many of these toxins. For example, the equilibrium concentration ratios between
the fat and the body and between the fat and the blood for 2,3,7,8-TCDD are
approximately 10 and 100 respectively on a tissue basis. The blood flows and
compartment volumes are well known. One parameter that had to be estimated when
applying this model to 2,3,7,8-TCDD was the clearance term K. This was done by first
allowing the model to simulate elimination as though exposure was suddenly terminated (D
becomes 0). Initial concentrations for the tissues were taken as those typically expected
in the general population (7.0 ppt in the fat). The value of K was then adjusted until the
model predicted a half life of 7 years. Values of clearance for the other compounds in this
series would be determined similarly. The necessary information includes the equilibrium
concentration distribution ratios, the half lives of the compounds, and some reasonable
approximation of steady state fat concentrations.
With an estimated value for the clearance, the model can now be used with various
exposure inputs to establish body burdens of toxin. The model can also be used to
estimate elimination profiles from the body. In addition, events such as lactation can be
incorporated into the model with knowledge about the appropriate parameters.
Figure 8-2 shows the results of the model run describing the elimination of 2,3,7,8-
TCDD from fat. The clearance rate was adjusted to predict a half life of 7 years. The
same clearance value was used with different starting conditions (concentration of TCDD
in tissues) and the model produced a half life of 7 years. Next the model was used with a
constant daily intake as an input. Figure 8-3 shows the resulting profile of TCDD in the
fat. Figure 8-3 shows the results of a model run using an input of 0.44pg/kg/day. Note
that the steady state fat concentrations are approximately 7.0 ppt. The clearance rate
used in this model run had been independently determined in the previous run based on
reported half life values. Thus, for these conditions this model does an adequate job of
predicting tissue levels of TCDD. Figure 8-4 shows a similar profile for a 0.30 pg/kg/day
8-14 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
10'
Q.
d
C
0
u
ra
10°--
10
-i
0.0
TCDD ELIMINflTION FROM FflT
5.0
10.0
15.0
time-years
20.0
25.0
Figure 8-2: Model Estimates of Elimination of 2,3,7,8-TCDD
from Fat.
8-15
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
FflT CON. UITH DfllLY: 0.44PG/KG/DflY
10
20 30 40
time-years
50
60
70
Figure 8-3: Accumulation of TCDD in Fat with 0.44 pg/kg/day
dose - Human.
8-16
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
FflT CON. UITH DRILY: 0.30 PG/KG/DflY
5.0-T
30 40
time-years
Figure 8-4: Accumulation of TCDD in Fat with 0.30 pg/kg/day
dose - Human.
dose. The daily intakes chosen were similar to those calculated by the steady state
equations in section 8.1.2 (Table 8-1). Note that the steady state fat levels predicted by
the pharmacokinetic model agree closely with those used as a starting point for the steady
state calculation using seven years for the half life. This three compartment model
appears to provide reasonable approximations of body burden, at least under the
circumstances that it has currently been tested. The necessary information for use with
other substances in this series include the equilibrium distribution ratios between fat and
8-17
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
blood and between fat and the rest of the body. An estimate of the half life is also needed
in order to establish an appropriate value for the clearance term used in the model. Table
8-3 shows the results when the model was adjusted for other substances. The TEF (Toxic
Equivalency Factors) based intakes, along with those for 2,3,7,8-TCDD and others, can
then added together as so desired to arrive at a TEF based total daily intake for all the
TCDDs, TCDFs, and PCBs of interest to a particular assessment.
Table 8-3: Model Determined Daily Intakes
Compound
OCDD
1,2,3,7,8
-PeCDF
2,3,4,7,8
-PeCDF
Half life
(yrs)a
50C
8
8
Fat Con.
(PPt)b
1174.0
2.8
13.0
Intake
(pg/kg/day)
23.48
0.55
0.93
Intake-TEF
(pg/kg/day)
0.02
0.03
0.47
aTaken from Table 8-2
bTake from Schecter (1991)
cValue determined by Schlatter (1991)
8.3.3 Determining Liver Concentrations from Fat Levels
As mentioned previously, higher resolution PBPK models are necessary to estimate
and predict concentration at cellular and sub-cellular targets (vidae supra). At the present
time many of the data necessary to develop and apply these models to estimate target
tissue or cellular dose in humans are lacking. Andersen and Greenlee (1991) provide an
approach to use the equations of a PBPK model (Leung et al., 1990) to predict liver
concentrations from monitored fat concentrations. It should also be noted that blood or
milk concentrations expressed on a per lipid basis could also be used. Basically the
8-18
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
approach calculates the ratio of fat to liver concentration by dividing the tissue
concentration equations of the PBPK model.
VL PF
BMi
BM2(T)
CVL
KB
CVL
CVF
(8-26)
Where:
CL,CF
PLV,LPF VF
BM1( KB.,
BM2(T)
KB0
Concentrations of toxin in liver and fat
Concentrations of toxin in venous blood from liver and fat
Livenblood and Fat:blood partition coefficients of toxin
Volume of Liver compartment
Forward and reverse binding constants for binding of toxin with Ah
receptor
Forward binding constant with microsomal binding protein - time
dependent upon the amount of inducible microsomal binding protein
Reverse binding constant for microsomal binding protein
Source: Andersen and Greenlee, 1991.
Andersen and Greenlee (1991) go on to further simplify the previous equation for
various limiting conditions. For the case of very low doses where CVL<
-------
DRAFT-DO NOT QUOTE OR CITE
BM2(max) = Maximally induced binding level of P450
For the case where CVL> >KB17 KB2 equation (8-26) becomes
ft = £t (8-29)
Andersen and Greenlee provide values for each of the above conditions for experimental
animals (1991). It can be readily observed from examining equation (8-26) that when
binding levels are very small the ratio of concentration between liver and fat is influenced
mostly by the ratio of partitioning coefficients.
With this approach and by knowing the necessary parameters equation (8-26) can
be used to estimate liver concentrations from fat (including blood lipid and milk lipid)
concentrations. Andersen and Greenlee (1991) further suggest that most of the necessary
binding parameters can be determined from in-vitro studies. Further, the pharmacokinetic
model from which equation (8-26) was derived can be used to estimate and predict cellular
concentrations (both free and bound) under various exposure conditions. The equations
used by Leung et al. (1990) or other like equations provide estimates of amount bound to
intracellular receptors sites, and can thus provide a relationship between multiple binding
sites. Denison et al. (1991) suggest that binding to the Ah receptor is only one of several
steps necessary for TCDD to have an intracellular toxic effect. As further knowledge
becomes available about the mechanism and kinetics of each step the model can be
expanded to include these other processes such as DNA enhancing and hormonal
modulation. The pharmacokinetic model will therefore become a pharmacodynamic model
which will more explicitly link exposure to effect. Also, other tissues can be more
specifically described in the model if the mechanism of action data so warrant.
8.4 BIOAVAILABILITY AND TISSUE DISTRIBUTION
In a companion document (US EPA, 1992) there is an extensive discussion on the
disposition and pharmacokinetics of these compounds of interest in animals and humans.
For detailed descriptions of the pharmacokinetic and bioavailability studies and findings the
8-20 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
reader is referred to that document. Following is a brief summary of the pertinent findings
related to bioavailability in humans.
8.4.1 Bioavailability
Following ingestion of a material containing 2,3,7,8-TCDD or other toxic species,
the toxic effect of the material is modified by the degree of absorption, principally in the
small intestine. In several experimental studies, investigators administered 2,3,7,8-TCDD-
containing environmental matrices to experimental animals, and measured parameters
relating to bioavailability. These studies included quantification of 2,3,7,8-TCDD in liver
and other tissues following treatment; comparison of toxicities of contaminated
environmental materials with pure 2,3,7,8-TCDD; and examination of enzyme induction.
The results of these different approaches, their limitations, and needs for further research
are discussed below.
8.4.1.1 Bioavailability Data
Umbreit et al. (1985, 1986a,b) conducted experiments in guinea pigs,
administering 2,3,7,8-TCDD in corn oil, 2,3,7,8-TCDD added to chemically
decontaminated soil, or soil from two industrial sites in Newark, New Jersey (a
manufacturing site and a salvage site) contaminated with CDDs. 2,3,7,8-TCDD was the
principal lower chlorinated isomer (dioxin or furan) present in the soil from the
manufacturing site (for which a chemical analysis was presented). Soil from the
manufacturing site was found to have 1,500 to 2,500 ppb 2,3,7,8-TCDD under soxhlet
extraction; release under ambient temperature manual solvent extraction was much lower,
reported as ">2.5 ppb." The soil from the salvage site was reported as approximately
180 ppb 2,3,7,8-TCDD under soxhlet extraction.
In this study groups of two or four male and two or four female guinea pigs
received single gavage doses of the test materials and were observed until death or
sacrifice at 60 days. 2,3,7,8-TCDD in corn oil or in recontaminated soil (6 g/kg in both)
proved highly toxic, without similar toxicity being observed in animals treated with up to
twice this dose of 2,3,7,8-TCDD in the soil from the manufacturing site. The limited data
on 2,3,7,8-TCDD levels in the liver showed much higher levels following administration of
recontaminated soil versus contaminated soil from the manufacturing site.
8-21 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Umbreit et al. (1986a) thus demonstrated that gavaged 2,3,7,8-TCDD containing
soil from the manufacturing site was substantially less toxic than equivalent doses of
2,3,7,8-TCDD in corn oil. However, quantitative comparison of the effective doses in this
study is difficult. Approaches to a quantitative comparison are outlined below.
(1) Guinea pigs receiving 12 ug/kg 2,3,7,8-TCDD in contaminated
soil experienced no deaths, while five out of eight guinea pigs
receiving 6 ug/kg 2,3,7,8-TCDD in corn oil died, with no
groups tested having lower doses in corn oil. Other authors
have provided data on the toxic effects of 2,3,7,8-TCDD in
corn oil which could aid in the comparison.
McConnell et al. (1984) observed one out of six animals dying at 1
ug/kg and six out of six animals dying at 3 ug/kg. Silkworth et al. (1982)
observed three out of six animals dying at 2.5 ug/kg and no deaths out of
six at 0.5 ug/kg. Comparing these data directly with the Umbreit et al.
results would suggest that the 2,3,7,8-TCDD in the Newark manufacturing
site soil was less effective, by a factor of 10 or greater, in producing toxicity
than 2,3,7,8-TCDD in corn oil.
(2) Umbreit et al. reported a "slightly reduced" weight gain in guinea pigs
receiving 6 ug/kg of 2,3,7,8-TCDD in Newark manufacturing site soil, and a
"greater reduction" at the 12 ug/kg dose. No other signs of toxicity were
noted in these groups. The animals receiving 6 ug/kg 2,3,7,8-TCDD in corn
oil, in contrast, exhibited a marked loss of body weight and showed toxicity
and mortality. Silkworth et al. (1982) also provided data on weights of
guinea pigs receiving 2,3,7,8-TCDD in corn oil. Those receiving 2.5 ug/kg
exhibited a marked reduction in weight gain among three out of six
survivors, while those receiving 0.5 ug/kg showed a weight gain comparable
to vehicle controls. Comparison of this weight data with that of Umbreit et
al. suggests that the 2,3,7,8-TCDD in corn oil was more than 5 times but
less than 25 times as potent as 2,3,7,8-TCDD in the Newark soil. This
comparison assumes that the effect of the Newark manufacturing site soil
on weight gain was due to 2,3,7,8-TCDD as opposed to other compounds in
the soil. Numerous other dioxin and furan compounds and other chemicals
8-22 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
have been identified in this soil (Umbreit et al., 1987a). It has not been
established that 2,3,7,8-TCDD is the sole or prime source of toxicity in the
soil.
(3) Umbreit et al. presented liver concentrations of 2,3,7,8-TCDD after death or
sacrifice at 60 days following gavage. Much lower concentrations of
2,3,7,8-TCDD were found in the livers of animals receiving soil from the
manufacturing site compared with those receiving the dose in corn oil.
There are, however, two factors that limit the conclusions than can be
drawn from this comparison.
First, the corn oil group experienced major toxicity and weight loss, particularly
complete loss of body fat. These changes may have affected the partitioning of 2,3,7,8-
TCDD within the body, leading to a higher concentration in the livers of the animals
experiencing toxicity. Second, the animals gavaged with corn oil died early-half were
dead by 26 days, while all of the guinea pigs treated with soil survived to 60 days (with
the exception of one gavage death). The U.S. EPA (1985c) reported a half-life for 2,3,7,8-
TCDD elimination of 30 Ji 6 or 22 to 43 days from two studies in guinea pigs.
Additionally, the U.S. EPA (1985c) stated that elimination in the guinea pig may follow
zero-order kinetics. Differences in elimination due to differences in periods of survival are
likely to have affected the relative quantities of 2,3,7,8-TCDD found in the livers of the
test groups.
Perhaps a more appropriate comparison can be made with the four animals
receiving 0.32 ug/kg of 2,3,7,8-TCDD in contaminated soil from the Newark salvage site.
These animals experienced no reported toxic signs (weight data not presented) and
survived the full 60-day experiment. Approximately 6% of the gavage dose was found in
the liver of these animals, while only about 0.06% of the gavage dose was found in the
livers of guinea pigs in the 12 ug/kg group receiving the Newark manufacturing site soil.
This would suggest that the 2,3,7,8-TCDD in the manufacturing site soil was 100 times
less bioavailable. However, given the different doses used and the fact that only a single
pooled sample was analyzed for 2,3,7,8-TCDD in each group, caution must be used in
interpreting this comparison.
The 2,3,7,8-TCDD in soil from the salvage site was substantially bioavailable,
based on the single liver tissue analysis. Approximately 6% of the administered dose was
8-23 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
recovered from the livers of these animals at 60 days. This can be compared with data on
hamsters given 2,3,7,8-TCDD in corn oil by McConnell et al. (1984), where approximately
8% of the 2,3,7,8-TCDD could be recovered in the 1 ug/kg dose group among survivors at
30 days.
McConnell et al. (1984) treated Hartley guinea pigs (2.5 weeks old) with single
gavage doses of either 2,3,7,8-TCDD or dioxin contaminated soil from two sites in
Missouri. The 2,3,7,8-TCDD concentrations from the two sites were reported at 700 and
880 ppb respectively; total tetrachlorodibenzofurans (TCDF) concentrations in the soil
were 40 to 80 ppb, and polychlorinated biphenyls (PCB) concentrations were 3 to 4 ppm.
Taking into account the relative toxicities, the authors concluded that toxicity from the
other compounds was likely to be small compared with that from 2,3,7,8-TCDD. Livers
were analyzed for 2,3,7,8-TCDD at death or sacrifice at 30 days following treatment.
Treatment deaths occurred between 5 and 21 days post-gavage.
Guinea pigs that died exhibited severe loss of body fat, markedly reduced thymus
and testicle size, and adrenal hemorrhage. No adverse affects were noted in animals
treated with decontaminated soil. For 2,3,7,8-TCDD in corn oil and for both contaminated
soils, there were clear dose-responses in mortality. The calculated LD50 values for the two
soil types were lower than the LD50 for 2,3,7,8-TCDD in corn oil by a factor of three to
four.
There was a dose-response between the liver concentration of 2,3,7,8-TCDD and
the gavage dose; the details of this relationship are complex. Animals dying during the
experiment had liver concentrations a factor of 1.4 to 3.2 higher than animals in the
same dose groups who survived 30 days. This observation makes quantification of the
dose-response relationships difficult (all or most of the animals in the low-dose groups
survived the experiment, while all of the animals in the high-dose groups died). When the
liver concentrations of 2,3,7,8-TCDD in animals dying early at the middle and high-dose
groups are compared, there appears to be a greater-than-linear increase in liver
concentration with dose for the Times Beach and Minker Stout soil groups, with a 3.3-fold
increase in dose producing a 10-to 13-fold increase in liver concentration.
Liver concentrations of animals in the different dosing groups can best be compared
among groups that experienced similar mortality.
(1) Animals in dose groups in which all animals died within 30 days: 2,3,7,8-
8-24 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
TCDD in corn oil, approximately 20% of the administered dose was in the
liver. For the soil-treated groups, 13% and 11% of the doses, respectively,
were in the liver. Comparison of these data suggest that 2,3,7,8-TCDD was
approximately twice as available through corn oil as through soil.
(2) Animals surviving the 30-day experiment (in groups where at least 4 out of
6 survived): For 2,3,7,8-TCDD in corn oil, 7.5% of the administered dose
was in the liver. For soil-treated animals < 3.6, 1.3, < 4.2, and 2.0% of
the doses, respectively, were in the liver. Comparison here would suggest
that 2,3,7,8-TCDD was approximately four times as available through corn
oil as through soil.
The authors note that the differences in liver concentrations observed in the study
may reflect varying partitioning of the 2,3,7,8-TCDD among internal organs, since dying
animals suffered major loss of body weight and fat content. In addition, surviving animals
would have had greater opportunity to metabolize and excrete 2,3,7,8-TCDD due to a
longer lifetime.
Umbreit et al. (1986a) reported additional chemical analyses of the Times Beach
soil. Soxhlet extraction of the Times Beach soil yielded a similar quantity of 2,3,7,8-TCDD
to the solvent extraction reported by McConnell et al. (1984). This is in contrast to the
Newark manufacturing site soil used in the Umbreit et al. (1987a) experiments, where only
a small fraction of soxhlet-extractable 2,3,7,8-TCDD was extractable by the solvent
extraction methodology used by McConnell et al. (1984).
McConnell et al. (1984) also reported an experiment in which groups of six
Sprague-Dawley rats were given single gavage doses of 2,3,7,8-TCDD in corn oil or
dioxin-contaminated soil from the Minker site. Induction of aryl hydrocarbon hydroxylase
(AHH) in the rat livers was measured at sacrifice 6 days after dosing. Experimental doses
ranged from 0.4 to 5.0 ug/kg 2,3,7,8-TCDD. Measured AHH induction was similar for
groups receiving 2,3,7,8-TCDD in corn oil or receiving contaminated soil containing nearly
equal doses of 2,3,7,8-TCDD. For example (based on the rate of formation of 3-
hydroxybenzo[a]pyrene), AHH activity was measured at 1,269 pmole min"1 mg~1 for the
group receiving 5 ug/kg 2,3,7,8-TCDD in corn oil and at 1,230 pmole min'1 mg"1 for the
group receiving 5.5 ug/kg 2,3,7,8-TCDD in contaminated soil. For the five dose groups,
the AHH activity for the soil group ranged from 50% to 110% of the activity in the corn
8-25 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
oil group.
The McConnell et al. rat data indicate that the bioavailability of 2,3,7,8-TCDD from
the Minker site soil was at least 50% of that of equivalent doses of 2,3,7,8-TCDD in corn
oil.
Lucier et al. (1986) provided additional information on the induction of hepatic
enzymes in rats by the 2,3,7,8-TCDD contaminated soil from the Minker site tested by
McConnell et al. (1984). AHH induction was similar for the groups of rats receiving
2,3,7,8-TCDD in corn oil and contaminated soil (within a factor of two) over a broader
range of doses (0.01 5 ug/kg to 5 ug/kg) than reported by McConnell et al. In a second
enzyme assay using the same animals, UDP glucuronyltransferase activity was found to be
slightly higher in groups receiving 2,3,7,8-TCDD in corn oil than groups receiving equal
doses in contaminated soil.
Liver concentrations of 2,3,7,8-TCDD for the rats were also reported. For the corn
oil vehicle the liver concentrations were 40.8 _+_ 6.5 ppb at the 5 ug/kg dose and 7.6 _+.
2.5 ppb at the 1 ug/kg dose. Assuming that the liver comprises 4.0% of body weight, the
retention rates for the 5 and 1 ug/kg doses were 33% and 30%, respectively. In rats
receiving 2,3,7,8-TCDD in contaminated soil, the 5.5 ug/kg group had liver concentrations
of 20.3 _+. 12.9 ppb, and the 1.1 ug/kg group had concentrations of 1.8 _i0.3. Thus,
retention rates for the 5.5 and 1.1 ug/kg groups are estimated at 14% and 7%,
respectively. These data indicate that liver retention in the soil group was 20% to 40% of
that in the corn oil vehicle groups.
Umbreit et al (1986b) report additional studies of mortality in guinea pigs treated
with soil containing 2,3,7,8-TCDD from Newark (manufacturing site) and Missouri (Times
Beach) previously tested by Umbreit et al (1985, 1986a) and McConnell (1984),
respectively. Guinea pigs received a single gavage dose of a soil suspension and were
observed for 60 days. After autopsy, deaths were classified as whether or not they
appeared to be due to TCDD toxicity. Substantial mortality (25% overall) from conditions
not attributed to TCDD was observed across all groups.
The data for both the Newark and Missouri sites are similar in trend for the previous
data on these sites; and clearly indicate the greater toxicity of the Newark soil for given
equal administered doses of 2,3,7,8-TCDD. With larger groups of guinea pig studied, a
toxicity-related death was observed in both the 5 and 10 mg/kg dose groups for Newark
8-26 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
soil while no deaths were observed in corresponding dose groups (6 and 12 mg/kg) with
fewer animals in Umbreit et al (1986a).
Comparing groups within this study, similar mortality (1 or 2 deaths in 10 to 16
animals) was seen in both the 5 and 10 ug/kg Newark groups and the 1 and 3 ug/kg
Missouri groups. These results suggest that the toxicity of these materials, differs by an
order of magnitude or less. As noted above the degree to which toxicity from these soils
can be attributed to 2,3,7,8-TCDD in the presence of numerous other related toxic
compounds is not known. 2,3,7,8-TCDD tissue concentrations were not reported in this
work.
In another comparative study Umbreit et al. (1987b) compared the Newark
manufacturing site and Times Beach soils in the induction of aryl hydrocarbon hydroxylase
(AHH) in rats. While the use of only single dose levels prevents detailed analysis, the two
soils proved quite similar in their ability to induce AHH. The explanation for the difference
in this finding from those observed in the toxicity studies discussed above is not clear, but
may relate to the presence of other toxic and/or AHH inducing compounds.
Umbreit et al. (1987a) report a reproductive toxicity study with soils from the
Newark manufacturing site and salvage yard previously studied by Umbreit (1986a).
Female mice were treated thrice weekly with soil from these sites, with treatment
continuing through fertilization to weaning of pups. The total doses of 2,3,7,8-TCDD
received by the mice were 720 ug/kg in manufacturing site soil, and 86 ug/kg in salvage
yard soil. A corn oil vehicle group and are contaminated soil group received a total of 225
ug/kg.
Deaths in animals showing "classic signs" of TCDD toxicity were observed in the
corn oil and recontaminated soil groups, and indicate appreciable bioavailability of 2,3,7,8-
TCDD. Deaths were also observed in animals receiving manufacturing site soil but the
authors did not observe "classic signs" of TCDD toxicity. Fewer live pups born and fewer
pups surviving until weaning were observed in the manufacturing site soil group compared
with those receiving decontaminated soil. TCDD completely blocked reproduction in the
corn oil and recontaminated soil groups. The results of this study demonstrate acute and
reproductive effects occurred in animals receiving manufacturing site soil. However, these
effects were of a lesser magnitude than those seen in animals treated with 2,3,7,8-TCDD
in corn oil at a dose three fold lower. The authors note the presence of substantial
8-27 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
quantities of other toxic substances in the manufacturing site soil (chemical analyses
presented). No toxic effects were noted in animals treated with salvage site soil, who
received a much smaller 2,3,7,8-TCDD dose. The data does not allow a quantitative
evaluation of the bioavailability of 2,3,7,8-TCDD.
Kaminski et al. (1985) and Silkworth et al. (1982) reported the results of a series of
studies on the toxicity of soot containing dioxin and furan compounds from a fire involving
transformer fluid containing PCBs. Hartley guinea pigs (500 to 600 g) received single oral
doses of soot in an aqueous vehicle, a soxhlet extract of the soot in the same vehicle, or
2,3,7,8-TCDD in either an aqueous vehicle or corn oil.
The soot was reported to contain 2.8 to 2.9 ppm 2,3,7,8-TCDD and 124 to 273
ppm 2,3,7,8-TCDF. The total polychlorinated dibenzofuran content was estimated at
5,000 ppm. Animal weights and mortality were recorded for 42 days, at which point the
survivors were sacrificed and LD50 values were calculated. Blood chemistry and a
pathologic examination were performed at sacrifice.
Silkworth et al. (1982) noted that the LD50's for contaminated soot and soot
extract were similar at 410 and 327 equivalent ug/kg, indicating that the matrix had only a
small effect on toxicity. If expressed in terms of the content of 2,3,7,8-TCDD, the LD50
from soot is 2.5 ug/kg, which is a factor of seven below the LD50 for 2,3,7,8-TCDD in an
aqueous vehicle, suggesting that other compounds contributed to the toxicity of the soot
and soot extract.
The authors stated that they adopted an aqueous vehicle in these experiments
because it was nontoxic and provided a stable suspension of soot; they regarded this
vehicle as more appropriate for modeling of human exposure conditions than an oil vehicle.
The data from these experiments also demonstrate that use of an oil vehicle leads to
substantially greater 2,3,7,8-TCDD toxicity than does an aqueous vehicle.
Comparison of mortality and weight loss in groups of female guinea pigs receiving
500 ug/kg of soot or the equivalent amount of soot extract suggests that the extract may
be somewhat more toxic; however, all six animals died in the 1,000 ug/kg soot group,
while four out of five died in the 500 ug/kg extract group. Taken together, these data
indicate that the soxhlet extract of soot in an aqueous vehicle was between one and two
times as toxic as the soot itself. It is likely that a larger difference in toxicity would have
been observed if the soot extract was in an oil vehicle.
8-28 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Van den Berg et al. (1983) fed small groups of male Wistar rats fly ash from a
municipal incinerator (pretreated with HCI) containing dioxins and furans, a soxhlet extract
of the fly ash, or a purified extract of the ash that was obtained using column
chromatography. 2,3,7,8-TCDD was present as 3.3% of the TCDD isomer group in the
fly ash extract. (The authors did not specify whether this reference was to crude or
purified extract.) 2,3,7,8-TCDF was present as 17.9% of the tetra-CDF isomer group in
the extract. The rats were fed 2 g/d fly ash mixed with diet or the residual from 2 mL/d
extract after the extract was mixed with diet and the solvent was evaporated. The animals
were exposed to the treated diet for 19 days, and then sacrificed, and the liver tissue was
analyzed for the presence of dioxins and furans.
Approximately 1% of the 2,3,7,8-TCDD dose from fly ash was retained in the liver,
and approximately 4% of the dose of this isomer from fly ash extract was so retained.
The corresponding percentages for 2,3,7,8-TCDF are 0.3 and 1.0. Data on the retention of
isomer groups in adipose tissue were presented for the extract-treated groups but not for
the fly-ash-treated group. The concentrations of the various isomers in adipose tissue are
comparable to, or less than, the concentrations in liver tissue.
The U.S. EPA (1985b) reported a half-life for elimination of 2,3,7,8-TCDD in the rat
of 20 days at high dose. If a similar half-life is assumed in this experiment, the quantities
of 2,3,7,8-TCDD in the animals at the end of the 19-day feeding experiment would be
significantly less than the absorbed dose, but still of the same order of magnitude.
However, the recovery percentages in this study are low for both the fly ash and fly ash
extract groups in comparison with other studies in which 2,3,7,8-TCDD was administered
to rats. Fries and Marrow (1975) fed rats diets containing 7 or 20 ppb of 2,3,7,8-TCDD
for a period of up to 42 days. After 14 days of feeding, the rat livers contained an
average of 32% of the cumulative administered dose; at 28 days, 21 % of the dose; and at
42 days, 18% of the dose. Thus, in the van den Berg et al. study, the liver retention of
2,3,7,8-TCDD for the fly ash extract group is a factor of five to eight below what could be
anticipated for the Fries and Marrow data, and the liver retention in the van den Berg
group fed soot is a factor of 20 to 30 lower than that seen by Fries and Marrow. Data
from Kociba et al. (1976), Rose et al. (1976), and Kociba et al. (1978) lead to similar
conclusions to those from the Fries data regarding the fraction of cumulative 2,3,7,8-
TCDD dose retained in the rat liver.
8-29 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
An explanation of the low level of recovery for the animals receiving the soxhlet
extract of soot is not apparent. It is possible that the presence of multiple compounds
affected absorption or metabolism in the rats fed soot and soot extract.
A second approach to the van den Berg et al. data is to compare the ratios of liver
concentrations for dioxins in fly-ash-treated animals to the concentrations in extract-
treated animals. These ratios, based on measurements in small numbers of animals,
indicate a substantial bioavailability of dioxin and furan compounds from the tested fly ash.
This availability varied among the different isomers with the value of 0.3 for 2,3,7,8-
TCDD, indicating that this isomer was three times as available from fly ash extract as from
fly ash.
Van den Berg (1985) fed fly ash (pre-treated with HCI) to Wistar rats, guinea pigs,
and Syrian golden hamsters. Fly ash was mixed with standard laboratory diet at 2.5% by
weight, and animals were allowed to eat ad libitum. The amount of fly ash consumed by
each group of five rodents was determined by the authors. For each species there were
three groups of animals each fed fly ash for approximately 32 days (group I), 60 days
(group II), or 94 days (group III). Concentrations of dioxin and furan isomer groups in the
food were presented, and include 1.4 ng/g TCDD compounds and 2.1 ng/g TCDF
compounds.
The authors presented calculated recovery percentages for the cumulative dose of
specific isomers in the rodent liver. For 2,3,7,8-TCDD in guinea pigs, 3.7%, 0.9%, and
1.4% of the administered dose was recovered in the liver in groups I, II, and III,
respectively. The 32-day (group I) recovery percentage is somewhat higher than seen in
the lower dose groups receiving 2,3,7,8-TCDD contaminated soil in McConnell et al.
(1984). The value in hamsters was approximately 2% (only reported for group II), and
analytical problems prevented this determination in rats. No other TCDD compounds were
quantified. Similarly, for 2,3,7,8-TCDF, guinea pigs showed retention of 4.7%, 2.2%,
2.5% of the administered dose in groups I, II, and III, respectively. For both 2,3,7,8-TCDD
and 2,3,7,8-TCDF the recovery percentages in guinea pigs at 32 days were approximately
a factor of 4 to 15 higher than that observed in the van den Berg et al. (1983) study in
rats.
Other TCDD compounds that were present showed comparable or somewhat lower
retention, averaging 1 % to 2% over the animals groups. No TCDD or TCDF compounds
8-30 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
were detected in hamster liver or analyzed for in rat liver. Higher chlorinated isomers most
typically showed retention in the range of 2% to 5% in rat liver and 1 % to 3% in guinea
pig liver, with the exception of 2,3,4,7,8-PnCDF (9.8%, 8.3%, and 11.3% in the hamster
groups). Few other compounds were found in hamster liver, but 2,3,4,7,8-PnCDF was
found with a recovery of 5% to 8% and 2,3,4,7,8-HxCDD was found at 3% to 7%.
As with other experiments in which the retention of dioxins in the liver has been
determined, these percentages place a lower bound on the bioavailability of the dioxins
but, because not all dioxin is localized in the liver, do not permit bioavailability to be
estimated without knowledge of the elimination of the administered dose over time and
the quantity of dioxins in the remainder of the organism. No positive control group
receiving 2,3,7,8-TCDD was included for comparison.
Poiger and Schlatter (1980) conducted several experiments in Sprague-Dawley rats
(180 to 220 g) in which liver concentrations of tritium label from 2,3,7,8-TCDD were
determined using various doses and vehicles. All experiments consisted of a single gastric
intubation of 2,3,7,8-TCDD-containing material, followed by animal sacrifice at
predetermined times. The doses used were well below the LD50 in the rat (the maximum
dose applied was 5 ug/kg), and no deaths or toxic effects were reported.
In a preliminary experiment, rats were treated with 14.7 ng/rat 2,3,7,8-TCDD in
ethanol. These data indicate substantial localization of 2,3,7,8-TCDD in the rat liver, with
a decrease of a factor of two in the fraction of the dose in the liver between 1 and 4
days. Poiger and Schlatter (1980) conducted all further studies with sacrifice at 24 hours
to maximize the recovery of 2,3,7,8-TCDD from the liver.
In a second experiment, the authors administered 2,3,7,8-TCDD doses in ethanol
ranging from 15 to 1,070 ng/rat to groups of six rats. They found a graded increase in
percentage retained in the liver from 37% _+_ 1 % at the 15 ng dose to 51 % _+ 4% at 280
ng. At the high-dose point, the percentage may have fallen (42% +_ 10% at 1,070 ng).
In a further experiment, 2,3,7,8-TCDD was administered at low dose in a series of
vehicles. These data demonstrate that administration of 2,3,7,8-TCDD in soil reduced the
retention of the dose in the liver to 66%, or 44% of the retention seen with 2,3,7,8-TCDD
in ethanol. The lower value, 44% was obtained for soil that was aged for 8 days at 30-40
°C following addition of 2,3,7,8-TCDD. This observation is consistent with the findings of
other studies reported here that 2,3,7,8-TCDD from environmental soil (naturally aged)
8-31 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
was generally less available than 2,3,7,8-TCDD freshly added to clean samples of these
soils. The aqueous suspension of 2,3,7,8-TCDD in activated carbon showed little
evidence of bioavailability; this is supported by the authors' measurements showing that
2,3,7,8-TCDD was only slightly extractable from the activated carbon matrix by various
solvents. In contrast, 58% to 70% of 2,3,7,8-TCDD could be recovered from soil samples
by washing with hexane/acetone (4:1 v/v).
Poiger and Schlatter (1980) also presented results from several skin application
experiments with TCDD-containing materials using rats and rabbits (not reviewed here).
Bonaccorsi et al (1984) reported the results of a study of gut absorption of 2,3,7,8-
TCDD from soil taken from the Seveso, Italy accident site. Soil containing 81 _+. 8 ppb
2,3,7,8-TCDD from the "highly contaminated" area in Seveso was administered to albino
male rabbits (2.6 +_ 0.3 kg) in daily gavage doses for seven days. Additional samples of
clean soil were spiked with 2,3,7,8-TCDD in the laboratory to yield 10 and 40 ppb
contamination levels and were administered to rabbits following the same protocol. For
comparison, rabbits were also treated with 2,3,7,8-TCDD in solution in acetone-vegetable
oil (1:6) or alcohol-water (1:1). Rabbits were sacrificed on the day after treatment
stopped and liver concentrations of 2,3,7,8-TCDD were measured. The authors did not
remark on the presence or absence of toxicity in the treated rabbits. EPA (1985a) reports
values for the single dose LD50 of 2,3,7,8-TCDD in rabbits of 11 5 and 275 ug/kg. The
total doses received by the rabbits in this study were approximately 54, 107, and 215
ug/kg over seven days. Based on this comparison, there is a likelihood that toxic effects
occurred in the Bonaccorsi work, and noted above, toxicity has the potential to affect the
tissue concentrations of 2,3,7,8-TCDD. For this reason the most appropriate comparisons
among these data are between groups showing similar liver concentrations of 2,3,7,8-
TCDD, which may then be inferred to have experienced similar toxic effects.
That this method of comparison is desirable can also be seen from the Bonaccorsi
data, where both solvent vehicle groups and the spiked soil groups show an increase of
the fraction of the dose in the liver at the higher administered doses. However, it should
be mentioned that use of two different solvent vehicles complicates interpretation. Similar
liver concentrations of 2,3,7,8-TCDD were seen in the 40 ug/d solvent vehicle and 80
ug/d Seveso soil groups. Comparing the percentage of liver retention in these two groups
indicates absorption from Seveso soil was 40% of that from the solvent vehicle. Using
8-32 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
the same approach, comparison of the 80 ug/d solvent vehicle and 160 ug/d Seveso soil
groups indicates that absorption from the soil was 41 % of that from the solvent.
The same approach can be used to compare absorption from the solvent vehicle
and from the spiked soil. In this case the 40 ug/d solvent vehicle group had the liver
concentrations closest to either the 40 or 80 ug/d spiked soil groups. Comparison of the
percentage of dose in the liver indicates absorption from spiked soil is 68-73% of that
from the solvent vehicle. Bonaccorsi et al (1984) work conducted with either aged or non-
aged spiked soil but do not present data to allow a comparison of these groups.
Shu et al. (1987, as cited by Leung and Paustenbach, 1987) study of 2,3,7,8-
TCDD from the Missouri site tested by McConnell et al. (1984). Their paper "reports an
oral bioavailability of approximately 43% in the rat dosed with environmentally
contaminated soil from Times Beach, Missouri. This figure did not change significantly
over a 500-fold dose range of 2 to 1450 ng 2,3,7,8-TCDD per kg of body weight for soil
contaminated with approximately 2, 30 or 60 ppb of 2,3,7,8-TCDD. The data from this
study is not now available to the Exposure Assessment Group for review.
8.4.1.2 Summary of Bioavailability
Table 8-4 summarizes data that are pertinent to the bioavailability of 2,3,7,8-TCDD
from environmental matrices. Studies of bioavailability, which examined soil samples,
soot, and fly ash, have utilized three methodologies: measuring acute toxicity, retention of
2,3,7,8-TCDD in the liver, and induction of hepatic enzymes.
Among the five samples of soil from contaminated sites that have been tested,
three have shown substantial bioavailability, e.g., 25% to 50%, when compared with
2,3,7,8-TCDD in corn oil gavage. A fourth soil sample was compared with 2,3,7,8-TCDD
administered in a solvent vehicle, and fell in this range. The fifth soil, tested by Umbreit et
al. (1986a,b; 1987a,b) showed bioavailability substantially less than the other soils tested.
While difficult to gauge quantitatively, dioxin from this soil may be an order of magnitude
less available than from the other soils.
Additionally, three samples of soil spiked with 2,3,7,8-TCDD have been tested for
bioavailability, including one sample in which the 2,3,7,8-TCDD was incubated with soil at
an elevated temperature. The 2,3,7,8-TCDD added to these soil samples proved to be
highly available (e.g., 40% to 70%).
8-33 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
In one study, soot from a transformer fire containing dioxins and furans proved
similarly toxic to a soxhlet extract of the soot in an aqueous vehicle. However, the soot
extract may have proved more toxic if delivered in corn oil, as was 2,3,7,8-TCDD in the
soil studies. The availability of 2,3,7,8-TCDD and other dioxins and furans from
incinerator fly ash have been addressed by van den Berg et al. in extended feeding studies.
In these studies, liver retention of 2,3,7,8-TCDD from either fly ash or fly ash extract
proved low, with availability from fly ash being approximately 25% of that from the
extract.
The individual studies reviewed have a variety of limitations, as discussed in the
preceding text. A notable limitation was that some experiments were conducted at using
highly toxic doses of 2,3,7,8-TCDD, so that determination of bioavailability was
complicated by wasting and early death of the test animals. It should also be noted that,
while the relative retention of 2,3,7,8-TCDD in the liver can serve as an appropriate
indication of differences in bioavailability between samples, the percentage of dose found
in the liver only places a lower bound on absorption. This is particularly relevant to
experiments where animals have been maintained for many weeks after dosing and an
undetermined quantity of 2,3,7,8-TCDD has been excreted.
Finally, toxicity data for mixtures for which both toxicity and bioavailability of
individual compounds may vary are difficult to interpret quantitatively in terms of
bioavailability.
As presented in U.S. EPA (1985c), Rose et al. (1976) determined gut absorption of
2,3,7,8-TCDD in a 1:25 mixture of acetone to corn oil (by volume) in the rat. In both
single dose and multiple dose experiments, measured absorption was approximately 85%.
Assuming that absorption from pure corn oil is similar to that from this mixture, and
assuming that absorption in other species for which data are not available is similar, the
85% factor can be applied to the data presented here to obtain an approximate range for
typical 2,3,7,8-TCDD absorption from soil. Using this factor, the estimated relative
bioavailability of 2,3,7,8-TCDD from soil is 25% to 50% and, when compared with corn
oil, provides an estimate of gut absorption of 20% to 40% of ingested 2,3,7,8-TCDD in
soil. This estimate is comparable with the 20% to 26% absorption from 2,3,7,8-TCDD
treated soil from the work of Poiger and Schlatter (1980).
Recognizing these limitations, the weight of evidence indicates that 2,3,7,8-TCDD
8-34 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
is often highly available from environmental materials. However, in one tested soil sample
the compound was substantially less bioavailable. While the data are too sparse to allow a
prediction as to whether a particular environmental sample will prove more or less
bioavailable, one important suggestion has emerged. In the two samples that have proved
least bioavailable (the Umbreit et at. (1986a) manufacturing site soil sample, and 2,3,7,8-
TCDD on activated carbon tested by Poiger and Schlatter (1980)) the 2,3,7,8-TCDD was
largely resistant to solvent extraction. This was not the case for more bioavailable
materials.
Further research, using short-term experiments in which animals are handled under
identical conditions and are fed dioxins in different media, is needed for an improved
comparison of absorption between different environmental samples. Acutely toxic doses
should be avoided to ensure that tissue concentrations are directly interpretable.
Experiments studying both tissue retention and enzyme induction should prove valuable for
this research. Whole-body levels of 2,3,7,8-TCDD need to be related to liver
concentrations, and the effects of metabolism need to be addressed. The vehicle of
administration has been shown to affect acute 2,3,7,8-TCDD toxicity, and vehicle effects
need to be considered in designing experiments.
8.4.2 Distribution
Ryan et al. (1985) examined the distribution of 2,3, 7,8-TCDD in two humans at
autopsy. On a weight basis, they determined that there were 6 ppt of TCDD in fat, 2 ppt
in liver and below levels of detection in kidney and muscle. They reported that on a per
lipid basis the levels were similar between tissues. It is important to note that one of
these subjects suffered from a fatty liver syndrome, possibly resulting in higher levels in
the liver than might normally be found in healthy individuals.
Poiger and Schlatter (1986) estimated that about 90% of the total body burden of
2,3,7,8-TCDD was sequestered in fat. Levels of 2,3,7,8-TCDD averaging 5-10 ppt have
been reported for background populations in St. Louis, MO, by Graham et al. (1986), in
Atlanta, GA, and Utah by Patterson et al. (1986). These data consistent with the lipid
bioconcentrations assumptions made in calculations of daily intakes (vidae supra).
Patterson et al. (1987) developed a high resolution gas chromatographic/high
resolution mass spectrometric analysis for 2,3,7,8-TCDD in human serum. A high
8-35 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
correlation between adipose tissue and serum concentrations when adjusted for total lipid
content. The reader is referred to other documents (US EPA, 1992; Schlatter, 1991;
Schecter, 1991) for more details on the distribution and elimination.
8.4.3 Determination of Daily Intake Dose from Exposure Concentrations
As was discussed in Section 8.3.1 it would be most advantageous to know more
about the kinetics of absorption in the various animal species and the human. This is
necessary for both extrapolation between species in the risk assessment and for
determining body burdens from levels in the exposure media. For the time being, until
more data become available regarding the kinetic absorption constants a slightly modified
approach can be used. Basically, the needed information is the concentration of the toxins
in the media, the fraction of toxin absorbed from each of the media, and the amount of
media coming into contact with the body. The following equation describes this in more
detail.
D = Y^fi ci ui (8-30)
Where:
D = Daily Intake
f, = Fraction absorbed from various i media
Cj = Concentration in various i media
Uj = Amount of Contact of i media, (gm/time for food, L/time for air, etc.)
This approach should be used with caution. The major assumption which impacts
upon equation (8-30) is that the fraction absorbed is constant across various
concentrations and doses. As discussed in Section 8.4.2 and in US EPA (1992) this
assumption cannot always be considered to be sound. It most probably only applies at
low doses and within any one species. It is an approach which, with care, can be used for
certain conditions to give estimates until more reliable kinetic absorption data become
available.
8-36 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
REFERENCES FOR CHAPTER 8
Andersen, Melvin E.; Greenlee, Willam F. (1991) Biological determinants of TCDD
pharmacokinetics and their relation a biologically based risk Assessment in
Biological Basis for Risk Assessment of Dioxins and Related Compounds. Michael
Gallo, Robert Scheuplein and Kees Van Der Heijden, eds.; Banbury Report 35, Cold
Spring Harbor Laboratory Press.
Bonaccorsi, A.; diDomenico, A.; Fanelli, R.; Merli, F.; Motta, R.; Vanzati, R.; Zapponi, G.A.
(1984) The influence of soil particles adsorption on 2,3,7,8-tetra-chlorodibenzo-p-
dioxin biological uptake in the rabbit. Arch. Toxicolo. Suppl. 7:431-434.
Bowman, R.E.; Schantz, S.L.; Weerasinghe, N.C.A.; Gross, M.L.; Barsotti, D.A. (1989)
Chronic dietary intake of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) at 5 or 25
parts per trillion in the monkey: TCDD kinetics and dose-effect estimate of
reproductive toxicity Chemosphere 18(1-6):243-252.
Brewster, D.W.; Elwell, M.R.; Birnbaum, L.S. (1988) Toxicity and disposition of 2,3,4,7,8-
pentachlorodibenzofuran (4PeCDF) in the rhesus monkey (Macaca mulatta).
Toxicology and Applied Pharmacology 93:231-246.
Denison, Machael S.; Phelphs, Cynthia L.; Dehoog, jane; Kim, Hun J.; Bank, Paula A.; Yao,
Eveline F. (1991) Species variation in Ah receptor transformation and DNA binding
in Biological Basis for Risk Assessment of Dioxins and Related Compounds, Michael
Gallo, Robert Scheuplein and Kees Van Der Heijden, eds.; Banbury Report 35, Cold
Spring Harbor Laboratory Press.
Fries, G.F.; Marrow, G.S. (1975) Retention and excretion of 2,3,7,8-tetrachlorodibenzo-p-
dioxin (TCDD) by rats. J. Agric. Food Chem. 23:265-269.
Furst, Per; Fiirst, Christiane; Wilmers, Klaus (1991) Body burden with PCDD and PCDF
from food in Biological Basis for Risk Assessment of Dioxins and Related
Compounds. Michael Gallo, Robert Scheuplein and Kees Van Der Heijden, eds.;
Banbury Report 35, Cold Spring Harbor Laboratory Press.
Gasiewicz, Thomas A.; Henry, Ellen C. (1991) Different forms of the Ah receptor:
possible role in species- and tissue-specific responses to TCDD. in Biological Basis
for Risk Assessment of Dioxins and Related Compounds, Michael Gallo, Robert
Scheuplein and Kees Van Der Heijden, eds.; Banbury Report 35, Cold Spring Harbor
Laboratory Press.
Graham, M.; Hileman, F.D.; Orth, R.G.; Wendling, J.M.; Wilson, J.W. (1986)
Chlorocarbons in adipose tissue from a Missouri population. Chemosphere 15:1595-
1600.
Huetter, R.; Phillipi, M. (1982) Studies on microbial metabolism of TCDD under laboratory
conditions. Permamon Ser. Environ. Sci. 5:87-93.
8-37 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Kaminski, L.S.; DeCapiro, A.P.; Gierthy, J.F.; Silkworth, J.B.; Tumasonis, C. (1985) The
rule of environmental matrices and experimental vehicles i chlorinated dibenzodioxin
and dibenzofuran toxicity. Chemosphere 14:685-695.
King, F.G.; Dedrick. R.L.; Collins, J.M.; Matthews, H.B.; Birnbaum, L.S. (1983)
Physiological model for the pharmacokinetics of 2,3,7,8-tetra-chlorodibenzofuran in
several species. Toxicology and Applied Pharmacology 67:390-
Kociba, R.J.; Deeler, P.A.; Park, C.N.; Gehring, P.J. (1976) 2,3,7,8-Tetra-chlorodibenzo-p-
dioxin (TCDD): results of 13-week oral toxicity study in rats. Toxicol. Appl.
Pharmacol. 35:553-573.
Kociba, R.J.; Deyes, D.G.; Beyer, J.E., et al. (1978) Results of a two year chronic toxicity
and oncogenic study of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. Toxicol. Appl.
Pharmacol. 46(2):279-303.
Leung, H.; Paustenbach, J. (1987) A proposed occupational exposure limit for 2,3,7,8-
tetrachlorodibenzo-p-dioxin. Environmental Health and Safety, Syntex, U.S.A. Inc.,
Palo Alto, CA.
Leung, H-W.; Ku, R.H.; Paustenbach, D.J.; Andersen, M.E. (1988) A physiologically based
pharmacokinetic model for 2,3,7,8-tetra-chlorodibenzo-p-dioxin in C57BL/6J and
DBA/2J mice. Toxicology Lett. 42: 15-28.
Leung, H-W.; Poland, A.; Paustenbach, D.J.; Murray, F.J. Andersen, M.E. (1990)
Pharmacokinetic of [125l]-2-iodo-3,7,8-trichlorodibenzo-p-dioxin in mice: Analysis
with a physiological modeling approach. Toxicology and Applied Pharmacology
103: 411-419.
Luceir, G.W.; Rumbaugh, R.C.; McCoy, A.; Haas, R.; Harvan, D.; Albro, P. (1986)
Ingestion of soil contaminated with 2,3,7,8-tetrachlordibenzo-p-dioxin (TCDD) alters
hepatic enzyme activities in rats. Fund. Appl. Toxicol. 6: 364-371.
McConnell, E.E.; Lucier, G.W.; Rubaugh, R.C.; et al. (1984) Dioxin in soil: Bioavailability
after ingestion by rats and guinea pigs.
Nau, H.; Bab, R.; Neuber, D. (1986) Transfer of 2,3,7,8-tetrachlorodibenzo-p-dioxin
(TCDD) via placenta milk, and postnatal toxicity in mouse. Arch. Toxicol. 59:36-
40.
Nessel, C.S.; Amoruso, M.A.; Umbreit, T.H.; Gallo, M.A. (1990) Hepatic aryl hydrocarbon
hydroxylase and cytochrome P450 induction following the transpulmonary
absorption of TCDD from intratracheally instilled particles. Fund. Appl. Toxicol.
15:500-509.
Patterson, D.G.; Holler, J.S.; Lapez, C.R., Jr. (1986) High resolution gas
chromatographic/high resolution mass spectrometric analysis of human adipose
tissue for 2,3,7,8-tetrachorodibenzo-p-dioxin. Anal. Chem. 58: 705-713.
8-38 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Patterson, D.G.; Hampton, L.; Lapeza, C.R., Jr; (1987) High resolution gas
chromatographic/high resolution mass spectrometeric analysis of human serum on
whole weight and lipid basis for 2,3,7,8-tetrachlorodibenzo-p-dioxin. Anal. Chem.
59:2000-2005.
Perdew, G.H.; Hollenbeck, C.E. (1990) Analysis of photoaffinity labeled aryl hydrocarbon
receptor heterogeneity by two dimensional gel electrophoresis. Biochemistry 29:
6210-6214.
Philippe, M.; Drasnobagew, V.; Zeyer, J.; Huetter, R. (1981) Fate of 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) in microbial cultures and soil under laboratory
conditions. FEMS Symp. 12:2210-2233.
Poiger, H.; Schlatter, C. (1980) Influence of solvents and adsorbents on dermal and
intestinal absorption of TCDD. Food Cosmet. Toxicol. 18: 477-481.
Poiger, H.; Schlatter, C. (1986) Pharmacokinetics of 2,3,7,8-TCDD in man. Chemosphere
15(9-12): 1489-1494.
Rose, J.Q.; Ramsey, J.C.; Wentzler, Th.H. (1976) The fate of 2,3,7,8-tetrachlorodibenzo-
p-dioxin following single and repeated oral doses to the rat. Toxicol. Appl.
Pharmacol. 36:209-226.
Ryan, J.J; Schecter, A.; Lizotte, R.; Sun, W.F.; Miller, L. (1985) Tissue distribution of
dioxins and furans in humans from the general population. Chemosphere. 14(6/7):
929-932.
Schecter, Arnold (1991) Dioxins and related chemicals in humans and in the environment
in Biological Basis for Risk Assessment of Dioxins and Related Compounds, Michael
Gallo, Robert Scheuplein and Kees Van Der Heijden, eds.; Banbury Report 35, Cold
Spring Harbor Laboratory Press.
Schlatter, Christian (1991) Data on kinetics of PCDDs and PCDFs as a prerequisite for
human risk assessment in Biological Basis for Risk Assessment of Dioxins and
Related Compounds, Michael Gallo, Robert Scheuplein and Kees Van Der Heijden,
eds.; Banbury Report 35, Cold Spring Harbor Laboratory Press.
Shu, J.; Paustenbach, D.; Murray, F.J. (1988) Bioavailability of soil bound TCDD: Oral
bioavailability in the rat. Fund. Appl. Toxicol. 10:648-654.
Silkworth, J.; Marti, D.; DeCapri, A.; Rej, R.; O'Keefe, P.; Kaminsk, L. (1982) Acute
toxicity in guinea pigs and rabbits of soot from a polychlorinated biphenyl-
containing transformer fire. Toxicology and Applied Pharmacology. 65:425-439.
Stanley, J.S.; Boggess, K.; Onstot, J.; Sack, T.; Remmers, J.; Breen, J.; Kutz, F.W.;
Robinson, P.; Mack, G. (1986) PCDDs and PCDFs in human adipose tissues from
the EPA FY82 NHATS repository. Chemosphere 15:1605-1612.
8-39 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Umbreit, T.J.; Patel, D.; Gallo, M.A. (1985) Acute toxicity of TCDD contaminated soil
from an industrial site. Chemosphere 14:945-947.
Umbreit, T.H.; Hesse, E.J.; Gallo, M.A. (1986a) Bioavailability of dioxin in soil from a
2.4.5.-T manufacturing site. Science 232:497-499.
Umbreit, T.H.; Hesse, E.J.; Gallo, M.A. (1986b) Comparative toxicity of TCDD
contaminated soil from Times Beach, Missouri, and Newark, New Jersey.
Chemosphere 15(9-12): 2121-2124.
Umbreit, T.H.; Hesse, E.J.; Gallo, M.S. (1987a) Reproductive toxicity of female mice of
dioxin-contaminated soils from a 2,4,5-trichlorooxacetic acid manufacturing site.
Arch. Environmental Contamination and Toxicology 16: 461-466.
Umbreit, T.H.; Hesse, E.J.; Gallo, M.S. (1987b) Differential bioavailabilty of 2,3,7,8-
Tetrachlorodibenzo-p-dioxin from contaminated soils. American Chemical Society
Symposium Series No. 338, "Solving hazardous waste problems: Learning from
dioxin.
US EPA (1985a) Compilation of air pollutant emission factors, Vol 1. Office of Air Quality
Planning and Standards, Research Triangle Park, N.C.
US EPA (1985b) Rapid assessment of exposure to particulate emission from surface
contamination sites. Office of Solid Waste and Emergency Response, Washington,
DC EPA-600/8-85-002.
US EPA (1985c) Health assessment document for polychlorinated dibenzo-p-dioxins.
Office of Air Planning and Standards, Research Triangle Park, NC. EPA-600/8-84-
014F.
US EPA (1992) Health Assessment For 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) and
related compounds. Preliminary Draft.
van den Berg, M.; OIlie, K.; Hutzinger, 0. (1983) Uptake and selective retention in rats of
orally administered chlorinated dioxins and dibenzofurans from fly-ash and fly-ash
extract. Chemosphere 12:537-544.
van den Berg, M.; Vroom, A.; van Greevenbroek, M.; Olie, K.; Hutzinger, 0. (1985)
Bioavailability of PCDDs and PCDFs adsorbed on fly ash in the rat, guinea pig, and
Syrian golden hamster. Chemosphere 14:865-869.
Whitlock, James P., Jr. (1991) Mechanism of dioxin action: Relevance to risk assessment
in Biological Basis for Risk Assessment of Dioxins and Related Compounds, Michael
Gallo, Robert Scheuplein and Kees Van Der Heijden, eds.; Banbury Report 35, Cold
Spring Harbor Laboratory Press.
8-40 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
9. DEMONSTRATION OF METHODOLOGY
9.1. INTRODUCTION
This document has provided methodologies and background information for
conducting site-specific exposure assessments to the dioxin-like compounds. Chapter 2
described physical and chemical properties of these compounds, and Chapter 3 described
the occurrence of these compounds in environmental and exposure media. Chapter 4
summarized an overall exposure assessment framework, and Chapter 7 provided details on
development of exposure scenarios including exposure pathways and associated
parameters. Chapter 5 provided methodologies to estimate exposure media concentrations
for four sources of contamination, which were termed source categories. Chapter 6
discussed a principal source of dioxin-like compounds in the environment, incinerators.
Incinerator stack emissions and disposal of incinerator fly ash were two of the four source
categories identified in Chapter 5. Chapter 6 also discussed and demonstrated the use of
the Industrial Source Complex (ISC) model to generate air concentrations of contaminants
and deposition rates at various distances from a hypothetical incinerator. Chapter 8 on
pharmacokinetics summarized information on uptake and distribution of dioxin-like
compounds and developed a model to predict blood levels resulting from exposure.
The purpose of this chapter is to put all this information together and demonstrate
the site-specific methodologies that have been developed. For this demonstration,
exposure scenarios are developed which are associated with the four source categories.
These categories were defined in Chapter 5, and are:
• On-site soil: The source of contamination is soil and both the source and
exposure site are on the same tract of land.
• Off-site soil: The source of contamination is soil and this source is located
distant and upgradient/upwind from the site of exposure.
• Incinerator stack emissions: Exposed individuals reside downwind of the
incinerator and are exposed to resulting air-borne vapor phase contaminants
originating at the incinerator, and soil on their property is impacted by
deposition of contaminated particulates.
Incinerator ash disposal in a landfill: Contaminated ash is spread onto the surface
of an active landfill, and exposed individuals reside upgradient/upwind of the
9-1 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
landfill.
The demonstration in this chapter is structured around what are termed exposure
scenarios. As defined in Chapter 7, an exposure scenario includes a description of the
physical setting of the source of contamination and the site of exposure, behavior of
exposed individuals, and exposure pathways. Chapter 7 also described the objective of
exposure assessors to determine "central" and "high end" exposure scenarios. This
objective was an important one for this demonstration, and the strategy to design such
scenarios is detailed in Section 9.2 below. Three dioxin-like compounds are demonstrated
for each of the exposure scenarios, including 2,3,7,8-TCDD, 2,3,4,7,8-PCDF, and
2,3,3',4,4',5,5'-HPCB.
Exposure results, expressed in lifetime average daily doses (LADDs in mg/kg-day),
are listed separately for each pathway and for each contaminant. These exposure
estimates could be summed across all contaminants for a given exposure pathway to
estimate the total risk from that pathway. This procedure uses Toxicity Equivalency
Factors (TEFs), as described in Chapter 4, Section 4.3. This summation was not done for
this demonstration since this document only focuses on exposure and not risk, and also
because TEFs for PCBs have not been well established.
Section 9.2 describes the strategy for development of the demonstration exposure
scenarios. Section 9.3. gives a complete summary of the example scenarios. Section 9.4
provides some detail on the three example compounds chosen for demonstration. Section
9.5 describes the source strength terms for the scenarios. Section 9.6 summarizes the
results for all scenarios, which are exposure media concentrations for all exposure
pathways, and exposure estimates which are Lifetime Average Daily Doses (LADD) for all
pathways and for all example compounds.
9.2. STRATEGY FOR DEVISING EXPOSURE SCENARIOS FOR
DEMONSTRATION PURPOSES
Chapter 4 of this document culminated with Figure 4.1, a roadmap for assessing
exposure to dioxin-like compounds. These procedures assess individual exposures to
known sources of contamination. Chapter 7 discussed central and high end exposure
patterns and developed parameters for each exposure pathway consistent with these
patterns. The demonstration in this chapter attempts to merge procedures for estimating
9-2 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
individual exposures to known sources of contamination and current thoughts on devising
central and high end exposure scenarios.
An exposure assessor's first task in determining patterns of exposure is to fully
characterize the exposed population in relation to the source of contamination. If the
extent of contamination can be characterized, then the exposed population would be
limited to those within the geographically bounded area. An example of this situation
might be an area impacted by incinerator stack emissions. Chapter 6 demonstrated the
use of ISC atmospheric dispersion model on a hypothetical but realistic incinerator using
meteorological data from Tampa, Florida, and showed that maximum ambient air
concentrations and particulate deposition rates occurred east of the hypothetical
incinerator, and that the maximum values of these quantities occurred 200 m east.
Simulations were performed for as far away as 40 kilometers in all directions, although
results were only displayed (for sake of brevity) for up to 5 kilometers in the easterly
direction. By overlaying the concentration isopleths onto a population density map, the
exposed population can be identified. If the extent of contamination is not as clearly
defined, such as extent of impact of nonpoint source pollution (impacts from use of
agricultural pesticides, e.g.) or the compound is found ubiquitously without a clearly
defined source, then the emphasis shifts from geographical bounding to understanding
ambient concentrations, exposure pathways and patterns of behavior in general
populations.
After identifying the exposed population, the next task is to develop an
understanding of the continuum of exposures. The exposures faced by the 10 percent of
the population most exposed has been defined as high end exposures. Those faced by the
middle of the continuum are called central exposures. Another important estimate of
exposure level is a bounding exposure, which is the one faced by the individual(s) most
exposed (EPA, 1991 a). Arriving at such an understanding can be more of an art than a
science. One consideration is the proximity of individuals within an exposed population to
the source of contamination. For the incinerator example discussed above, one might
begin an analysis by assuming that bounding or high end exposures occur 200 meters east
of the incinerator, the location modeled as having the highest ambient air concentrations
and particle deposition rates. Another important consideration is the relative contribution
of different exposure pathways to an individual's total exposure. While individuals residing
9-3 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
200 meters east of the incinerator might experience the highest inhalation exposures, they
may not experience other exposure pathways associated with contaminated soil on their
property - such as consumption of home grown vegetables, dermal contact, or soil
ingestion. Families with home gardens and individuals who regularly work in those
gardens may reside over a kilometer from the incinerator and possibly be more exposed
because of their behavior patterns. Screening tools, such as the spreadsheets developed
for this assessment, can be used in an iterative mode to evaluate the interplay of such
complex factors. When applied to a real world situation, information should be sought as
to the makeup and behavior patterns of an exposed population.
The demonstration in this chapter attempts to be consistent with the goal of
quantifying central and high end exposures. However, it is not exhaustive in its analysis,
nor should be construed as a case study with generalizable results. Following are bullet
summaries of key features of the structure and intent of the demonstrations.
• Exposed populations: Exposed individuals are assumed to reside in a rural
setting. Exposures occur in the home environment, in contrast to the work
environment or other environments away from home (parks, etc.). The
presumption is made that the sources of contamination of this assessment
can occur in rural settings in the United States. Sources demonstrated
include soil with concentrations characteristic of background levels,
incinerator stack emissions, ash landfills, and areas of soil contamination
with concentrations that have been found in industrial sites (see Section 9.5.
below). It is further assumed that the behavior patterns associated with the
exposure pathways can exist in rural settings. Several of these behaviors
characterized as high end relate to individuals on farms as compared to
behaviors characterized as central for individuals not on farms. The
exposed population for this demonstration, therefore, consists of rural
individuals in farming and non-farming residences. For each of the four
source categories demonstrated, the exposed populations can be further
defined:
On-site soils: The on-site source category is demonstrated with soil
concentrations that have been found and characterized in the
literature as "background" and "rural", or not associated with an
9-4 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
identified source. For this source, the exposed population includes all
individuals within a rural area for which the background concentration
can be considered representative.
Incinerator stack emissions: As indicated earlier, the populations
exposed to incinerator stack emissions can be identified by overlaying
results of an atmospheric dispersion modeling exercise (or using
measured air concentration data) over a map containing population
density information. Such an exercise was not done for this
demonstration. Instead, simulated ambient air concentrations and
deposition rates were taken from tables in Chapter 6 for two
locations, one for use in a central scenario and another for a high end
scenario.
Off-site soil and ash landfills: The two remaining source categories
are both finite areas of soil contamination. The off-site soil source
category is demonstrated with a 10-acre site having elevated soil
concentrations that have been found in the United States and
elsewhere in industrial sites. The other is an ash landfill, whose size
was estimated using information on the quantity of ash generated by
the hypothetical incinerator. The soil concentrations were estimating
based on concentrations estimated to be in the ash. For these source
categories, a working hypothesis is made that the population most
exposed are those residing very near these sites. Their soil is
assumed to become contaminated over time due to the process of
erosion; these processes normally do not carry contaminants long
distances across land, particularly land developed with residences or
where erosion is interrupted with ditches or other surface water
bodies. People from the surrounding community can be impacted by
visiting or trespassing on the contaminated land, volatilized residues
may reach their home environments, they may obtain water and fish
from impacted water bodies, and so on. It seems reasonable to
assume that those residing near these sites comprise the principally
exposed individuals, or equivalently, the individuals experiencing the
9-5 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
high end or bounding exposures associated with these areas of soil
contamination.
Proximity to sources of contamination: As noted above, the on-site soil
contamination source category was demonstrated using soil concentrations
typical of background levels that have been found in rural settings. In this
case, proximity to the source of contamination was not an issue. Proximity
to an incinerator was identified as an important determinant for identifying
the continuum of exposures. Assuming there is a uniform distribution of
exposure-related behaviors among exposed populations, i.e., their behavior
patterns are not a function of where they live in relation to the incinerator,
the most exposed individuals will be those exhibiting high end exposure
behavior nearest the incinerator. This was the assumption made for
purposes of this demonstration. A set of high end exposure behaviors and
pathways were demonstrated for individuals residing 500 meters from the
incinerator, and a set of central exposure pathways were demonstrated for
individuals residing 2000 meters from the incinerator. The highest
exposures would occur at 200 meters, using the model output given in
Chapter 6. Since the intent is to characterize the upper 10 percent with the
high end scenario, and not the bounding estimate, it was felt that 500
meters might be more reasonable. Figures 6.2 and 6.3 show that ground
concentrations at 500 meters are 60% of what they are at 200 meters.
These figures also show that concentrations at 2000 meters are 40% of
what they are at 200 meters, and at 10000 meters are 10% of what they
are at 200 meters. Without rigorous justification, the model output
(concentrations and deposition rates) at 500 and 2000 meters was felt to
appropriately characterize high end and central exposures. The above bullet
justifies a definition of principally exposed individuals as those nearest both
source categories of off-site soil contamination, while recognizing that lesser
exposures can occur for other individuals in the community. These lesser
exposures will not be demonstrated. Instead, the off-site and ash landfill
source categories will only be demonstrated with a single, high end scenario.
Individuals exposed will be assumed to reside 150 meters downgradient
9-6 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
from the sites of soil contamination.
Central and high end exposure patterns: Chapter 7 described the exposure
pathways that are considered in this methodology, and justified assignment
of key exposure parameters (contact rates and contact fractions, exposure
durations, and so on) as central or high end estimates. That chapter notes
that the exposure pathways identified were those that were consistent with
the sources of contamination, and consistent with literature which identified
predominant media where these compounds were found. The bullet above
discussing exposed population indicated that several of the behavior
assumptions were specific to individuals on farm, and that these behavior
patterns were evaluated as high end. High end behaviors assumed to be
different for individuals on farms vs central behaviors for individuals not on
farms include: residing on larger tracts of land (10 acres assumed for
farmers; 1 acre assumed for non-farmers), ingestion of home produced and
impacted beef and milk, tendencies to reside in a single location longer (20
years versus 9 years), tendencies to be present in the home environment
longer (90% of the time versus 75% of the time), and patterns of soil
dermal contact designed to be plausible for farmers working with soil versus
those incidentally contacting soil. Other patterns of behavior modeled as
central and high end are not specifically associated with farming and not
farming, but are assumed to be plausible for individuals in rural settings.
These include home gardening for fruit and vegetables, inhalation exposures,
children that ingest soil, and the use of impacted surface water bodies for
drinking and fish to be ingested.
Plausibility of source strength terms: The objective to determine plausible
levels of source strength contamination was an important one for this
demonstration. The source terms are soil concentrations, incinerator ash
concentrations, and incinerator stack emissions. Section 9.5 describes the
source terms in detail.
Appropriate estimation of exposure media concentrations: The realism of
estimated exposure media concentrations is dependent on the
appropriateness of the models used for such estimations and the assignment
9-7 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
of parameter values for those models. One way to arrive at a judgement as
to the realism of estimated concentrations is to compare predictions with
observations. To the extent possible (i.e., given the availability of
appropriate data), model predictions of exposure media concentrations are
compared with occurrence data in Chapter 10 on Uncertainty. As will be
shown, predictions fell within the realm of observed data. Chapter 5
describes the justification of all model parameter values. Many of the
parameters are specific to the contaminants. Several contaminant properties
were estimated as empirical functions of the contaminant's octanol water
partition coefficient, Kow, and others were measured values. For non-
contaminant parameters such as soil properties, patterns of cattle ingestion
of soil, ash disposal practices, and many others, values were selected with
the incentive to be mid-range and plausible.
9.3. EXAMPLE EXPOSURE SCENARIOS
As noted above, all exposures occur in a rural setting. Exposure pathways were
those which could be associated with places of residence in contrast to the work place or
other places of exposure. The example scenarios are structured so that all the behaviors
associated with high end exposures are included in the "high end" scenarios and all the
central behaviors are in the scenarios characterized as "central". To summarize, the
components which distinguished the high end exposure scenarios in contrast to the central
scenarios include:
• Individuals in the central scenarios lived in their homes and were exposed to
the source of contamination for only 9 years, in contrast to individuals in the
high end scenarios, who were exposed for 20 years (except for the exposure
pathway of soil ingestion, where the individuals are assumed to be children
ages 2-6, and in both the central and high end scenarios, the exposure
duration is 5 years).
• Individuals in the central scenarios lived on properties 1 acre in size, whereas
individuals in the high end scenarios lived on properties 10 acres in size.
• Individuals in the high end scenario associated with the stack emission
source category lived 500 meters from the incinerator, whereas individuals
9-8 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
in central scenario lived 2000 meters from the incinerator.
• High end individuals obtained a portion of their beef and milk using home
produced beef and milk - such individuals are obviously farmers. Central,
and non-farming rural individuals, were not exposed to contaminated beef or
milk.
• Ninety percent of the inhaled air and ingested water by the high end
individuals were assumed to be contaminated, whereas only 75% of these
exposures were with impacted media for the central individuals. This is
based on time at home versus time away from home assumptions for central
vs. high end individuals.
• Although their total intake of fruit and vegetables was assumed to be the
same, a larger proportion of the intake of those food products in the high
end household was home grown and impacted as compared to the central
household.
• The rates of ingestion of soil by children and of recreationally caught and
impacted fish were higher for the high end individuals than the central
individuals.
These are the distinguishing features for the central and high end exposure
scenarios. For the sake of convenience mainly, all the scenarios defined below as high end
are called "farms", and all central scenarios are called "residences". The assertion is not
being made that all behaviors are likely to occur simultaneously (or in some cases, simply
to occur) on a farm or a non-farm residence, although several of the high behavior patterns
are specific to farms. In an exhaustive site-specific analysis, one might begin by
evaluating all possible pathways, further evaluating pathways of most exposure, and then
determining what pathways occur simultaneously for identified individuals in the exposed
population. Only then can be the assessor begin to define a continuum of exposures.
The following bullets describe six exposure scenarios that are demonstrated. The
numbering scheme and titles will be referenced for the remainder of this chapter:
Exposure Scenarios 1 and 2: On-site Soil Contamination, Residence and Farm
Surface soil on a 4,000 m2 (1-acre roughly) rural residence (Scenario 1) and on a
40,000 m2 (10-acre roughly) rural farm (Scenario 2) is contaminated with the three
9-9 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
example contaminants. The concentrations of the contaminants are uniformly set at 1
part per trillion, which was evaluated as reasonable background levels (see Section 9.5
below). Bottom sediment in a nearby small river becomes contaminated. Water and fish
in that river are subsequently impacted. Fish are recreationally caught and eaten, and the
water is extracted for drinking purposes (perhaps at an upstream water system intake,
although such configurations do not impact water concentration predictions;
concentrations for exposure are those that are calculated to occur above contaminated
bottom sediment).
Exposure Scenario 3: Off-site Soil Contamination, Farm
A 40,000 m2 rural farm is located 1 50 m (500 ft roughly) from a 40,000 m2 area
of bare soil contamination; an area that might be typical of abandoned industrial property.
The surface soil at this property is contaminated with the three example compounds to the
same concentration of 1 part per billion. This is evaluated as reasonable for industrial sites
of contamination of dioxin-like compounds, and three orders of magnitude higher than
concentrations for Scenarios 1 and 2. As in the above and all scenarios, bottom sediment
in a nearby river is impacted, which impacts the drinking water supply and fish which are
recreationally caught and consumed by members of this farming household.
Exposure Scenarios 4 and 5: Incinerator Stack Emissions, Residence and Farm
A 4,000 m2 rural residence (Scenario 4) is located 2000 meters from an
incinerator, and a 40,000 m2 (Scenario 5) rural farm is located 500 meters downwind
from an incinerator emitting the three contaminants. The emission rates were developed
from a hypothetical but realistic incinerator: incinerator capacity, stack pollution control
technology, and rates of contaminant emissions were all developed using current
information. The modeling of the transport of these contaminants using the Industrial
Source Complex (ISC) model was also realistic, using a weather data set from Tampa,
Florida. Details on the incinerator and the ISC model application are found in Chapter 6. A
nearby impacted river supplies the drinking water and a portion of the fish diet for both
families.
9-10 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Exposure Scenario 6: Incinerator Ash Disposal in a Landfill Near a Farm
Combined (fly + bottom) ash generated by the hypothetical incinerator is disposed
of in 100+ hectare landfill, in which a 270,000 m2 area (27 hectares, or 67 acres
roughly) is assumed to be active and bare at any time. The landfill is 150 m upgradient
from a 40,000 m2 rural farm. The concentration of the contaminants on the ash were also
hypothetical but realistic. These concentrations are consistent with the incinerator
capacity and pollution control technology of the incinerator designed for Scenarios 4
and 5. Ash mixes with landfill soil and the concentrations at the landfill are assumed to be
half of what they are in the delivered ash. A nearby impacted river supplies the families
with their drinking water and a portion of their fish diet.
9.4. EXAMPLE COMPOUNDS
For purposes of illustration, one compound was arbitrarily selected from each of the
major classes of dioxin-like compounds. They are: 2,3,7,8-tetrachlorodibenzo-p-dioxin,
2,3,4,7,8-pentachlorodibenzofuran, and 2,3,3',4,4',5,5'-heptachloro-PCB. For the
remainder of this chapter, these compounds will be abbreviated as 2,3,7,8-TCDD,
2,3,4,7,8-PCDF, and 2,3,3',4,4',5,5'-HPCB.
These compounds demonstrate a range of expected results because of the
variability of their key fate and transport parameters. The log octanol water partition
coefficients (log Kow) for 2,3,7,8-TCDD, 2,3,4,7,8-PCDF, and 2,3,3',4,4',5,5'-HPCB were
6.64, 6.92, and 7.71, respectively. Whereas the span of reported log Kow ranged from
less than 6.00 to greater than 8.00, only a few reported values were at these extremes.
Increasing log Kow translates to the following trends: tighter sorption to soils and
sediments and less releases into air and water, less accumulation in plants and in cattle
products (beef, milk), and more accumulation in fish. The Henry's Constants for the three
compounds span the range of reported values, with the value of the PCS compound the
highest of all reported at 3.0 * 10~3. There were few values less than the 4.99 * 10~6
reported for 2,3,4,7,8-PCDF. Higher Henry's Constants translate to greater amounts of
volatilization flux. Further information on the chemical properties and fate parameters can
be found in Chapters 2 and 5, respectively.
9-11 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
9.5. SOURCE TERMS
This section describes the source terms for the five settings. Source terms include
the areas of contamination and soil concentrations. The source terms for the stack
i
emission scenarios, 5 and 6, strictly speaking are the emission rates of contaminants from
the incinerator stacks. These are provided in Chapter 6 and not repeated here. This
section does summarize the soil concentrations that result from stack emission
depositions. It also summarizes the exposure site soil concentrations that result from
erosion of contaminated soil from the nearby soil contamination sites (Scenarios 2, 3, 7,
and 8).
Key source terms are summarized in Table 9-1. Following now are discussions on
these terms for all scenarios.
Scenarios 1 and 2
The residence in Scenario 1 is 4,000 m2 and in Scenario 2 is 40,000 m2 in size.
Table 3.1 listed literature values for soil concentrations of the dioxin-like compounds. As
noted in that chapter, concentrations of the coplanar PCBs were not found in the
literature; soil concentrations assigned for 2,3,3',4,4',5,5'-HPCB will be the same as the
other two compounds. These scenarios were designed to demonstrate exposures to low
concentrations which might be considered "background" soil concentrations. Soil
concentrations of 2,3,7,8-TCDD and 2,3,4,7,8-PCDF described as "background" or "rural"
by researchers were found in the non-detect to low ng/kg (ppt) in Illinois, Ohio, and
Minnesota in the United States (EPA, 1985; Reed, et al., 1990), and in Sweden (Broman,
et al. 1990) and England (Greaser, et al., 1989; Stenhouse and Badsha, 1990). Tier 7 of
EPA's National Dioxin Study (EPA, 1987) consisted of "background" sites, or sites that did
not have previously known sources of 2,3,7,8-TCDD contamination. The purpose of this
tier was to provide a basis for comparison for the other 6 tiers of study, which did include
sites of known or suspected 2,3,7,8-TCDD contamination. The results were that 17 of
221 urban sites and only 1 of 138 rural sites had detectable levels of 2,3,7,8-TCDD, with
a range of positives of 0.2 to 11.2 ng/kg (ppt). While the value of 1 ng/kg selected for
these scenarios may not be a true "background" concentration, the intent in designing
Scenarios 1 and 2 was to select a concentration that might be typical of areas where no
known identifiable source impacts the soil.
9-12 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 9.1. Summary of key source terms for the six exposure scenarios and the three
example compounds
I. Soil Concentrations, //g/kg (ppb)
Source Category 1: On-Site Soil
Scenario 1. Central
Scenario 2. High End
Source Category 2: Off-Site Soil
Off-site soil concentration
Scenario 3. High End No-till
Tilled
Source Category 3: Stack Emissions
Scenario 4. Central No-till
Tilled
Scenario 5. High End No-till
Tilled
Source Category 4: Ash Landfill
Landfill soil concentrations
Scenario 6. High End No-till
Tilled
II. Land Areas, m2
Source Category 1: On-site Soil
Scenario 1. Central 4,000
Scenario 2. High End 40,000
Source Category 2: Off-site Soil
Off-site soil 40,000
Scenario 3. High End 40,000
2,3,7,8-
TCDD
0.001
0.001
1.000
0.613
0.031
0.0008
0.00004
0.0013
0.00006
0.700
0.563
0.028
2,3,4,7,8-
PCDF
0.001
0.001
1.000
0.613
0.031
0.009
0.0005
0.014
0.0007
8.200
6.590
0.330
2,3,3',4,4',5,5'-
HPCB
0.001
0.001
1.000
0.613
0.031
0.003
0.0001
0.005
0.0002
2.200
1.770
0.088
Source Category 3: Stack Emissions
Scenario 4. Central 4,000
Scenario 5. High End 40,000
Source Category 4: Ash Landfill
Landfill size 270,000
Scenario 6. High End 40,000
Scenario 3
This scenario was designed to be plausible for properties located near inactive
industrial sites with contaminated soil. The selection of 1 //g/kg (ppb) for the three
compounds was based on 2,3,7,8-TCDD findings associated with the Dow Chemical site
in Midland, Ml (EPA, 1985; Nestrick, et al. 1986) as well as the 100 industrial sites
9-13
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
evaluated in the National Dioxin Study (which included the Dow Chemical site; EPA,
1987). In that study, most of the sites studied had soil concentrations in the parts per
billion range. The farm size was 40,000 m2, as in all high end scenarios. Table 9-1
shows these concentrations for the example compounds at the site of contamination, and
also for the tilled and untilled condition at the sites of exposure. Exposure site soil is
assumed to become contaminated over time due to erosion of soil from the contaminated
site. The "tilled" condition distributes the eroded contaminants to a depth of 20 cm and
impacts the estimated concentrations on home-grown vegetables and fruit. The "untilled"
condition distributes the eroded contaminants only to a depth of 1 cm, and results in a soil
concentration for which soil exposure pathways, ingestion and dermal contact, are
estimated. Note that there are no differences in concentrations at the exposure site
among the three example contaminants. It is assumed that eroded soil contains
concentrations identical to the concentrations in soil at the landfill site. Therefore, the
same amounts of contaminants erode onto and contaminate the soil on the exposure site
properties to the same levels.
Scenarios 4 and 5
Chapter 6 described the application of the ISC model to estimate air-borne
concentrations and deposition rates of the contaminants at the residence and farm. Wet
plus dry particle deposition rates, in units of g/m2-yr, were determined for all dioxin-like
compounds, for two pollution control technologies, and at various distances from the
stack. Results from this exercise can be found on Tables 6-18 and 6-19 in Chapter 6.
The properties of Scenarios 4 and 5 are located 500 and 2000 meters, respectively, from
the incinerator. Assuming 500 meters is the midpoint of the larger property, and that
property is square-shaped at 200 meters per side, then it begins roughly 400 meters from
the stack and ends roughly 600 meters away. Strictly speaking, the deposition rate for
modeling can be estimated as the average of several rates between 400 and 600 meters.
For demonstration purposes, the deposition rates at 500 meters as listed in Tables 6-18
and 6-19 were used. Finally, the emission control technology assumed for the
hypothetical incinerator in these scenarios was the fabric filter combined with semi-dry
alkaline scrubbers (abbreviated DSFF in Chapter 6). Deposition rates for this technology
9-14 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
are given in Table 6-19, and for the three compounds at 500 meters are: 1,32*10~9 g/m2-
yr (2,3,7,8-TCDD), 1.42* 10^ g/m2-yr (2,3,4,7,8-PCDF), and 4.72MO'9 g/m2-yr
(2,3,3',4,4',5,5'-HPCB). At 2000 meters, the deposition rates are: 8.08MO"10 g/m2-yr
(2,3,7,8-TCDD), 9.45MO'9 g/m2-yr (2,3,4,7,8-PCDF), and 3.15MO'9 g/m2-yr
(2,3,3',4,4',5,5'-HPCB). The soil concentrations resulting from these depositions are
listed in Table 9-1.
Scenario 6
Concentrations on combined (fly + bottom) ash can be estimated using information
provided in Chapter 6. Chapter 6 indicates that negligible amounts of PCDDs and PCDFs
have been found on bottom ash. This is consistent with data on fly, bottom, and
combined ash concentrations of PCDDs, PCDFs, and PCBs summarized in EPA (1991b).
For purposes of this demonstration, it will be assumed that bottom ash contains 1/10 the
concentrations assumed for fly ash. Chapter 6 also indicates that the mass of bottom ash
exceeds that of fly ash by about a factor of ten, and estimates 273 tons of bottom ash
generated daily given a solid waste load of 2727 tons daily. Tables 6-9 through 6-12 list
ash emission factors for PCDDs, PCDFs, and PCBs, for the two control technologies.
These factors are given in units of gms of contaminant in the fly ash per metric ton waste
incinerated. Since the hypothetical incinerator handles 2727 metric tons/day, the total
daily contaminant in fly ash is estimated as these emission factors times 2727. Also
discussed in Chapter 6 is the estimation of 30 metric tons fly ash generated per day with
this load of solid waste, and using the fabric filter combined with semi-dry alkaline
scrubbers. Tables 6-11 and 6-12 provide the emission factors for this technology, and
now concentrations can be demonstrated. From Table 6-11, the emission factor for
2,3,7,8-TCDD is 8MO'5 g/MT solid waste, and the ash concentration of 2,3,7,8-TCDD is
(using conversion factors 1000 kg/MT and 1000/yg/mg):
(8 x 10 5 g/MT) (2727 MT) (106 ng/g) = _ ..
(30 MT) (1000 kg/MT) My/ *
The concentrations for 2,3,4,7,8-PCDF and 2,3,3',4,4',5,5'-HPCB are 86/vg/kg and 23
9-15 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
/yg/kg on fly ash, respectively. Combined ash concentration can then be estimated given
273 MT bottom ash at 1/10 the concentration just estimated:
[(7 fig/kg) (30 Mf)] +[(0.7Mg/£g) (273 MT)] _
-
30 MT + 273 MT
for 2,3,7,8-TCDD, and 16.3 and 4.3/yg/kg for 2,3,4,7,8-PCDF and 2,3,3',4/4',5,5'-HPCB,
respectively. The final assumption made is that the ash mixes with equal amounts of soil
at the landfill, so that landfill concentrations are half these concentrations at 0.7, 8.2 and
2.2 /;g/kg. The concentrations of contaminants at the farm for the untilled condition are:
0.56/yg/kg for 2,3,7,8-TCDD, 6.59/yg/kg for 2,3,4,7,8-PCDF, and 1.77/yg/kg for
2,3,3',4,4',5,5'-HPCB.
9.6. RESULTS
The results of this exercise include the exposure media concentrations for all
exposure pathways and settings, and the LADD exposure estimates. These two
categories of results are summarized in Tables 9-2 and 9-3. Following now are several
observations from this exercise. It is important to understand that these observations are
not generalizable comments. Different results would arise from different source strength
characteristics, proximity considerations, model parameter values, different models
altogether, and so on. Chapter 10 on uncertainty describes many areas of this assessment
which should be considered when evaluating the methodology or viewing the results.
9.6.1. Observations Concerning Exposure Media Concentrations
• Soil Concentrations:
The background soil concentrations at the site of exposure assumed for Scenarios 1
and 2, 0.001 /yg/kg (1 ppt) were not very different from concentrations estimated at the
sites of exposure 500 and 2000 meters from the hypothetical incinerator, 0.0008-0.014
/yg/kg (.8-14 ppt) for Scenarios 4 and 5. Higher soil concentrations at the site of exposure
resulted from erosion of contaminated soil originating at the 10-acre abandoned industrial
site, Scenario 3, and at the 67-acre ash landfill, Scenario 6. Concentrations at the sites of
exposure were 0.613/yg/kg (613 ppt) for Scenario 3 (off-site soil with 1 /yg/kg soil
9-16 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 9-2. Exposure media concentrations estimated for all scenarios and pathways
Exposure pathway/
scenario
2,3,7,8-
TCDD
2,3,4,7,8-
PCDF
2,3,3',4,4',5,5'-
HPCB
Concentration of contaminants in soil
for soil ingestion and dermal contact
pathways, ywg/kg (ppb)
#1 On-site soil, central
#2 On-site soil, high end
#3 Off-site soil, high end
#4 Stack emissions, central
#5 Stack emissions, high end
#6 Ash landfill, high end
Concentration of contaminants in air
in vapor phase for vapor inhalation
pathway, //g/m3
#1 On-site soil, central
#2 On-site soil, high end
#3 Off-site soil, high end
#4 Stack emissions, central
#5 Stack emissions, high end
#6 Ash landfill, high end
Concentration of contaminants in air
in particulate phase for particulate
inhalation pathway, /vg/m3
#1 On-site soil, central
#2 On-site soil, high end
#3 Off-site soil, high end
#6 Ash landfill, high end
Concentration of contaminants in water
for water ingestion pathway, mg/L
#1 On-site soil, central
#2 On-site soil, high end
#3 Off-site soil, high end
#4 Stack emissions, central
#5 Stack emissions, high end
#6 Ash landfill, high end
0.001
0.001
0.613
0.0008
0.0013
0.563
0.001
0.001
0.613
0.009
0.014
6.590
0.001
0.001
0.613
0.003
0.005
1.770
4*10'11
4*10'9
4*10'11
6*10'9
6MO'13
5*10'12
2MO-10
6*10-io
4MO'12
4*1Q-12
7MQ-11
9*10-14
-14
-10
9*10
3*10
2*10
-11
2*10
4*10
7*10
3*10
-9
-10
-10
-8
6*1Q-13
5*10'12
2*10-10
6*10'9
3*10'12
3*10'12
5*10'11
6*1Q-13
-13
-9
6*10
2*10
1*10-10
1*10-10
2MQ-8
1MO
-10
8 * 1 0
'8
6*1Q-13
5*1Q-12
2*10-10
2*10'9
6*10
6MO
-13
'13
4*10-14
4*10-14
1MO-10
(continued on the following page)
9-17
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 9-2. (continued)
Exposure pathway/
scenario
2,3,7,8-
TCDD
2,3,4,7,8-
PCDF
2,3,3',4,4',5,5'-
HPCB
5. Concentration of contaminants in
fish for fish ingestion pathway,
mg/kg
#1 On-site soil, central
#2 On-site soil, high end
#3 Off-site soil, high end
#4 Stack emissions, central
#5 Stack emissions, high end
#6 Ash landfill, high end
6. Concentration of contaminants in below
ground vegetables (no below ground fruit
was assumed) for their respective
pathways, mg/kg fresh weight
#1 On-site soil, central
#2 On-site soil, high end
#3 Off-site soil, high end
#4 Stack emissions, central
#5 Stack emissions, high end
#6 Ash landfill, high end
7. Concentration of contaminants in above
ground fruit and vegetables for their
respective pathways, mg/kg fresh weight
#1 On-site soil, central
#2 On-site soil, high end
#3 Off-site soil, high end
#4 Stack emissions, central
#5 Stack emissions, high end
#6 Ash landfill, high end
8. Concentration of contaminants in
beef fat for beef ingestion
pathway, mg/kg dry weight
#2 On-site soil, high end
#3 Off-site soil, high end
#5 Stack emissions, high end
#6 Ash landfill, high end
7 * 1CT8
7*10'8
1 * 1 cr6
2.10-9
2*icr9
6*10
-4
4*10-11
4*icr11
4*icr9
3*icr9
5*1Q-9
6*icr9
4*io~7
2*io-4
1 * 10'6
2*10'4
9 * 1 cr8
9*1Q-8
2MO-6
2*10'8
2*icr8
1 * 1 or9
1*10-9
4*10'8
6*icr11
9*icr11
4*icr8
1*1Q-9
i*icr9
4*icr8
6MO-io
9»10-io
4MCT7
9*10
9*10
11
-11
9*10'9
4*10'8
5*10
8
2*10'7
1 * 10-4
1 *10"5
1 * 10'3
3*10'6
4*10'5
2*10'7
2*10'7
6*10-4
8*10
8*10
3*10
1*10
2*10
7*10
-10
-10
-8
-10
-10
-8
2*1Q-11
2MO-11
3*10"9
2*10'8
2*10
1*10
2*10"5
4*1Q-7
9-18
(continued on the following page)
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 9-2. (continued)
Exposure pathway/
scenario
2,3,7,8-
TCDD
2,3,4,7,8-
PCDF
2,3,3',4,4',5,5'-
HPCB
Concentration of contaminants in
milk fat for milk ingestion pathway,
mg/kg dry weight
#2 On-site soil, high end 1 * 1CT7 1 * 10'7 9 * 1 (T9
#3 Off-site soil, high end 5*1(T5 3*10'5 4*1(T6
#5 Stack emissions, high end 8MO'7 8*1CT6 3*1CT7
#6 Ash landfill, high end 5MCT5 4*1CT4 IMO'5
9-19 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 9-3. Lifetime average daily dose, LADD, estimates for all scenarios and exposure
pathways (all results in mg/kg-day)
Scenario description
exposure pathway
2,3,7,8- 2,3,4,7,8- 2,3,3',4,4',5,5'-
TCDD PCDF HPCB
#1 On-site soil contamination
Central exposure scenario
a. Soil ingestion
b. Soil dermal contact
c. Inhalation-vapor
d. Inhalation-particle
e. Water ingestion
f. Fish ingestion
g. Fruit ingestion
h. Vegetable ingestion
#2 On-site soil contamination
High end exposure scenario
a. Soil ingestion
b. Soil dermal contact
c. Inhalation-vapor
d. Inhalation-particle
e. Water ingestion
f. Fish ingestion
g. Fruit ingestion
h. Vegetable ingestion
i. Beef ingestion
j. Milk ingestion
#3 Off-site soil contamination
High end exposure scenario
a. Soil ingestion
b. Soil dermal contact
c. Inhalation-vapor
d. Inhalation-particle
e. Water ingestion
f. Fish ingestion
g. Fruit ingestion
h. Vegetable ingestion
8*10~13
4*1Q-14
1MO-15
2MO-17
8*1Q-15
2MO-13
1 *10"15
2MO-14
8MO-13
4*10-14
5*1Q-16
2MO-17
5*10-15
2»10-13
3*1Q-15
2MO-14
8«10-13
4. 10-14
3*10'15
2MO-17
1 *10~15
6*10'12
5*10"16
1MO-14
3MQ-12
9MO-13
3MO-16
2*10-14
4*10'15
2MO-12
2*10-9
3MCT11
3*icr13
2*10-14
4*10'13
2MCT11
4*10-13
3MCT12
3*10"12
g.10-13
3MQ-16
2MO-12
7MO-14
2MQ-12
2*10"9
3*10-11
2MQ-14
2MQ-13
1MO-12
9*1Q-13
7»10-15
3MO-16
3*10'15
2MO-15
1*10-12
2*10-13
2*10'9
3*1Q-11
1MO-12
5*10-14
7*10-10
3*10"13
2*10-12
(continued on the following page)
9-20
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 9-3. (continued)
Scenario description
exposure pathway
#3 Off-site soil contamination
High end exposure scenario
(continued)
i. Beef ingestion
j. Milk ingestion
#4 Stack emissions
Central exposure scenario
a. Soil ingestion
b. Soil dermal contact
c. Inhalation-vapor
d. Water ingestion
e. Fish ingestion
f. Fruit ingestion
g. Vegetable ingestion
#5 Stack emissions
High end exposure scenario
a. Soil ingestion
b. Soil dermal contact
c. Inhalation-vapor
d. Water ingestion
e. Fish ingestion
f. Fruit ingestion
g. Vegetable ingestion
h. Beef ingestion
i. Milk ingestion
#6 Ash landfill
High end exposure scenario
a. Soil ingestion
b. Soil dermal contact
c. Inhalation-vapor
d. Inhalation-particle
2,3,7,8-
TCDD
7*10'9
8MO-10
7*10-13
4*10-14
1MO-15
2*10-16
3MO-15
1 *10~13
1MO-13
4*10'12
5*10-14
5*1Q-15
5*10"16
3*10-14
5*10'13
6*10'13
5*10'11
1*10-11
2*10-9
2*10-11
5*10'13
4MO-14
2,3,4,7,8-
PCDF
4*10'9
6*10-10
8*1Q-12
4*1Q-13
1*10-14
1*10'15
5*10-14
1MO-12
1MO-12
5*1Q-11
6*1Q-13
5*10-14
3MO-15
3*10-13
6*10'12
7*10'12
5*10-10
1MO-10
2*10-8
3MO-10
2,10-12
5*10'13
2,3,3',4,4',5,5'-
HPCB
7MO-10
8*1Q-11
3*1Q-12
1MO-13
4*10'15
4MO-13
5*1Q-13
6*10'13
2MO-11
2MO-13
2*10-14
2*10-16
3*1Q-12
2*1Q-12
2MO-12
2*10"11
4*10-12
6*10-9
8*1 0'11
6*10'12
1MO-13
(continued on the following page)
9-21
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 9-3. (continued)
Scenario description
exposure pathway
2,3,7,8-
TCDD
2,3,4,7,8-
PCDF
2,3,3',4,4',5,5'-
HPCB
#6 Ash landfill
High end exposure scenario
(continued)
e.
f.
g.
h.
i.
j-
Water ingestion
Fish ingestion
Fruit ingestion
Vegetable ingestion
Beef ingestion
Milk ingestion
2»10-12
9MO'11
6MQ-13
3*1Q-12
6MO'9
8«10-10
1*10-"
1 * 1 0'9
2»10-12
4MO'11
5 * 1 0'8
7*10'9
7*
1 *
1*
5*
2*
2*
10-13
10'8
1C'12
10"12
io-9
10-10
concentrations) and 0.563 to 6.590/yg/kg (563 to 6590 ppt) for Scenario 6 (ash landfill
with 0.7-8.1 /;g/kg concentrations). Exposure site soil concentrations for Scenario 3 were
61 % of what they were at the off-site area of soil contamination, and for the ash landfill,
were 80% of what they were at the ash landfill for Scenario 6. The distance from the off-
site location of soil contamination was the same in these two scenarios at 150 meters; the
ratios were higher for the landfill scenarios because of significantly higher amounts of
contaminated eroded soil at the 67-acre landfill compared to the 10-acre site. The soil
concentrations were uniform for all three compounds within Scenarios 1, 2, and 3. For
Scenarios 1 and 2, soil concentrations were initialized at 0.001 /yg/kg. For Scenario 3,
exposure site soil concentrations resulting from erosion were also the same for all three
compounds because concentrations at the site of contamination were uniformly set at 1.0
//g/kg. The process of erosion and mixing with exposure site soil was not a function of
chemical properties. For the other four scenarios, different exposure site soil
concentrations resulted because stacks emitted different quantities of compounds
(Scenarios 4 and 5), and because concentrations at the ash landfill were different for each
compound (Scenario 6). One final note is that the soil concentrations listed for Scenarios
3 through 6 are those that result from the assumption of a 1-cm mixing depth, and are the
concentrations used for dermal and soil ingestion exposure estimates. A second soil
concentration, the tilled concentrations, is 20 times less than those listed for Scenarios 3-
9-22
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
6, and is used to estimated concentrations of underground vegetables.
• Vapor-Phase Air Concentrations:
One statement to make up front about vapor-phase air concentrations is that using
the descriptor "vapor-phase" does not necessarily mean that the contaminants are
expected to remain in a pure vapor state while air-borne. As discussed in Chapter 6, the
hypothesis is that contaminants sorb to aerosol-size particles, particles < 1 //m. Because
the atmospheric residence time is in the order of days for aerosol-size particles,
contaminants sorbed to aerosols are expected to remain air-borne for inhalation exposures.
Residues which volatilize from the soil are expected to initially be a vapor phase.
However, it has also been theorized that dioxin-like compounds released into the air this
way would not remain in vapor phase, but would also sorb to air-borne particulates,
aerosol size or otherwise. The term vapor-phase is used more to distinguish this reservoir
of air-borne contaminants from the one directly linked to the suspension of particulates.
Processes which suspend particulates include wind erosion, ash management at the ash
landfill, and the release of stack emissions.
Concentrations of contaminants in the vapor phase range from 10~11 to 10~8/yg/m3.
Similar and lower concentrations, 10~11 to 10"10/yg/m3, resulted from the volatilization of
background soil concentrations of 0.001 //g/kg, Scenarios 1 and 2, and the transport of
vapor-phase contaminants originating from the incinerator stack 500 and 2000 meters
away. Scenarios 4 and 5. Like the trend for soil concentrations above, volatilization from
off-site and more extensive soil contamination, followed by transport 1 50 meters away to
a site of exposure, resulted in higher exposure media concentrations, 10'8 to 10'9 //g/m3
air concentrations for Scenarios 3 and 6. One interesting trend of note is that the vapor-
phase concentrations for the central and high end scenarios of Scenarios 1 and 2 are
similar for each compound; i.e., the 2,3,7,8-TCDD concentration for Scenario 1 is the
same as the 2,3,7,8-TCDD concentration of Scenario 2 (although they are different within
a Scenario for different compounds; that will be discussed shortly). This is, in fact, the
result of two inverse trends of the solution algorithm. First, the average volatilization flux
(mass/area-time) will always be lower for the high end scenario as compared to the central
scenario. This is due to the solution algorithm assumption that residues available for
volatilization originate from deeper in the soil profile over time, so that the average flux is
9-23 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
lower for longer periods of volatilization. This is seen in the volatilization flux equation -
Equation (5-12), Chapter 5 - which has a time term (ED, or exposure duration) in the
denominator. The high end scenarios assume 20 years exposure duration compared to 9
years for the central scenarios. This alone would have resulted in lower air concentrations
in the high end as compared to the central scenario. However, the dispersion of volatilized
residues is a direct function of the area over which volatilization occurs. This is expressed
in terms of a side length, parameter "a" in Equation (5-15), Chapter 5, which is in the
numerator. This by itself would have resulted in higher air concentrations at the larger
farm site of the high end scenario, 10 acres, as compared to the smaller residence of the
central scenario, 1 acre. The two trends cancel each other and vapor phase
concentrations for a given compound are similar for both scenarios. However, for
different compounds within the same scenario, vapor phase concentrations are different.
This difference is due to chemical parameters, principally the Henry's Constant, H.
2,3,3',4,4',5,5'-HPCB had the highest value for H, and it was 2 orders of magnitude
higher than the value for 2,3,7,8-TCDD and 3 orders of magnitude higher than the value
for 2,3,4,7,8-PCDF. This drove the trend for air concentrations, as 2,3,3',4,4',5,5'-HPCB
had the highest air concentrations, followed by 2,3,7,8-TCDD at a concentration 1 order
of magnitude lower and 2,3,4,7,8-PCDF at slightly lower than 2,3,7,8-TCDD.
• Particulate-Phase Air Concentrations:
In all cases, particulate-phase air-borne concentrations of contaminants were 1 to 3
orders of magnitude lower than vapor-phase concentrations for the same scenario.
Concentrations resulting from wind erosion of background soil concentrations. Scenarios 1
and 2, were 2 to 3 orders of magnitude lower than those which originate 1 50 meters
away at the off-site locations of contaminated soil, Scenario 3 and 6. This was due
principally to 3-4 orders of magnitude higher soil concentrations at these off-site soil
contamination locations. Another trend is that the concentrations are the same for all
three compounds within Scenarios 1-3. This is because the algorithm to estimate air-
borne concentrations of particulate phase contaminants is independent of chemical
properties. The concentrations are different for the different compounds in the ash landfill
scenario, 6, because of different initial concentrations in the landfill soil. The trend
discussion above concerning vapor phase concentrations resulting from volatilization and
9-24 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
dispersion is not true for particulate phase estimation. In this case, a steady flux is
estimated which is not a function of time. The same dispersion algorithm is used,
however, so that the high end concentrations in Scenario 2 are higher than the central
concentrations in Scenario 1. One final observation is that concentrations for the ash
landfill scenario, 6, are 4 to 10 times higher than those from the off-site scenario, 3. Part
of this is due to different initial soil concentrations, but part is also due to more sources of
particulate emissions at the ash landfill. Sources of particulate emissions at the ash landfill
not assumed for the off-site soil contamination include: vehicular resuspension of
contaminated roadway dust, emissions off trucks in transit, unloading, and spreading and
compacting. An evaluation of the ash landfill flux contributions from the various sources
show that the flux due to these ash management activities was twice the flux due to wind
erosion alone. (Chapter 6 discussed the use of the ISC model to estimate exposure site
concentrations, and it was noted there that air-borne particulate-phase contaminants are
not assumed to result from stack emissions.)
• Drinking Water and Fish:
Concentrations of the example contaminants in water were 10~14 to 10~9 mg/L
(ppm). Concentrations in fish ranged from 10~9 to 10~4 mg/kg. One trend of note is that
fish concentrations of the 2,3,3',4,4',5,5'-HPCB are 1-2 orders of magnitude higher than
2,3,7,8-TCDD or 2,3,4,7,8-PCDF. This is because the key variable estimating fish tissue
concentrations, the Biota Sediment Accumulation Factor, BSAF, is 2.0 for the example
PCB while it is 0.09 for the example dioxin and furan. Another noteworthy trend is that
the water and fish concentrations are the same for each compound for the pair of
scenarios, #4 and #5, which demonstrate the stack emission source category. The key
difference between the central and high end example scenarios is the distance from the
stack to the site of exposure - it is 500 m for the high end and 2000 for the central
scenario. This impacts all exposure media estimations except the fish and water
estimates. Those two are a function of average watershed impact to the stack emissions,
not impact to the site of exposure. All other exposure media concentrations for the stack
emission source category are a function of specific impact to a site, which will differ as a
function of distance from the stack.
9-25 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
• Fruit and Vegetable Concentrations:
Concentrations in these foods ranged from 10~11 to 10~7 mg/kg (ppm) expressed on
a fresh weight basis. Concentrations in below ground vegetables are found to exceed
those in above ground vegetables when the source of contamination is soil - the on-site,
off-site, and ash landfill scenarios of this assessment. When the source of contamination
is stack emissions, above ground concentrations exceed those of below ground. The main
cause for this trend was that particle depositions have a large impact on above ground
vegetation for the stack emission source category, whereas particle depositions have a
small impact on above ground vegetation for the other three soil source categories (this is
further explained in Chapter 10, Section 10.2.11.3). As in the air and soil trends
discussed above, off-site soil contamination in the range of 1 yug/kg (1 ppb; example
Scenarios #3 and #6) results in higher concentrations than on-site background soil
concentrations of 0.001 jug/kg (1 ppt; example Scenarios #1 and #2). Another trend
noted for Scenarios 1-3, where the initial soil concentrations were the same among the
three compounds, is that transfers from soil to plant are driven by chemical parameters,
particularly the octanol water partition coefficient, Kow. 2,3,3',4,4',5,5'-HPCB had the
highest Kow, with 2,3,4,7,8-PCDF and 2,3,7,8-TCDD at lower but similar Kow. Higher
Kow translates to tighter sorption to soil, and less transfer to plant, either through root
uptake or air-to-leaf transfer. This trend translated to the lowest fruit/vegetable
concentrations for 2,3,3',4,4',5,5'-HPCB. 2,3,7,8-TCDD and 2,3,4,7,8-PCDF had similar
fruit/vegetable concentrations for Scenarios 1-3.
• Beef and Milk Concentrations:
These concentrations were comparable to those of fish and ranged from 10~9 to
10~3 mg/kg. Milk concentrations were lower than beef concentrations in all cases. This
was due to assumptions concerning apportioning of total dry matter intake between
contaminated soil, contaminated pasture grass, and home-grown contaminated fodder.
Beef cattle were assumed to take in 4 times as much soil as dairy cattle, 8% of their dry
matter intake versus 2%. Another observation that can be made is similar to the
observation concerning the key biotransfer term for fish, the Biota Sediment Accumulation
Factor, BSAF. In that case, it was noted that the BSAF for the PCB example compound
was much higher than the BSAF for either the dioxin or the furan - hence higher fish
9-26 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
concentrations result. In this case, however, the PCB example compound had the lowest
beef/milk biotransfer factor, F. The F for the dioxin and furan example compound was 5.0
and 3.0, respectively, while the PCB F was 0.5. This resulted in uniformly lower PCB beef
and milk concentrations.
9.6.2. Observations Concerning LADD Exposure Estimates
Much of the differences between exposure pathways and scenarios is due to
differences in exposure media estimation. Therefore, much of the above discussion is also
appropriate for trend analysis of Lifetime Average Daily Dose, LADD, estimates. What will
be noted below are unique observations:
• General:
LADD estimates ranged from 10"17 mg/kg-day to 10~8 mg/kg-day. The highest
exposures were associated with the ash landfill and the off-site soil contamination
scenarios. Scenarios #3 and #6. As discussed above, these scenarios had the highest
exposure media concentrations for all exposure media. LADDs for Scenarios 3 and 6
ranged from 10~14 to 10~8 mg/kg-day. Exposures associated with stack emissions,
Scenarios #4 and #5, and on-site soil contamination with low soil concentration. Scenarios
#1 and #2, were fairly similar with a range of 10~17 to 10~10 mg/kg-day.
• "Central" vs. "High End":
Differences between analogous "central" and "high end" exposures were, in all
cases, near or less than an order of magnitude (inhalation exposure for the central on-site
scenario and the inhalation exposure for high end on-site scenario are analogous
exposures). This is because the exposure parameters used to distinguish typical and high
end exposures, the contact rates, contact fractions, and exposure durations, themselves
did not differ significantly. In the stack emission scenario, placing exposed individuals
either 500 or 2000 meters away from the incinerator did not significantly impact the
results. The high end scenarios were modeled after a rural farm and did have exposures
from home grown beef and milk food products. The central scenarios were modeled after
a non-farming rural residence, and did not have beef and milk exposure pathways. Since
beef and milk exposure pathways were noted as the highest exposures, along with
9-27 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
ingestion of fish (see next bullet), it would be appropriate to conclude that farming families
ingesting a portion of their home produced beef and milk are more exposed than non-
farming families without these exposures.
• Exposure pathway analyses:
It is inappropriate to compare and rank exposure pathways across all scenarios
because the source terms are different. However, relationships between different
pathways within each scenario can be discussed. Table 9.4 was constructed by summing
the LADDs for all pathways, and then determining the percent contribution by each
pathway. Before the summation, LADDs were corrected to account for absorption - all
ingestion LADDs assumed 50% absorption and inhalation LADDs assumed 75%. The
dermal contact LADD was the only one where absorption was already considered in its
estimation: absorbed dose was estimated as 3% of dose contacting the body. Also, this
exercise assumes all pathways occur simultaneously, and so on. Table 9.4. was
generated only for the 2,3,7,8-TCDD example compound, and the rows are listed generally
from the highest to lowest percentage contribution. The following observations are made
from that table:
• In high end scenarios which assumed exposure to home grown beef and milk -
Scenarios 2, 3, 5, 6 - the beef exposures dominated the results.
• Milk exposures were lower than beef exposures because of less milk fat
ingestion (10.5 g/day milk fat vs. 22 g/day beef fat) and lower concentrations in milk as
compared to beef (this was discussed above). The general dominance of beef and milk
exposures underscore the importance of food chain exposures.
• Soil ingestion exposures dominated the results when beef exposures were not
considered, and was generally the second highest exposure when beef was considered.
Dermal exposures were non-trivial. These demonstrations also show the importance of
direct soil exposures
• Fish exposures were critical when exposure site soil concentrations matched
concentrations throughout the watershed, which was true for example Scenarios 1 and 2.
Proximity to the site of high soil contamination, the off-site example Scenario 2 and the
ash landfill Scenario 2, led to high beef, milk, and soil related exposures which
overshadowed fish exposures. The particular incinerator modeled for example Scenarios 4
9-28 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 9.4. Percent contribution of the different exposure pathways within each exposure
Exposure Pathway
Meat Ingestion
Soil Ingestion
Milk Ingestion
Fish Ingestion
Soil Dermal
Vegetable Ingestion
Fruit Ingestion
Water Ingestion
Vapor Inhalation
Particle Inhalation
1
NA
75
NA
14
8
2
0
1
0
0
2
63
15
9
5
7
4
0
0
0
0
Scenario #
3 4
70
21
8
0
1
0
0
0
0
0
NA
71
NA
1
7
11
10
0
0
NA
5
73
6
19
0
0
1
1
0
0
NA
6
69
21
8
1
1
0
0
0
0
0
* Assumes exposed individual experiences all relevant pathways and exposures are
additive; see text for further explanation.
and 5 apparently had a relatively low impact to the water body in comparison to exposures
at the site of residence or farming. This led to marginal water and fish exposures for the
incinerator scenarios.
• Fruit and vegetable exposures were noteworthy only in the stack emission
source category where individuals had home gardens but did not raise cattle.
• Water ingestion, vapor and particle inhalation exposures were very low in
comparison to the other exposures in these scenarios.
The LADD estimates of all example scenarios were derived assuming a limited
duration of exposure to the dioxin-like compounds, and also limited contact with exposure
media. A pattern of childhood soil ingestion was assumed to occur over a five-year period.
The central scenarios assumed a nine-year duration of exposure to the contamination, and
the high end scenarios assumed a twenty-year exposure period. The contact with
impacted media was only assumed to occur in the home environment - only a portion of an
individual's meat, milk, water, and fruit and vegetable ingestion was evaluated. This is
only one approach to scenario development; other approaches might consider the quality
of exposure media not associated with the home environment. For example, if the bulk of
an individual's ingestion of produce comes from local farms, and local farms may be
9-29 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
impacted by an incinerator, then perhaps 90-100% of an individual's fruit and vegetable
ingestion, rather than the 20-40% assumed in this assessment, should be considered
impacted.
Another issue to consider while interpreting these scenarios is their relation to
background exposures. As discussed in'Chapter 3, dioxin-like compounds are commonly
found throughout the environment, leading to "background" exposures even in situations
where known sources are not present. Since these scenarios are developed around
defined sources, they could be interpreted as incremental exposures beyond "background"
levels. This interpretation is less satisfying for Scenarios 1 and 2 where the ambient soil is
assumed to be contaminated at 1 ng/kg (ppt) level which, as discussed in Chapter 3 and
earlier in this Chapter in Section 9.5, may be close to a background level. In this sense.
Scenarios 1 and 2 are more representative of background exposures than incremental
exposures. However, diet fractions were applied indicating that only a portion of the food
supply was contaminated. Also, the exposure duration was defined as less than a lifetime.
For purposes of a true background exposure analysis, it would be more appropriate to
assume 100% for diet fractions and lifetime exposures for example Scenarios 1 and 2.
An exercise was undertaken making the following changes in example Scenarios 1
and 2: 70 years exposure duration was assumed for all pathways except the childhood
pattern of soil ingestion (which remained at 5 years), and all contact fractions were set
equal to 1.00. All contact rates were unchanged for this exercise. The one contact rate
which might be changed is the rate of fish ingestion. For that pathway, the 1.2 (central)
and 4.1 (high end) g/day rates were estimated assuming a number of meals and meal sizes
that a rural individual would recreationally obtain from a nearby impacted water body. For
background impacts, it may be more appropriate to have an average ingestion rate, and
assume that the 0.63 pg/g fish concentration estimated for example Scenarios #1 and #2
is typical of all fish ingested. That was not done for the results of this exercise that are
displayed in Table 9.5, but was done separately and will be discussed in the bullets below.
That table includes LADDs and the percent contribution from each pathway for the original
and the amended example Scenarios #1 and #2, and only for 2,3,7,8-TCDD. Observations
from this exercise are:
• All exposures increased except soil ingestion. Mostly the increase in LADD
were within an order of magnitude. The LADDs are still about an order of magnitude
9-30 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 9.5. Exposures to low soil concentrations of 2,3,7,8-TCDD assuming lifetime
exposure durations and unlimited contact with impacted media, compared with exposures
assuming limited durations and limited contact.
Pathway
Limited Exposures
LADD, mg/kg-day Percent
Lifetime Exposures
LADD, mg/kg-day Percent
Example Scenario #1
On-site soil contamination
Concentration = 1.0 ng/kg
Central exposure scenario
Soil ingestion
Soil dermal contact
Inhalation-vapor
Inhalation-particle
Water ingestion
Fish ingestion
Fruit ingestion
Vegetable ingestion
8»10-13
4*10-14
1*10-15
2*10-17
8»10-15
2»10-13
1MO-15
2*10-14
75
8
0
0
1
14
0
2
8MO-13
3*icr13
4*10-15
2MQ-16
9»icr14
i*icr12
2*1Cr14
1MO-12
24
20
0
0
2
37
0
17
Example Scenario #2
On-site soil contamination
Concentration = 1 ng/kg
High end exposure scenario
Soil ingestion
Soil dermal contact
Inhalation-vapor
Inhalation-particle
Water ingestion
Fish ingestion
Fruit ingestion
Vegetable ingestion
Beef ingestion
Milk ingestion
3MQ-12
9*10'13
3*1CT15
3*10-16
2*10-14
1*1(r12
4»10-15
7"10-14
Tier11
2*icr12
15
7
0
0
0
6
0
0
63
9
3*i(r12
3*icr12
7MQ-15
1MQ-15
4MO-14
4*1(r11
3*10-14
6MQ-13
1MO-10
1MO-11
3
5
0
0
0
3
0
0
77
12
9-31
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
lower than limited exposures estimated to occur from living near an area of high soil
contamination, such as ash landfill, as can be seen from comparing these lifetime results
with those in Table 9.3. for Example Scenarios #3 (off-site source category) and #6 (ash
landfill source category).
• The increases to vapor phase inhalation exposures were small relative to other
increases. This occurs because average volatilization flux decreases as a function of time -
average fluxes over long periods of time are lower than average fluxes over short periods
of time (see Section 5.3.2 for further explanation). This has a rippling effect on fruit and
vegetable concentrations as well as beef and milk concentrations. A trend of decreasing
volatilization flux is realistic when highly contaminated soils become depleted over time.
This was a key principal of the volatilization flux algorithm. However, this is probably not
a realistic trend for ubiquitous concentrations which are likely to be replenished in soil over
time. Sensitivity analysis in Chapter 10 (Section 10.2.9.4) shows that increasing
exposure duration from 20 to 70 years decreases average flux by a factor of 2. By
implication, vapor inhalation, fruit and vegetable ingestion, and beef and milk ingestion, are
all underestimated for a lifetime of exposure for example Scenario #2, by around the same
factor of 2.
• The relative impact of soil ingestion dropped when assuming lifetime exposures.
It is interesting that direct soil impacts, a childhood pattern of soil ingestion and a lifetime
of soil dermal contact, still are noteworthy impacts, particularly when not considering beef
or milk exposures.
• Fish ingestion exposures became the predominant lifetime exposures when beef
and milk were not considered. Interestingly, home grown vegetable exposures also gained
in significance when considering a lifetime of exposure, being 17% of total exposures
when beef and milk were not considered.
• Different ingestion rates of fish were evaluated for example Scenario #2. The
rate of 4.1 g/day was increased to 6.5, 30, and 140 g/day. The 6.5 g/day was used in
the Ambient Water Quality Criteria document for 2,3,7,8-TCDD, and was described as an
average daily per capita consumption of freshwater and estuarine fish and shellfish (EPA,
1984). The 30 and 140 g/day were recommended as 50th and 90th ingestion rates in
EPA (1989) for recreational fisherman in an area where there is a large water body present
and widespread contamination is evident, and where site-specific information is
9-32 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
unavailable. Increasing to 6.5, 30, and 140 g/day increased 70-year exposure LADDs to
7*10~12, 3*10~11, and 1 *10~10, respectively. The percent of total exposure in this test
increased from 3% to 5, 20, and 54%, respectively.
An evaluation of total human exposure, and the relative impact of different
exposure pathways for 2,3,7,8-TCDD, was also undertaken by Travis and Hattemer-Frey
(1991). Their approach was based on modeling, using their Fugacity Food Chain model.
This model requires as input emission rates into air, soil, and water; these emission rates
are then transformed into concentrations. Those concentrations are used to estimate food
crop concentrations, beef, milk, and fish concentrations. In the application of this model
by Travis and Hattemer-Frey (1991), emission rates were calibrated to arrive at air, soil,
and water concentrations that were supported by the literature. Resulting concentrations
in these three media were: 0.02 pg/m3 in air partitioned as 0.016 pg/m3 in particulate
form and 0.004 pg/m3 in vapor form, 0.96 pg/g in soil, and 0.003 pg/L in water. Using
the model to estimate exposure media concentrations, they then assumed contact rates as
given by Yang and Nelson (1986) to estimate human exposure. Assuming 100%
absorption, they concluded that typical human exposure is on the order of 35 pg/day
2,3,7,8-TCDD.
Their analysis will be compared to the analysis of this methodology. The example
scenario most like their's was example Scenario #2, which estimated exposures to soil
levels of 1.0 pg/g. The assumption used in the above sensitivity exercise of 100%
contact fractions rather than partial contact is also the appropriate assumption for
comparison. A key difference to keep in mind is that air and water impacts are estimated
given soil concentrations in this methodology, whereas air and water impacts are specified
with their model. This comparison is given in Table 9.6 below.
While the exposure of 9.5 pg/day estimated in Scenario 2 (assuming 100% contact
fractions) is lower than the 34.3 pg/day estimated by Travis and Hattemer-Frey, there are
a few critical differences in the two approaches. One was in the contact rates. However,
replacing the contact rates used by Travis and Hattemer-Frey with the contact rates used
in this methodology would not particularly change the total exposure estimate - it would
increase only to 9.9 pg/day using the concentrations estimated in this methodology. The
second and more important difference is that air concentrations are derived from soil
concentrations in this methodology whereas they are input for the Fugacity Food Chain
9-33 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 9.6. Comparison of exposure pathway contributions to total daily exposure as
estimated in example Scenario #2 and in Travis and Hattemer-Frey (1991).
Quantity
Travis & Hattemer-Frey (1991)
Scenario #21
Inhalation
concentration, pg/m3
contact rate, m3/day
intake, pg/day
0.02
20
0.40
0.000045
20
0.0009
Water Ingestion
concentration, pg/L
contact rate, L/day
intake, pg/day
Soil Ingestion
concentration, pg/g
contact rate, g/day
intake, pg/day
Fruit and Vegetables
concentration, pg/g
contact rate, g/day
intake, pg/day
Milk
concentration, pg/g
contact rate, g/day
intake, pg/day
Beef
concentration, pg/g
contact rate, g/day
intake, pg/day
Fish
concentration, pg/g
contact rate, g/day
intake, pg/day
0.003
1.4
0.004
0.96
0.02
0.02
0.06
20
1.2
0.02
400
8.0
0.2
90
18.2
0.38
17.6
6.7
0.004
1.4
0.006
1.0
0.06
0.06
0.001 & 0.00042
192
0.09
.0035
300
1.05
.08
100
8.0
0.07
4.1
0.29
TOTAL INTAKE, pg/day
34.3
9.5
1 Quantities in this column were transformed from what is otherwise displayed in this document to be analogous to the
quantities given in Travis and Hattemer-Frey (1991). Specifically, the following was done: 1) all contact rates were
displayed as contact rate (20 m3 inhalation) * contact fraction (0.90; to get 20.7 m3 contact rate); 2) rate of child soil
ingestion for this scenario, 0.8 g/day, was normalized over a lifetime to obtain 0.06 g/day ingestion, and 3) beef and milk
ingestion were expressed in total ingestion rather than in fat ingestion; also beef and milk concentrations were expressed in
whole product rather than in terms of fat.
2 below and above ground vegetative concentrations, respectively.
9-34
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
model. Air concentrations are used for inhalation exposures, to estimate fruit and
vegetable concentrations, but most importantly, beef and milk concentrations in a food
chain model. The air concentrations used by Travis and Hattemer-Frey, in the order of
10~2 pg/m3, are nearly three orders of magnitude higher than are predicted by the
volatilization/dispersion modeling used in this methodology. Section 10.2.9.1. (Chapter
10) discusses the fact that air concentrations of 2,3,7,8-TCDD in the 10~2 pg/m3 range
were found in urban environments, that urban sources of air contamination are likely to
exceed sources in rural environments, and that had the model predicted such
concentrations given low soil concentrations perhaps typical of rural environments, the
model should be questioned. In the same vein, it seems unlikely that concentrations
typical of urban environments would be impacting beef and dairy cattle operations. In any
case, air concentrations used by Travis and Hattemer-Frey were input into the food chain
models of this assessment to evaluate the impact to predicted beef and milk
concentrations. It was found that whole beef concentrations rose three orders of
magnitude to 2.9 pg/g to now exceed the beef concentrations estimated by Travis and
Hattemer-Frey by an order of magnitude, and that milk concentrations also rose three
orders of magnitude to 0.28 pg/g to similarly exceed estimations of Travis and Hattemer-
Frey by an order of magnitude. Finally, above ground concentrations of fruit/vegetables
rose three orders of magnitude to 0.02 pg/g. Using these air, fruit/vegetable, beef, and
milk concentrations, all other concentrations estimated by this methodology, and all the
contact rates as given by Travis and Hattemer-Frey raises the daily intake to 375 pg/day.
The principal difference between 375 pg/day estimated this way and the 38.4 pg/day
estimated by Travis and Hattemer-Frey is how the models of this assessment take air
concentrations and translate them to vegetative, beef, and milk concentrations.
This exercise has demonstrated the complex interplay of fate and transport
modeling, food chain modeling, assumptions regarding the physical environment, and
assumptions regarding exposure behavior. Chapter 10 on Uncertainty critically evaluates
the modeling approaches used in this assessment. Information in that Chapter 10 should
be reviewed when evaluating the validity of the approaches demonstrated in this Chapter.
9-35 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
REFERENCES FOR CHAPTER 9
Broman, D.; Naf, C.; Rolff, C.; Zebuhr, Y. (1990) Analysis of Polychlorinated Dibenzo-P-
Dioxins (PCDD) and Polychlorinated Dibenzofurans (PCDF) in Soil and Digested
Sewage Sludge from Stockholm, Sweden. Chemosphere (21): 1213-1220.
Greaser, C.S.; Fernandes, A.R.; AI-Haddad, A; Harrad, S.J.; Homer, R.B; Skett; P.W; Cox,
E.A. (1989) Survey of Background Levels of PCDDs and PCDFs in UK soils.
Chemosphere 18: 767-776.
Nestrick, T.J.; Lamparski, L.L.; Frawley, N.N.; Hummel, R.A.; Kocher, C.W.; Mahle, N.H.;
McCoy, J.W.; Miller, D.L.; Peters, T.L.; Pillepic, J.L.; Smit, W.E.; Tobey, S.W.
(1986) Perspectives of a Large Scale Environmental Survey for Chlorinated Dioxins:
Overview and Soil Data. Chemosphere 15: 1453-1460.
Reed, L.W.; Hunt, G.T.; Maisel, B.E.; Hoyt, M.; Keefe, D.; Hacknew, P. (1990) Baseline
Assessment of PCDDs/PCDFs in the Vicinity of the Elk River, Minnesota Generating
Station. Chemosphere 21: 1 59-171.
Stenhouse, I.A.; Badsha, K.S. (1990) PCB, PCDD, and PCDF Concentrations in Soils from
the Kirk Sandall/Edenthorpe/Barnby Dun Area. Chemosphere 21: 563-573.
Travis, C.C., and H.A. Hattemer-Frey. (1991) Human exposure to dioxin. The Science of
the Total Environment 104: 97-127.
U.S. Environmental Protection Agency. (1984) Ambient water quality criteria document
for 2,3,7,8-tetrachiorodibenzo-p-dioxin. Office of Water Regulations and Standards,
Washington, D.C. EPA-440/5-84-007.
U.S. Environmental Protection Agency. (1985) Soil Screening Survey at Four Midwestern
Sites. Region V. Environmental Services Division, Eastern District Office, Westlake,
Ohio, EPA-905/4-805-005, June 1985.
U.S. Environmental Protection Agency. (1987) National Dioxin Study. Office of Solid
Waste and Emergency Response. EPA/530-SW-87-025. August 1987.
U.S. Environmental Protection Agency. (1989) Exposure Factors Handbook. Exposure
Assessment Group. Office of Health and Environmental Assessment, Office of
Research and Development, Washington, D.C. EPA/600/8-89/043.
U.S. Environmental Protection Agency. (1991a) Guidelines for Exposure Assessment.
Draft final submitted to the Science Advisory Board. August 8, 1991. OHEA-E-
451.
9-36 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
U.S. Environmental Protection Agency. (1991b) Methodology for Assessing
Environmental Releases of and Exposure to Municipal Solid Waste Combustor
Residuals. Exposure Assessment Group, Office of Health and Environmental
Assessment, Office of Research and Development, U.S. Environmental Protection
Agency, Washington, DC; EPA/600/8-91/031. April 1991.
Yang, Y.Y., and C.R. Nelson. (1986) An estimation of daily food usage factors for
assessing radionuclide intake the U.S. population. Health Phys., 50: 245-257.
9-37 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
10. UNCERTAINTY
10.1 INTRODUCTION
This chapter addresses uncertainty in dioxin exposure assessment performed with
the methodologies presented in this document. Some discussion of the issues commonly
lumped into the term "uncertainty" is needed at the outset. The following questions
capture the range of issues involved:
(1) How certain are site specific exposure predictions that can be made with the
methods?
(2) How variable are the levels of exposure among different members of the
exposed local population?
(3) How variable are exposures between different sites?
The emphasis in this document is in providing the technical tools needed to perform site
specific exposure and risk assessments. For the assessor focusing on a particular site,
question (1) will be of preeminent importance. Therefore the emphasis of this Chapter is
to elucidate those uncertainties inherent to the exposure assessment tools presented in
this document. For the reader focusing on a particular site, the scenario calculations
presented here will be of secondary importance, because site specific data on
concentrations of dioxin-like compounds, land use, and resident populations are, at least in
theory, available. Specific issues that will be addressed are:
- The validation and expected uncertainties of the pollutant fate and transport
models utilized;
- Uncertainties in physical and chemical coefficients used in calculations; and
- Uncertainties in estimated human contact rates with contaminated media and
other human behavioral parameters.
Uncertainties in these site specific input data will, of course, need to be examined by the
site assessor and may contribute significantly to the overall uncertainty in site risks.
A site specific assessment will also need to address the variability of risks among
10-1 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
different members of the exposed population. The level of detail with which this can be
done depends on the assessors knowledge about the actual or likely activities of these
residents. In this document, we illustrate one approach to evaluating this variability.
Separate "central" and "high end" scenario calculations are presented to reflect different
patterns of human activities at a site. The examples presented focus on two areas.
"Central" scenarios are constructed to represent typical behavior patterns for residential
exposures. "High end" calculations focus on a farming scenario where individuals raise
much food for their own consumption. It should be emphasized that high end calculations
could also have been developed for residential exposures by making, for example, higher
range assumptions about duration of residence or contact rates with the contaminated
media. Indeed, this would be recommended for an assessment where considerable
emphasis was placed on residential exposures.
With regard to question (3), this document does not present a detailed
consideration of how exposure levels will vary between sites. As the scenario calculations
in Chapter 9 are intended to be illustrative rather than predictive, the exposure levels that
are obtained here are not intended to be typical of actual site exposures for the pathways
assessed. However, some readers will not be performing site specific exposure and risk
assessments but will refer to this document to obtain a general understanding of pathways
of exposure to dioxin-like compounds. To that end, the example scenarios in Chapter 9
were carefully crafted with regard to source strength, chemical and physical parameters,
and so on, to be plausible and meaningful. Some readers might ideally wish information
on both the magnitude of actual exposures and the variability of these exposures among
sites where dioxin-like compounds are present. However, the analysis presented here can
not support so broad a goal. Representative data to address the variation of dioxin risks
between sites are generally not available and therefore this document can only provide a
partial understanding of this variation through use of sensitivity calculations and
distributional data on some of the input parameters. Therefore, this analysis will include
discussion of the plausibility of site specific assumptions that are made in the example
scenarios to the extent that this is possible. Note, however, that the frequency with
which the basic exposure setting used in the scenarios actually occur, is not addressed in
this document. The structure of the exposure scenario will be taken as basic to the
analysis. For example, the frequency farmers raise dairy cattle near an incinerator is not
10-2 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
addressed. A discussion of observed environmental concentrations of 2,3,7,8-TCDD
compared with modeled concentrations will provide a perspective on the validity of models
used to make such estimations.
10.2. UNCERTAINTIES IN SPECIFIC METHODS APPLIED
Sections 10.2.1. to 10.2.12 discuss the uncertainties associated with specific
assumptions, parameters, and methodologies for estimating individual exposures. Sections
10.2.1 through 10.2.4. are on topics pertinent to more than one exposure pathway,
whereas sections 10.2.5. through 10.2.12. are specific to exposure pathways. Much of
the discussion refers to the example scenarios described in Chapter 9; familiarity with
these examples would be helpful. Also, much of the discussion focuses of 2,3,7,8-TCDD,
one of the three example compounds in Chapter 9. Unless otherwise noted, sensitivity
and uncertainty discussions below are specific to this compound. Sections 10.2.1.
through 10.2.12 contain all or some of the following general subsections and information:
• A discussion of the uncertainties associated with the exposure parameters
including ingestion, inhalation, and contact rates;
• A subsection comparing estimations of exposure media concentrations with
concentrations found in the literature;
• A subsection evaluating alternate modeling approaches to estimating exposure
media concentrations;
• A subsection quantifying the impact of variable and uncertain parameters
required to estimate exposure media concentrations;
• Tables summarizing key uncertainties.
10.2.1. Environmental Chemistry of Dioxin-Like Compounds
This assessment assumed that levels of dioxin-like compounds in soil and sediment
were constant over the period of exposure, with two exceptions. One circumstance was
when contaminated soil eroded from one site and deposited on a site of exposure nearby -
the off-site and ash landfill source categories. The other was when stack emitted
particulates deposited onto a site of exposure - the stack emission source category. In
both these instances, it is assumed that only a relatively thin layer of surface soil would be
impacted, and that this thin layer is subject to dissipation processes - erosion,
10-3 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
volatilization, possibly degradation. Data in Young (1983) implied a soil half-life of 10
years for surficial 2,3,7,8-TCDD residues, although the circumstances of the soil
contamination were not analogous. Specifically, a 37 ha test area at the site had received
an estimated 2.6 kg of 2,3,7,8-TCDD over a two year period. Soil sampling which
occurred over 9 years from the last application suggested that less than 1 percent
remained at the test area. Although Young hypothesized that photodegradation at the
time of application was principally responsible for the dissipation of residues, other
mechanisms of dissipation including volatilization, erosion, and biological removal may also
have contributed to the loss of residues. Soil sampling over time after application implied
a dissipation half-life of 10 years for soil residues of 2,3,7,8-TCDD. This value was
assumed to apply to the other dioxin-like compounds, introducing further uncertainty.
Section 10.2.3.2 below discusses the impact of this assumed half-life for erosion of
contaminated soil. As will be noted, alternate dissipation half-lives did not have a great
impact on estimated soil concentrations.
Section 2.5 in Chapter 2 reviewed the literature on degradation of dioxin-like
compounds. As discussed, biological transformations as well as chemical processes
(oxidation, hydrolysis, and reduction) do not appear to result in substantial degradation of
these compounds. There is evidence of photolysis, particularly when dissolved in solution
and when organic solvents are present. Most of these data are specific to 2,3,7,8-TCDD,
introducing further uncertainty when applied to the other dioxin-like compounds.
Dissipation of surficial residues could translate to lower soil-related exposures
including particulate inhalations, soil ingestion, and soil dermal contact. However, it is not
clear that reductions in exposure would, in fact, occur, particularly if the soil is
contaminated below the surface. Processes such as wind erosion, soil erosion, or
volatilization originating from deeper in the soil profile, could serve, in a sense, to replenish
reservoirs at the soil surface. Given very low rates of degradation (for all degradation
processes except photolysis), the assumption of no degradation is reasonable with
moderate, but unquantifiable uncertainty.
In evaluating an assumption of no degradation, another issue to consider is the
duration of exposure. If such a duration is assumed to be very long, then degradation or
dissipation of soil residues would be more critical than if the duration were relatively short.
Uncertainties associated with the duration of exposure are discussed in Section 10.2.2.
10-4 7/31/92
-------
DRAFT--PO NOT QUOTE OR CITE
below.
Sections below summarize the impact of key chemical-specific parameters to
predictions of exposure media concentrations. The following provides a background and
discussion of these parameters.
The most critical contaminant parameter required for the procedures in this
assessment is the octanol water partition coefficient, Kow, although none of the fate and
transport algorithms directly require a Kow. Rather, empirical parameters are a function of
Kow. As reported in Chapter 2, log Kow estimates for dioxin-like compounds range from
6.00 to 8.5, with higher log Kow associated with higher chlorination. However, this is not
a certain parameter. Estimates in literature for 2,3,7,8-TCDD, for example, range from
6.15 to 8.5.
The most important chemical-specific parameter estimated from Kow is the organic
carbon partition coefficient, Koc. Koc describes the steady state partitioning between soil
or sediment organic carbon and water; it impacts the volatilization flux from soils, and the
partitioning between suspended sediment and water in the water column. Koc is used to
estimate in-situ partitioning using a fraction organic carbon in the soil or sediment, OCs(,
OCsed, and OCssed, as Koc*OCs|, etc. The resulting chemical-specific parameter is termed
the soil (or sediment) partition coefficient, Kds (or Kdsed/ Kdssed). The empirical equation
used to estimate Koc from Kow was derived by Karickhoff (1979). This equation was
chosen over others available (Lyman, 1982) because it was derived from laboratory testing
of 10 hydrophobic contaminants. Others available would have led to lower estimates of
Koc. For example, using a relationship developed by Kenaga and Goring (1980) would
estimate a 2,3,7,8-TCDD Koc (given log Kow for 2,3,7,8-TCDD of 6.64) of 97,500. The
Koc for 2,3,7,8-TCDD estimated for this assessment using Karickhoff's relationship was
2,700,000. Some data implies that this estimate itself may be low for 2,3,7,8-TCDD.
Jackson, et al. (1986) obtained soil samples contaminated with 2,3,7,8-TCDD from 8 sites
in the Times Beach area of Missouri, and 2 from industrial sites in New Jersey. These
contaminated soils had 2,3,7,8-TCDD concentrations ranging from 8 to 26,000^g/kg
(ppb), and organic carbon fractions ranging from 0.015 to 0.08. They determined Kds for
these soil samples, and using the OCS, data, estimated Kocs for 2,3,7,8-TCDD. The mean
Koc from these ten samples was roughly 24,500,000.
The transfer of contaminants from soils to plants requires assignment to two
10-5 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
empirical parameters: the root concentration factor, RCF, and the air-to-leaf biotransfer
parameter, Bvpa. Both these parameters are empirical functions of Kow. Section
10.2.11. describes the impact to estimated vegetation concentration if these parameters
were estimated assuming a different log Kow for 2,3,7,8-TCDD than was assumed for this
assessment.
Another critical contaminant parameter is the Henry's Constant. Table 2-2 (in
Chapter 2) provides estimates of Henry's Constants, H, for dioxin-like compounds, most of
which were estimated given vapor pressure and water solubility data. As seen, the PCDDs
and PCDFs were in the 10"6 to 10~5 atm-m3/mol range, while coplanar PCBs were in the
10~5 to 10~4 range, with one high value at 3x10"3 atm-m3/mol. Sections below on air and
vegetation concentrations describe the impact of this parameter on media concentrations.
Finally, the contaminant molecular diffusivity in air is required for estimates of
volatilization flux from soils. The molecular diffusivity in air is set at 0.05 cm2/sec for all
dioxin-like compounds. Molecular diffusivity is a property of both the chemical and the
medium. It represents the propensity of a chemical to move through a medium. It is
recognized to be largely a function of molecular weight. The values selected are evaluated
as reasonable for all dioxin-like compounds, since the molecular weight for these
compounds are similar.
10.2.2. Lifetime, Body Weights, and Exposure Durations
As discussed in EPA (1989), values for lifetime of 70 years and adult body weight
of 70 kg are derived from large national studies and are not expected to introduce
significant uncertainty into exposure estimates. The assumed child body weight of 17 kg
(for ages 2-6) is similarly well founded and not expected to introduce much uncertainty
into soil ingestion exposure estimates (Section 10.2.5. discusses further issues in soil
ingestion exposure estimation).
Assumptions on exposure durations are the most uncertain of the three parameters
discussed here. A value of 9 years assumed for central exposure scenarios was an
average derived from census survey data (EPA, 1989) which only asked of respondents
the amount of time they lived in their current residences. It is likely to therefore be an
underestimate as an average amount of time spent in one residence (i.e., respondents are
expected to continue to live at their residence). The estimate of 20 years for the average
10-6 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
residence time of farming families (used to define high end exposure scenarios) was not
based on data but rather on judgement that farming families live at their farm site longer
than non-farming families.
Exposure durations are also tied to assumptions about source strength over time.
Assuming 20 years of exposure to incinerator stack emissions, for example, assumes that
the incinerator will be (or has been) in operation for this length of time with the same stack
emission controls in place. If the source is contaminated soil, assumptions include
whether or not the soil will be removed, the site will be capped, and so on. Another
consideration is the dissipation of soil residues. Section 10.2.1. discussed uncertainties
with the assumption of degradation of dioxin-like compounds in soil. As noted, this
assumption is the only one that can appropriately be made. It is reasonable to assume
dioxin concentrations in soil remain relatively constant over the durations of 9 and 20
years, given the data which implies little degradation of dioxin-like compounds in the soil
environment. For the example scenarios of this assessment, it was assumed that the
source strength remains constant over the duration of exposure. In site specific
assessments, which are either based on past or projected exposures, more precise
statements should address the strength of the contamination source over time.
Exposure estimates are linearly related to all three exposure parameters - increasing
body weight and lifetime decreases exposures in an inverse linear fashion, while increasing
exposure durations increase estimates in a direct linear fashion.
Uncertainties associated with body weight, lifetime, and exposure durations are
summarized in Table 10.1.
10.2.3. Soil Erosion Algorithm
This section addresses uncertainties associated with overland transport of
contaminated soil from one site to a nearby site. Soil erosion and deposition was used for
the two source categories evaluating the impact of elevated levels of soil contamination at
a site away from the site of exposure, the off-site and ash landfill source categories.
Transport to an exposure site is modeled using the Universal Soil Loss Equation coupled
with a soil delivery ratio. Two soil concentrations are estimated at the site of exposure.
One assumes delivered soil mixes to a depth of 20 cm. This depth is associated with
tillage operations including vegetable gardening and farming. The other assumes delivered
10-7 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.1. Uncertainties associated with the lifetime, body weight, and exposure duration parameters
Assumption/
Method
Approach
Rationale
Uncertainty
Comments
Lifetime
70 yrs
Body Weight 70 kg adult
Exposure
duration
9 & 20 yrs
Standard EPA assumption
and based on data
Standard EPA assumption
and based on data
Based on assumptions
for central and high
end exposure scenarios.
9 years based on data
for time spent in one
residence; rural farming
families assumed to
live in one location
longer than non-farming
families in rural settings.
Actuary data indicate
that lifetime may
may be increasing
Not much uncertainty
Can vary for popula-
tions in rural settings;
also important to con-
sider how long exposure
has been occurring for
retrospective site-
specific assessments
or how long exposure
may occur for prospec-
tive assessments.
Not a major source of uncertainty
Not a major source of uncertainty
Assuming non-farming families
are more transient than farming
families is probably reasonable,
although data is unavailable
to verify that assumption.
Considering exposure durations
to be in the range of 10-20 yrs
rather than 70 years is felt
to be more appropriate.
Overall: Of these three parameters, the exposure duration is the most uncertain. The estimates of 9 and 20 years were made in this
assessment for non-farming residents in rural settings, and farming residents in rural settings. These values were based on assumptions of
time living at one residence.
soil only mixes to a depth of 1 cm; this is the no-till assumption. The no-till soil
concentration is used to estimate dermal exposures and soil ingestion exposures; tilled
soils are used only to estimate concentration in underground home-grown vegetables.
This section will first compare results from the example scenarios with data from
the literature. Then key assumptions and parameters of the algorithms will be evaluated
for sensitivity and certainty. A summary of the uncertainties using this approach is given
in Table 10.2.
10.2.3.1. Comparison with liters ture da ta
A contaminant concentration ratio is defined for purposes of this discussion as the
ratio of soil concentration at the site of exposure to the soil concentration at the site of
contamination. For example Scenario 3, soil eroded from a 10-acre contaminated site was
assumed to partially deposit onto a 10-acre exposure site. The contaminant concentration
ratio was 0.61 for the 1-cm depth of mixing at the site of exposure and 0.03 for the 20-
cm mixing depth. For Scenario 6, a 67-acre ash landfill eroded 150 meters to a 10-acre
10-8
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.2. Summary of uncertainties associated with the soil delivery algorithm.
Assumption/
Method
Approach
Rationale
Uncertainty
Comments
Quantity of erosion
from contaminated
area
Used average erosion
determined from a survey
of 70 landfills: 62
T/ac-yr.
Use of landfill data
judged reasonable
for example scenarios.
Landfill erosion
estimates ranged from
0.6 to 306 T/ac-yr; 10th
and 90th percentile
were 2 and 163 T/ac-yr.
Reducing erosion to
5 T/ac-yr reduced exposure
site concentrations 64-77%;
increasing to 106 T/ac-yr
increased exp. site
concentrations 5-11 %.
Delivery of soil
to exposure site
Used soil delivery equa-
and additional correc-
when contaminated site
size exceeded exposure
site size
Soil delivery equation
was empirically developed;
with 150 m distance,
calculated soil delivery
ratios were 0.26
and 0.10.
A reasonable range for
delivery 150 meters
speculated as 0.10 to
0.40.
If this range is reasonable,
than exposure site cone.
would be reduced by 1/2 at
the low end and increased
1/5 at the high end.
Concurrent depo-
sition of clean
soil onto
exposure site
"Clean" soil is assumed
to erode at 6 T/ac-yr,
or 1/10 of 62 T/ac-yr
erosion off cont. site;
difference is due to
vegetation on clean soil;
no soil delivery ratio
applied to clean soil
to clean soil.
The mathematical model
assumes clean + cont.
soil erodes to exp. site;
this amount is matched by
an equal amount leaving
exp. site since soil does
not build up on exp. site.
It is unlikely that clean
soil would erode less than
1/10 rate of cont. soil
already assumed - it might
erode more than 1/10;
assuming 1/2 erosion rate
decreases exp. site
concentrations 40-44%.
Assumptions made for
example scenarios are
plausible; more erosion
of "clean" soil could
occur, which would
reduce exposure site
concentrations by less
than 50%.
Rate of dissi-
pation of delivered
contaminant
A rate corresponding to
to a 10-year half-life
assumed.
Dioxin-like compounds
resist degradation,
although evidence suggests
photodegradation for resi-
dues exposed to sunlight;
for deposition of resi-
onto otherwise "clean"
soil, dissipation will
likely occur; 10-year
half-life estimated from
field data.
Half-life of 4 years
decreases estimates of
exposure concentrations
up to 22%; assuming
non-degradation increases
concentrations up to 28%
4 years is likely a
rapid rate of degra-
dation, but non-degrada-
is also unlikely; thus
a degradation assumption
is not critical to model
predictions.
Steady state
assumption
Erosion has occurred
for a sufficiently
long time such that
the system has reached
steady state
All methodologies of this
assessment essentially
assume steady state; bio-
transfer factors, average
volatilization rates, etc.
are based on this assump-
tion.
Testing the analytical
equation including the
time term showed that
concentrations after 5
years were 81% of steady
state, and after 10 years
were 96% of steady state.
A sound assumption if
contamination source
has existed for more
than 5-10 years; could
be critical for evaluation
of planned landfill and
subsequent short-term
impacts.
(continued on next page)
10-9
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.2. (cont'd)
Assumption/
Method
Approach
Rationale
Uncertainty
Comments
Depth of
mixing
1-cm depth assumed
for non-tilled concen-
trations, and 20 cm
for tilled concentra-
tions; non-tilled used
for soil related expo-
sures - ingestion and
dermal exposure; tilled
used for underground
vegetable concentration
estimation.
Evidence for fallout
Plutonium suggests
5-cm for uncultivated
situations and 20-cm
for cultivated situa-
tions. Other analogous
exposure assessments
have adopted two mixing
depths reflecting tilled
and unfilled conditions.
The 20-cm depth
is judged to be
reasonable; the 1-cm
depth may be low.
Increasing it to
2 and 5 cm decreased
exposure site concen-
trations by 18 and 46%.
Of all parameter and
assumptions made for
the erosion algorithm,
this one might be con-
sidered the least cer-
tain. However, given
the tight sorption to
soils and the evidence
for plutonium, a depth
greater than 5 cm would
not be warranted. Hence,
the 1-cm assumption might
be leading to predictions
up to two times too high
Evaluation: Whereas there is some uncertainty with each of the six factors associated with estimating exposure site soil concentrations,
reasonable variations on assumptions and parameters from what was assumed for example Scenarios 3 and 6 would result in changes in
estimated exposure site concentrations within an order of magnitude and mostly within 50% of estimated concentrations (with parameter
changes made individually, not in a Monte Carlo mode). Most of the likely variations would tend to reduce rather than increase exposure site
concentrations. The amount of soil eroding off the contaminated site, 62 t/ac-yr, might be high, and the erosion of clean soil, estimated at 6
t/ac-yr, might be low. A mass balance analytical equation assumed clean and contaminated soil mix with exposure site soil to be matched by
an equal amount of soil eroding off the exposure site; the assumption that clean soil was less erodible than contaminated soil was reasonable.
Assuming steady state for situations where the contamination existed over 5-10 years is justified given that 96% of steady state is reached
after 10 years. The 10-year half-life rate of dissipation in soil was based on data on 2,3,7,8-TCDD, and the model is reasonably insensitive to
changes in that assumption. The model is sensitive to assumed depth of mixing, particularly for the unfilled situation. Concentrations for the
untilled situation were used to estimate soil related exposures - dermal and soil ingestion. If the 1-cm depth, which was assumed in this
assessment, were increased to 5 cm, soil concentrations and related exposures would decrease by 46%. The 1-cm assumption for untilled
situation is not based on data, although others have made the same assumption. This parameter might be considered the most uncertain
because of this lack of data.
exposure site. The contaminant concentration ratios were 0.80 for the 1-cm depth and
0.04 for the 20-cm depth. The impact of a larger size contaminated site is noted for
Scenario 6: going from a 10-acre contaminated site to a 67-acre contaminated site
increased the contaminant concentration ratio from 0.61 to 0.80.
Data to rigorously validate the approach taken in this assessment is unavailable.
However, there have been documented evidence of migration of 2,3,7,8-TCDD away from
industrial sites with soil contamination of 2,3,7,8-TCDD, resulting in off-site soil
contamination. Off-site soil concentrations of concern were identified in 7 of 100 Tier 1
and Tier 2 sites of the National Dioxin Study (EPA, 1987). The study noted that in most
cases, 2,3,7,8-TCDD had not migrated off-site. Most, but not all, Tier 1 and 2 sites did
10-10
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
have some off-site soil sampling without detection. It should be noted, however, that soil
detection limits for most of these 100 Tier 1 and 2 sites were at 1 ppb; this would have
precluded finding concentrations less than 1 ppb in some of the off-site soil sampling,
particularly important for many of the sites where on-site detections were in the low ppb
range. Summary data from the 7 sites noted above is provided in Table 10.3.
Contaminant concentration ratios cannot be evaluated by this summary because of lack of
detail provided in the National Dioxin Study.
Further detail on the 1984 sampling at the Dow Chemical site in Midland is
provided in Nestrick, et al. (1986). An evaluation of the information in that reference is
more informative than the Dow Chemical summary in Table 10.3. The entire site is 607
hectares. On-site sampling included areas identified as chlorophenolic production areas, a
waste incinerator area, and "background" areas. Background areas were within the 607
ha site but away from production areas. Two of the on-site areas were further identified
as areas with Localized Elevated Levels (LELs). These two areas comprise less than 0.5%
of the total site area, but had the three highest occurrences of 2,3,7,8-TCDD at 25, 34,
and 52 ppb. Including these three high occurrences in the total of 33 samples taken on-
site at sites of concern (i.e., not including the background sites) leads to an average
concentration of 4.3 ppb; excluding them leads to an average of 1.0 ppb. The average of
11 background samples (including two ND assumed to be 0.0) was 0.1 5 ppb. A
contaminant concentration ratio of 0.035 is calculated assuming an average concentration
for contaminated soil of 4.3 ppb (0.15/4.3 = 0.035), and a ratio of 0.15 is calculated if
the average soil contamination concentration is more like 1.0 ppb rather than 4.3 ppb.
This ratio of 0.035 is about 1.5 orders of magnitude lower than the 0.61 and 0.80
ratios estimated assuming the shallow 1-cm depth of contamination, although the ratio of
0.1 5 is within a factor of five of these modeled ratios. The depth of 20 cm leads to
modeled ratios of 0.03-0.04, which is more in line with the Dow contaminant ratios of
0.035 or 0.15. The 1-cm depth ratios are probably more pertinent for comparison since it
is unlikely that there were tillage operations (or other soil practices which would distribute
residues) in background areas of the 607 ha Dow site.
It appears reasonable that the no-till contaminant ratios of 0.61 and 0.80 are higher
than the Dow ratios for several reasons. First, the contaminated areas sampled were
those likely to be of concern and comprising only a small percentage of the total 607
10-11 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.3. Summary of off-site soil contamination from Tier 1 and 2 sites of the National Dioxin Study.
Site name
On-site
# samples/range (ppb)
Off-site
# samples/range (ppb)
Comments
Diamond
Alkali
Newark, NJ
Brady Metals
Newark, NJ
9/60-51,000
537/ND-725
10/1.9-3500
30/1.7-1156
Facility involved in the manufacture of 2,4,5-T; off-site sampling
covered a 4000-ft radius including public areas such as a public
housing unit, park, streets, and river. Two of 11 samples from
a park were positive at 1-3.1 ppb; detection limit was 1 ppb.
Other off-site positives were from streets and river sediments.
Site directly associated with the Diamond Alkali site summarized
above; text did not provide any further detail on off-site soil
sampling.
Love Canal
Niagara, NY
NA/NA-6.7
20/3-263
Love Canal contamination well documented elsewhere; few details
provided in reference for soil sampling programs; it was noted
that 3,000 cubic yards of fly ash and BHC cake were taken from
Love Canal in 1954 and used as fill at the nearby 93rd Street
School, a subsurface sample 3 + ft deep showed a concentration
of 6.7 ppb. The off-site summary provided here was from an area
identified as Hyde Park.
Vertac 45/<1-1,200
Jacksonville, AR
320/<1-33.4
A site manufacturing 2,4,5-T; it is not clear than any of the off-
site sampling was for surface soil - summary tables identified it
as "various"; text description did not mention off-site soil conta-
mination and indicated that solid and liquid waste were buried on-
site in a series of landfills. 2,3,7,8-TCDD was found in fish as
far away as 100 miles.
Hooker Chemical 17/ND-18,600
Niagara, NY
4/ND-1.1
A site manufacturing 2,4,5-TCP; subsurface soil sampling ranged
from ND to 18.6 parts per million; one off-site surface soil
detection noted at 1.1 ppb.
Bliss Tank
Property
Rosati, MO
NA/ND-430
NA/ND-430
No summary text provided in primary reference; tabular summary
identified soil sampling as on/off-site soil; non-detects were
noted in 13 off-site dust sampling.
Dow Chemical
Property
Midland, Ml
#1: 43/.041-52.0
#2: 106/ND-1500
11/.0006-.45
42/.003-2.03
Site most extensively studied of those in National Dioxin Study;
Data identified as #1 was a summary of 1984 data supplied in
NDS; #2 was a summary of 1985 data; the 1984 data was further
detailed in Nestrick, et al. 1986; see text for further discussions on
this site.
10-12
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
hectare site. That might question the representativeness of 4.3 ppb as average soil
contamination in impacted areas; the three highest concentrations came from specifically
identified LELs comprising only 0.5% of the 607 ha site area. Second, a map provided in
Nestrick, et al. (1986) including a distance scale clearly shows that all of the background
samples were much further than 150 meters from the contaminated sample points, with
several sample points hundreds to over a thousand meters from the contaminated sample
points. The contaminant concentration ratios, 0.61 and 0.80, were estimated with a
distance of 150 meters. Third, the example scenarios had specific assumptions about
erosion which may or may not have been appropriate for application to the Dow site.
Ideally validation of this model would involve direct application at the DOW site and
comparison of predicted values to measured values. This was not feasible due to lack of
information regarding the DOW site. Instead, this analysis has shown that the model
predictions of contaminant concentration ratios differ logically from observed ratios at the
DOW site. This comparison raises confidence in the model but cannot be considered a
validation exercise.
10.2.3.2. Key Parameters and Assumptions of the Erosion Algorithm
The algorithm to estimate exposure site concentration is comprised of six key
features:
• the total amount of erosion off the contaminated site;
• a soil delivery ratio;
• the concurrent erosion of "clean" soil - soil between the contaminated and
exposure site - which mixes with the eroded contaminated soil upon delivery to the
exposure site;
• the rate of dissipation of contaminants which have reached the exposure site;
• the steady state simplification;
• the depth to which contaminants are assumed to mix in at the site of exposure.
Each of these factors will be examined separately.
1. Erosion from the contaminated site
The quantity of material eroded from the contaminated site is estimated using the
Universal Soil Loss Equation (USLE). The USLE is a widely used empirical model for
10-13 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
agricultural applications. It estimates the edge-of-field soil loss and has been extensively
used to also estimate loss of contaminants bound to soil. The accuracy and precision of
USLE estimates are not reviewed for this assessment. However, the USLE was developed
using literally hundreds of plot-years of erosion (Wischmieir and Smith, 1965), so that
estimates of edge-of-field erosion can be considered reasonable.
USLE parameters used for the example scenarios were derived from a survey of 70
landfills conducted for EPA (SAIC, 1986). The survey, which was based on interviews
with site managers and local officials and review of USGS topographic maps, ascertained
estimates of the parameters related to rainfall, soil properties, topography, and site
management practices. For the 70 landfills, estimated erosion rates ranged from 0.6 to
306 T/ac-yr, with a mean calculated value of 62 T/ac-yr. The 10th to 90th percentile
range was 2 to 163 T/ac-yr. The soil loss estimate for the example scenarios, Scenarios 3
(off-site bare soil contamination with soil concentrations based on concentrations that
have been found in sites of industrial contamination) and Scenario 6 (active ash landfill
with soil concentrations based on estimated ash concentrations), were 62 T/ac-yr. It is
noteworthy that sites with low erosion potential, principally because of vegetative cover,
may have two orders of magnitude less erosion than 62 T/ac-yr. An assumption of dense
vegetative cover would reduce the estimate of 62 T/ac-yr to 6 T/ac-yr.
Sensitivity runs were done for the two scenarios involving off-site soil
contamination, Scenarios 3 and 6. Reducing the erosion by a factor of ten to 6 T/ac-yr
reduces exposure soil concentrations by 64% for Scenario 6 (ash landfill) and 77% for
Scenario 3 (off-site soil contamination). Interestingly, doubling the amount of erosion
leaving the contaminated site to 124 T/ac-yr has a marginal impact on exposure site
concentrations: they would be increased only by 5% (ash landfill) and 11 % (off-site soil
contamination). The estimate of 62 T/ac-yr is generally high - it assumes bare soil
conditions, but an alternate estimate of 6 T/ac-yr is low for contaminated sites such as
industrial sites of landfill. This low erosion estimate is reasonable for land with dense
vegetation such as pasture or undeveloped land. In summary, the high estimate of erosion
used in the example scenarios may lead to an overestimate of soil concentrations at a
nearby site of exposure - perhaps by a factor of 2.
10-14 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
2. Soil Delivery Ratio
The soil delivery ratio is defined as the ratio of soil delivered downgradient at a
specified distance and the edge-of-field erosion losses. The algorithm in this assessment
uses an empirical equation which is a function of length (Equation (5-5)), corrected by
consideration of size discrepancies between contaminated and exposure site. Both
example scenarios involving off-site soil contamination, Scenarios 3 and 6, had exposure
sites located 150 meters from the site of contamination. The calculated soil delivery ratios
for these scenarios were 0.26 (Scenario 3) and 0.10 (Scenario 6).
The soil delivery ratio was varied to evaluate its impact on exposure site soil
concentrations. The soil delivery ratio of 0.26 for Scenario 3 was reduced ten-fold to
0.026, five-fold to 0.05, and to 0.10. Exposure site soil concentrations were 77%, 61 %,
and 38% less with these three reductions as compared to predictions at the 0.26 soil
delivery ratio. Increasing the delivery ratio to 0.40 and 0.50 increased exposure site soil
concentrations by 16 and 24%. Similar exercises were done for example Scenario 6, the
ash landfill. The delivery ratio was reduced ten-fold to 0.01, two-fold to 0.05, and
doubled to 0.20. Exposure site concentrations were reduced 63% and 16%, and
increased 11 % with these changes.
Like the USLE, the soil delivery equation was developed using field data. However,
it was developed for erosion from mining sites to nearby surface water bodies. Its use
here assumes that delivery of erosion from mining sites is similar enough to delivery of
erosion from contaminated soil sites. Given this caveat, estimates from the soil delivery
equation might be reasonable enough to rule out sensitivity estimates at 10-fold reductions
or twice as much delivery, the extremes tested above. If so, one might conclude that the
uncertainty based on this factor is in the range of 1/2 less than estimated concentrations
to 1/5 more.
3. Concurrent Deposition of Clean Soil
Currently, the assumptions for example scenarios #3 and #6 is that the
contaminated site is bare, and that the land area between the contaminated site and the
exposure site has a dense vegetative cover. This translates to a reduction of "C" factor,
or cover, from 1.0 for the contaminated sites to 0.1 for the areas between the
contaminated sites and the exposure site, and a resulting 1/1 Oth unit edge-of-field soil
10-15 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
reduction for this "clean" soil area compared to the contaminated area. All other USLE
parameters are assumed to be the same for both sites for the example scenarios in
Chapter 9. This led to a unit erosion of clean soil of 6 T/ac-yr. Like erosion from a
contaminated site, this erosion estimate is subject to variation in site-specific assessments.
The bare soil assumption applied to the contaminated site might be appropriate for an
active landfill, might be an overestimate for sites of industrial contamination (there would
likely be at least sparse vegetation), but is unreasonable for land separating a
contaminated site and the nearest exposure site.
Another difference between erosion of soil from the contaminated site and from the
clean area is that there is no soil delivery ratio applied to erosion of the clean soil. The
USLE estimates edge-of-field losses, and the clean area is configured to be adjacent to the
exposure site.
Intuitively, the larger the amount of clean soil which mixes with the contaminated
soil, the lower will be the estimated exposure site concentrations. This hypothesis was
tested with increases to the soil loss estimate applied to the clean soil area. When it was
increased to 62 T/ac-yr, the same unit erosion amount as from the contaminated site,
amounts of depositing clean soil increased by ten and exposure site soil concentrations
were decreased between 48 and 59% for the two scenarios. When the unit soil loss was
increased to 31 T/ac-yr, exposure site concentrations decreased slightly less at 40-44%.
The possibility that the land area between the contaminated site and the exposure
site is more erodible than the contaminated site is unlikely, principally because of the likely
existence of vegetation. It can be concluded that if the land area between the
contaminated site and exposure site is more erodible than assumed (but still less erodible
than the contaminated site), than exposure site concentrations were overpredicted by no
more than 50%.
4. The rate of contaminant dissipation
The rate of dissipation of transported residues was assumed to be .0693 yr"1,
which corresponds to a half-life of 10 years. This parameter was changed to 0.1735 yr"1
(half-life = 4 yrs), 0.0198 yr"1 (35 yrs), and 0.0 yr"1 (no degradation). Results indicate
only a modest change in exposure site soil concentrations for the example Scenarios #3
and #6. Decreasing the half-life to 4 yrs reduced exposure site concentrations by 12 and
10-16 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
22% for Scenarios 3 and 6. Increasing the half-life to 35 years increased concentrations
between 7 and 16%, and assuming no degradation increased concentrations between 11
and 28%.
Indications are that photodegradation may be a relevant dissipation process for
dioxin-like compounds. Volatilization is also a process which dissipates residues.
However, these compounds are known to resist degradation. The assumption that in
areas of elevated soil contamination {industrial sites, landfills, etc.), contamination extends
below the surface and residues are assumed not to degrade is thought to be a reasonable
one. The assumption that contaminants transported to otherwise residue-free soil to
impact a thin surface layer are subject to dissipation processes is also a reasonable one.
The assumed half-life of 10 years has been estimated from field data, and variations are
shown to not greatly impact exposure site soil concentrations. The sensitivity exercise
showed that concentrations might be over or underpredicted by no more than 30%.
5. The steady-state assumption
Section 5.4.1, Chapter 5, described the derivation of the algorithm estimating
exposure site concentrations given contaminated site concentrations. The analytical
equation, Equation (5-27), had an exponential term including a time variable. Assuming
long periods of time results in this exponential term approaching 1.0. The term was,
therefore, dropped, to arrive at the steady state version of Equation (5-27), which was
Equation (5-28).
Tests were run including the exponential term with t equal to 1,2, 3, 4, 5, 10, and
1 5 yrs. Expressing results in terms of the percent of steady state concentration reached
at the end of each year, the results for these time intervals are 28, 49, 63, 74, 81, 96,
and 99%, respectively. As seen, 81% of steady state is reached after 5 years, and 96%
is reached after 10 years. Therefore, if off-site contamination has existed for 5-10 years
or more, estimates of exposure site concentration (and related exposures) with a steady
state assumption should be reasonable. However, if an assessment is to be done for a site
to be newly impacted, such as a planned landfill, than the steady state approach would
lead to some overprediction of concentrations and exposures. For the first five years,
concentrations would average about 60% of steady state, and for the first 10 years,
concentrations would average about 75% of steady state. This would be of most
10-17 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
concern for a childhood pattern of soil ingestion, which would be 40% lower for the first
five years of a landfill operation as compared to a steady state assumption. Otherwise, it
is seen that the steady state assumption does not greatly impact exposure site
concentrations.
6. Depth of mixing
Section 5.4.1. describes the justification for the 1-cm and 20-cm depths of mixing
for situations where contaminants from a distant source are transported to a soil surface in
off-site locations. Non-tilled depths of 2, 5, and 10 cm were tested. Results indicate that
exposure site concentrations were reduced by 18%, 46%, and 66% for these three
depths. If a tillage depth of 15-20 cm is judged reasonable, then a 10 cm depth for non-
tilled situations should probably be judged too high. If so, this exercise indicates that non-
tilled exposure site soil concentrations could be overestimated by less than 50%.
10.2.4. Surface Water Suspended and Bottom Sediments
Concentrations of contaminants in surface water sediments are assumed to be
correlated to concentrations in surface soils in the watershed draining into the surface
water body. This section will first compare results from the example scenarios with data
from the literature. An alternate approach to estimating sediment concentrations from soil
concentrations will be evaluated. Then key assumptions and parameters of the algorithms
will be evaluated for sensitivity and certainty. Table 10.4 summarizes the uncertainties
with this algorithm.
10.2.4.1. Comparison With Literature Data
Assuming sediment concentrations are a function of surface soil concentrations is a
reasonable approach when the only source of water body contamination is soil
contamination. However, the algorithm would not be appropriate if sediments and water
were impacted by industrial discharges, which has often been cited as the cause for
sediment and water impacts (see Bopp, et al., 1991; Norwood, et al., 1989; e.g.).
Sediment concentrations of note have also been found in surface water bodies near urban
settings, with car and industrial stack emissions cited as likely causes (Gotz and
Schumacher, 1990; Rappe and Kjeller, 1987). Rappe, et al. (1989) collected samples from
10-18 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.4. Summary of uncertainties associated with the surface water sediment algorithms.
Assumption/
Method
Approach
Rationale
Uncertainty
Comments
Site and watershed
soil concentrations
Site concentration is
site-specific factor;
watershed concentration
assumed to be 0.0 except
for "background" example
scenario, where it was
1 ppt
Assuming no residues in
watershed is the most
appropriate assumption
unless a background
concentration can be
established.
Assuming a background
concentration of 1 ppt
which is 3 orders of
magnitude less than
the site contamination of
1 ppb resulted in only
a 5% increase in sediment
concentrations.
No change in assumption
is currently warranted.
Site and watershed
size and delivery
ratios
Quantity of erosion
from contaminated
area and watershed
Stack emission
algorithm
Sizes are site-specific
and known with certainty;
delivery ratios empiri-
estimated given distance
to water body for site,
and watershed size
Like soil erosion
algorithm, assumed 62
T/ac-yr for contaminated
sites; 6 T/ac-yr
assumed for watershed.
ISC modeled watershed
representative deposition
rates mixes in represen-
tative mixing zone depth
to estimate soil concen-
tration for watershed;
sediment concentration
equals this concentration
Approach appropriately
considers sizes and
distances, and uses
developed empirical
relationships to estimate
delivery ratios
Assuming contaminated
site is more erodible
than average for water-
is not inappropriate,
although 10 times more
erodible might be high
Approach for stack
emissions is consis-
with other approaches
which assume sediment
in water body is a
function of soil concen-
trations in watershed
Delivery ratio for site
developed from surface
mining data and for
distances up to 250
m, use for sites further
from water body would
likely overestimate
delivery of contaminated
soil
Site-specific estimates
would lead to erosion
rates that are
known with reasonable
certainty.
Changes in sizes and
distances show direct
linear impact to sediment
concentrations; watershed
size appropriately selected
as median value from survey
of landfills.
Sediment concentrations are
linearly related to estimates
of unit soil loss from both
the site and the watershed.
Does not consider
direct depositions to
water body which could be
critical if water body is
very large or if incinera-
is near where water is
withdrawn; picking an ISC
modeled deposition rate
is a site-specific exercise;
depth of mixing in watershed
is uncertain
Example scenarios picked
ISC deposition rate at
5000 meters; exposure sites
were at 500 and 2000 m;
representative mixing depth
for watershed assumed to
be 10 cm.
Overall: Data was unavailable on the impact of surface water sediments from erosion of soil contaminated with dioxin-hke compounds. Most,
if not all, of the literature on sediment impacts attribute causes to direct industrial discharges or, where the source is not a point source, to an
"incineration pattern", which links concentrations found in sediments to concentrations found in air and air particulates where sources are
speculated to be incinerator, car, or other industrial air emissions. The algorithm estimates sediment concentrations as a linear function of soil
concentrations within the watershed. Sensitivity analysis showed mostly a linear impact to sediment predictions with all parameters - some of
the impacts are direct and some are inverse. An alternate modeling approach was compared with the approach in this assessment. Both
approaches predicted similar soil/sediment concentration ratios (ratio of concentration on contaminated land area within watershed to
concentration on bottom sediments). Changes in predicted sediment concentrations from changes in any one parameter would result in
concentrations within an order of magnitude of currently modeled concentrations.
10-19
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
the Baltic Sea, which were described as background samples. They note that the pattern
of tetra-CDF congener concentrations found in the Baltic Sea were typical of what they
termed the "incineration patterns" - air and air paniculate concentrations that were
attributed to sources such as incineration, car exhausts, steel mills, etc. On the other
hand, sediment samples collected between 4 and 30 km downstream from a pulp mill
revealed a congener pattern typical of bleaching mills. Different approaches to estimating
sediment concentrations as a function of direct industrial discharges or direct air-borne
depositions would be required for these sources. The effluent discharge issue is further
discussed in Appendix D.
Data was not found linking sediment concentrations to soil concentrations, in an
urban or more pertinent to this assessment, a rural setting. Some sampling has occurred
in areas described as rural or background. Sediment sampling in Lake Orono in Central
Minnesota in such a setting found no tetra- and penta-CDDs, although occurrences of total
hexa-CDDs were found in the low ng/kg (ppt) level, occurrences of hepta-CDDs to a high
of 110 ppt, and total OCDD concentrations ranged from 490-600 ppt for three samples
(Reed, et al., 1990). A report on sampling of several estuaries in Eastern United States
included a "reference" or relatively clean site, central Long Island Sound. There were no
occurrences of 2,3,7,8-TCDD, although 2,3,7,8-TCDF was found at 15 ppt in this clean
site. Other sites had identified industrial source inputs and higher noted concentrations
(Norwood, et al., 1990). 2,3,7,8-TCDD was extensively found in sediments of Lake
Ontario (EPA, 1990a). The average of samples from all depths of sediment collection from
49 stations including 55 samples was 68 ppt. The average of 30 surficial sediment
samples was 110 ppt. A modeling exercise implied that an annual load of 2.1 kg/year
into Lake Ontario corresponds to a concentration of 110 ppt. One identified source was
the Hyde Park Landfill, located about 2000 feet from the Niagara River, which drains into
Lake Ontario. Between 1954 and 1975, an estimated 0.7 to 1.6 tons of 2,3,7,8-TCDD
were deposited in the landfill. A principal conclusion from the modeling exercise, however,
was that a characterization of historical loadings of 2,3,7,8-TCDD into the lake was not
available and would be necessary to evaluate the contributions by the Hyde Park Landfill.
The sediment concentrations of 2,3,7,8-TCDD predicted to occur in the river for the
six example Scenarios in Chapter 9 were 0.005 ppt (Scenarios 4 and 5 - stack emission
source category), 0.2 ppt (Scenarios 1 and 2 - on-site source category), 4 ppt (Scenario
10-20 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
3 - off-site source category), and 17 ppt (Scenario 6 - ash landfill). The highest
concentrations in sediment were predicted to occur from sites of elevated concentrations
located 150 meters from the river. 2,3,7,8-TCDD was set at 1 //g/kg (ppb) at a 4 ha site
for Example Scenario 3 demonstrating the off-site source category, and was set at 0.7 ppb
for the 27 ha site of Example Scenario 6, demonstrating the ash landfill scenario. These
elevated concentrations for the small land areas were three orders of magnitude higher
than soil concentrations estimated or assumed in the other four Scenarios. One ratio of
note for Scenarios 3 and 6 is the ratio of soil concentration at the site of contamination to
soil concentration in the sediments. These contaminant ratios equal 0.004 for Scenario 3
and 0.024 for Scenario 6. No literature was found which listed contaminated soil and
resulting surface water body sediment concentrations, which would be necessary to
evaluate these ratios. The next section, however, does attempt an alternate modeling
approach to arrive at analogous ratios.
10.2.4.2. An Alternate Approach
As noted, the validity of these results cannot be rigorously evaluated because of a
lack of data on impact to surface water bodies by soil erosion of dioxin-like compounds.
However, the dilution of contaminated sediments entering a river system can be estimated
using an alternate approach.
The average runoff rate for the midwestern U.S. is about 15 inches/year (Linsley, et
al., 1982). For a 10,000-acre watershed (4,000 hectares; the watershed size and
effective drainage area for Scenarios 3 and 6), this yields a stream flow of about 17.2
ft3/sec. The sediment yield can be estimated from the stream flow as follows (Linsley, et
al., 1982): Qs = aQn, where Qs = sediment flow rate (Eng. T/yr); Q = stream flow
rate (ft3/sec); and a and n are empirical constants, reflecting the vegetative cover in the
watershed. Linsley, et al., (1982) recommend using a = 3,500 and n = 0.82 for coniferous
forest and tall grass, and a = 19,000 and n = 0.65 for scrub and short grass. Substituting
these into the equation above (and Q = 17.2 ft3/sec) gives an annual sediment flow rate
of 36,000 to 121,000 T/yr. Total sediment flows will be comprised of contaminated as
well as uncontaminated sediments. A "contaminant concentration ratio" can be calculated
by estimating the sediment contributed by the contaminated areas and dividing by this
sediment flow rate range; this assumes all other sediment contributions are
10-21 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
uncontaminated. The annual soil loss for Scenarios 3 and 6 was 62 T/ac-yr. The
contaminated site area for Scenario 3 was 10 acres (4 ha) and the area for Scenario 6 was
67 ac (27 ha). The total soil erosion contributed by these sites equals: the unit soil loss *
area * soil delivery ratio; for Scenario 3, this equals 62*10*0.26, or 161 T/yr, and the
total amount contributed by Scenario 6 was 62*67*0.26, or 1080 T/yr. The contaminant
concentration ratio is (161 T/yr)/(36,000 to 121,000 T/yr), or a range of 0.001-0.004.
This can be compared to the contaminant ratio estimated in the previous section for
Scenario 3 at 0.004. For Scenario 6, the range is 0.009-0.03, which compares to 0.024
estimated from above.
This implies that both approaches estimate similar contaminant ratios, although
they are both subject to key assumptions. A lower runoff rate (instead of 1 5 inches/year)
would have resulted in lower total sediment yields using the approach given in Linsely, et
al. (1982). Also, several assumptions were made to estimate sediment concentrations in
the approach of this assessment. The impacts of these assumptions are discussed below.
Of particular note is the difference in erosion rates assumed for the contaminated site in
comparison to the watershed. Contaminated site erosion was high at 62 T/ac-yr, and
watershed rate was low at 6 T/ac-yr. If both rates were assumed to be equal at 31 T/ac-
yr, estimated contaminant ratios would drop by a factor of ten: the contaminant
concentration ratio for example Scenario 3 would be 0.0004 instead of 0.004.
Recalculating the ratio range using the Linsley method above would yield 0.0005-0.002.
Now, 0.0004 is slightly lower than the range of 0.0005-0.002. The contaminant
concentration ratio for Scenario 6 would be 0.0024, also lower than the analogous range
of 0.0045-0.015. In short, both methods are based on assumptions about erodibility and
both appear to be consistent with each other.
10.2.4.3. Sensitivity to Key Parameters and Assumptions
The parameters estimating sediment concentrations are of three types: site-specific,
empirical, and modeled parameters. Site specific parameters include: concentrations at the
site of contamination and average for the watershed, Cs and Cw, distance from
contaminated site to water body, DIS, and areas of contamination, As, and of the
watershed, Aw. The empirical parameters include the sediment delivery ratios assigned to
the contaminated site and for the watershed, SDS and SDW, and the depth of mixing
10-22 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
assumed for incinerator stack depositions in order to estimate watershed average soil
impact. The modeled input were the incinerator stack deposition rates assumed to be
average for the watershed and the unit soil loss assumed for the contaminated site and the
watershed, SLS and SLW.
The impact to estimated sediment concentrations in the river to reasonable changes
in most of these parameters is minimal. Sensitivity tests described in this section were
conducted using example Scenario 3 from Chapter 9, which demonstrated the off-site soil
source category. For that demonstration, a 4 ha area was initialized to 1 //g/kg (ppb)
within a 4,000 hectare watershed. Background watershed concentrations were set at 0.0
for Scenarios 3.
1. Site and Watershed Contaminant Concentrations
Assuming that only the contaminated site has residues and that all other watershed
soils are residue free may be inappropriate and would underestimate surface water
impacts. The evaluation of the Dow Site in Midland, Ml (Nestrick, et al., 1986; described
above in Section 10.2.2.1) showed part per trillion levels of soil concentrations within the
city of Midland, which was three orders of magnitude less than concentrations at the site
of contamination. When the watershed concentrations in Example Scenario 3 were set at
1 ppt, three orders of magnitude lower than the site concentrations of 1 ppb, river
sediment concentrations increased by less than 5%.
Sediment concentrations are roughly linearly related to soil concentrations at the
site of contamination. Increasing the concentration from 1 //g/kg (ppb) to 10 ppb
increased sediment concentrations by a factor of 9.5; decreasing it to 0.1 ppb has roughly
the same order of magnitude reduction.
2. Site Size and Delivery Ratio and Watershed Size and Concurrent Sediment
Delivery Ratio
Sediment delivery ratios had an impact on sediment concentrations. When the
contaminated site was 50 meters instead of 1 50 meters from the rivers edge,
concentrations increased by roughly 27%. When the sites are set at 500 meters from the
rivers edge, the concentrations decreased by 23%. These distances are used to estimate
the sediment delivery ratio from a site to the edge of the water body. The range of ratios
10-23 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
estimated with these distances (see Equation (5-5), was 0.20 (at 500 meters) to 0.33 (at
50 meters), which is a rather narrow range. It was noted in Section 5.3.1 that the
sediment delivery ratio equation was developed from construction sites and from data up
to 250 meters from a water body. The delivery ratio estimated with this equation at 500
meters of 0.20 might be high. The delivery ratio was reduced to 0.05, a factor of 5
reduction from the estimated ratio of 0.26 at 150 meters, and sediment concentrations
were also reduced by 5. If the contaminated site drained directly into the river - the
delivery ratio equals 1.0, and concentrations would increase by 4.0. In other words,
sediment concentrations are linearly related to the sediment delivery ratio applied to the
contaminated site. It also seems reasonable to assume, unless the site were adjacent to a
water body, that a range of reasonable sediment ratios are in the 0.10 to 0.40 range; the
value of 0.26 estimated at 150 meters would appear to be a reasonable midrange value.
The size of the contaminated site was found to be roughly linearly related to
sediment concentrations. Increasing the contaminated site from its initial size of 4 ha to
40 ha increased concentrations by a factor of 9; decreasing the size to 0.4 ha decreased
concentrations by an order of magnitude.
Sediment delivery ratios applied to a watershed are narrowly defined. As seen in
Figure 5.1, the watershed sediment delivery ratio, SDW, is approximately 0.1 2 for a 1,000
hectare watershed, and 0.22 for a 10,000 hectare watershed. The value selected for the
example scenario for a 4,000 ha watershed area was 0.1 5. Changes in watershed
sediment delivery ratios should accompany changes in assumed watershed size. Tests
were run increasing the watershed size to 10,000 ha with an accompanying watershed
delivery ratio to 0.12, and reducing the watershed size to 1,000 ha with the appropriate
ratio of 0.22. Increasing the watershed size decreased the sediment concentrations by
1/2; and decreasing the watershed size increased sediment concentrations by 2.5.
A small watershed may not have a water body adequate to support fish for
consumption or water for drinking exposures. SAIC (1986) reports watershed areas for
most of the 70 landfills for which erosion rates were calculated. These watershed sizes
vary considerably with 10 of 51 watersheds being smaller than 1,000 acres (400 ha) and
1 6 of the 51 being 100,000 acres (40,000 ha) or above (median 11,000 acres - 4500 ha;
mean 29,000 acres - 12,000 ha). Thus the assumed value of 10,000 acres, or 4,000 ha,
falls in the middle of the observed values.
10-24 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
3. Unit Soil Loss Estimates for Site and Watershed
Concentrations are also impacted by assumptions for unit soil loss, including that
from the contaminated site, SLS, and that from the watershed on the average, SLW. The
unit soil loss from the contaminated site was modeled from the landfill data earlier
described (SAIC, 1986). The value of 62 T/ac-yr is high - mainly because of the
assumption of bare soil. All other parameters were reasonable mid-range values (see
Section 5.3.1). Reducing SLS by 1/2 to 31 T/ac-yr decreased sediment concentrations by
1/2. The unit soil loss assumed for the watershed on the average, on the other hand, was
relatively low, at 6 T/ac-yr. The only difference was the assumption of erosion retarding
ground cover (grass, crops, etc.). If the average soil loss over the watershed were instead
5 times higher at 31 T/ac-yr, sediment concentrations would be diluted and would be five
times lower. As seen, there is a linear relationship between soil loss estimates and
estimated sediment concentrations.
4. Stack Emission Deposition Rate and Mixing Depth
Average concentrations over a watershed resulting from stack emissions are a
function of an average watershed contaminant deposition rate, a representative watershed
mixing zone depth, and an assumed dissipation rate for depositing residues. The impact
of the assumption of a dissipation rate corresponding to a 10-year half-life was minimal, as
discussed above in Section 10.2.3.2.
The validity of deposition rates modeled with the ISC model is discussed in Section
10.3. A representative deposition rate for the example scenarios was one given in Table
6.19 corresponding to the stack emission controls which were demonstrated, and at a
distance of 5,000 meters; the exposure sites of the demonstration were at 500 and 2,000
m from the incinerator. For a 4,000 hectare watershed, depositions modeled to occur at
5,000 meters could be high, low, or appropriate depending on: 1) where water is
withdrawn and fish caught for consumption within the watershed, 2) where the incinerator
is located within the watershed, and 3) climate, terrain, and all other assumptions that are
required for atmospheric transport using the ISC model. For the example scenarios, water
was assumed to be withdrawn, and fish caught for consumption, at the bottom of the
watershed - this translates to the assumption that all land draining into the river system
impacts the sediment quality. If water is withdrawn from the middle of the watershed, the
10-25 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
appropriate assumption would be that all land upgradient of the withdrawal point impacts
the sediment quality. Selection of an appropriate deposition rate is a site-specific issue.
Selecting the rate at 5000 meters was done for this assessment also because it was felt
that the average watershed soil concentration should be lower than concentrations
assumed to occur at relatively close exposure sites (500 and 2000 meters). Ideally, this
analysis would be done by estimating the deposition in numerous small subsections of the
water shed and using an integration approach to estimate the total contaminant load to the
stream.
In any case, the average watershed soil and hence sediment concentrations are a
linear function of deposition rates. Figure 6-3 shows that deposition rates at 10 km are
half of what they are at 5 km, and they are likely to be half again at 20 km. Sediment
concentrations using a deposition rate corresponding to 10 km would, therefore, be one-
half what they were assuming the rate at 5 km.
The assumed average mixing depth over the watershed was 10 cm. This is
midway between the assumption of 1 cm for the non-tilled mixing depth and 20 cm for
tilled situations. All example scenarios were described as rural settings. If the rural
setting were predominantly agricultural, then an average mixing depth might be closer to
20 cm. However, agricultural experts are currently promoting no-till or minimum till in
order to reduce erosion losses. With these practices, a mixing depth due to tillage may
not be as high as 20 cm. Also, a rural setting which is not predominantly agricultural may
also not call for an average mixing depth of 20 cm. In any case, average watershed
concentrations and hence sediment concentrations are inversely related to mixing depth
assumptions: doubling it to 20 cm would halve the estimated sediment concentrations,
and reducing it to 5 cm would double sediment concentrations.
10.2.5. Soil Ingestion Exposure
This exposure is directly a function of the concentration of contaminants in surface
soil layers. For example Scenarios 1 and 2, demonstrating the on-site soil source
category, soil at the site of exposure was contaminated to a specified level. For example
Scenarios 3 and 6, demonstrating the off-site and ash landfill source categories, erosion
onto the site of exposure deposited residues into a thin, no-till, surface layer of 1 cm, and
a thicker, 20-cm, till layer of soil. Soil ingestion exposures were based on concentrations
10-26 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
in the 1-cm layer. In Scenarios 4 and 5 demonstrating the stack emission source
category, contaminated particles deposited onto the exposure site creating a till and a no-
till concentration as well.
Uncertainties associated with the erosion methodologies are discussed in Section
10.2.1 above, and uncertainties associated stack emission and deposition and discussed in
Section 10.3. below. Of particular note from these discussions is the uncertainty
associated with a 1-cm depth of mixing. As was noted, others have also assumed depths
of mixing this shallow for analogous applications. Evidence from radioactive fallout
suggests depths no deeper than 5 cm. Increasing the depth to 5 cm would reduce surface
concentrations by a factor of five and estimated soil ingestion exposures as well.
Another issue is whether children should be assumed to be exposed to tilled soils,
tilled by home gardening, farming, etc., or untilled soils. It is feasible that children would
be exposed to tilled soils, which have lower concentrations, during periods of home garden
preparation. However, it is more reasonable to assume that they generally play outside in
areas that are not mechanically tilled.
The estimated soil ingestion quantity is based on field measurements, using trace
elements, of soil ingested by relatively small groups of children over brief periods.
Methodological issues in these studies remain to be addressed. In particular, ingestion
estimates may have been lower if dietary intake of the trace elements was taken into
account. Research is underway to refine soil ingestion estimates obtained through trace
element measurements. Given the available data, 0.2 g/day is used as a typical value for
soil ingestion in young children. Due to the behavior known as pica, some children are
known to be high ingesters of various non-food materials; although no quantitative data on
soil ingestion are available for children known to exhibit pica, the use of the high-end
estimate of 1.0 g/day may better reflect such behavior.
Because of remaining methodological research needs, no quantitative estimate of
the uncertainties in these estimates is made here.
Soil ingestion exposure estimates also depend on the duration of the period over
which children are assumed to ingest soil. Data on soil ingestion by age are not available,
and the estimate that significant ingestion occurs between ages 2 and 6 is broadly
supportable on behavioral grounds.
No measurement data are available on soil ingestion in infants (0-2 yrs. old) or in
10-27 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
older children or adults, and no ingestion is assumed for these groups. While some soil
ingestion will occur in these groups, e.g., through contact of soiled hands with food, it is
plausible that such ingestion is of a lesser degree than occurs in early childhood. If
Hawley's (1985) estimate that an adult ingests an average 0.060 g/d of soil is used, after
accounting for differences in exposure duration (9-20 yrs vs. 5 yr) and body weight (70 kg
vs. 1 7 kg), the adult soil ingestion exposure is close to the estimated exposure for children
(at 0.2 g/d). The high end example scenarios in Chapter 9 assumed that the exposed
family was involved in farming operations. One implication is that individuals on the farm
would be working closely with the soil, which may result in some soil or dust ingestion
(dust ingestion is distinct from the particulate inhalation exposure pathway). The other
implication is that, should this be the case, they would be in contact with tilled soil, whose
concentration is 20 times less than the no-till soil for which children are assumed to be
exposed.
Considering these uncertainties, the soil ingestion exposure estimates presented for
children are plausible. Further consideration may be warranted for considering adult soil
ingestion, particularly in farming situations. Uncertainties associated with the soil
ingestion pathway are summarized in Table 10.5.
10.2.6. Soil Dermal Contact Pathway
Estimates of dermal exposure to soil rely largely on four factors unique to this
pathway: exposed skin area, soil adherence, frequency of soil contact and fraction of
contaminant absorbed. The uncertainty in these three terms are discussed below.
The uncertainty in the assumed value for exposed skin area reflects primarily
population variability. As reported in EPA (1992), relatively accurate measurements have
yielded a good data base on total skin area. Thus the uncertainty in this factor is derived
more from the assumptions of how much of the total skin area is exposed. EPA (1992)
recommends approaching this issue by determining the coverage of normal apparel in the
exposed population and assuming exposure is limited to the uncovered skin. As discussed
in EPA (1992), this assumption could lead to underestimates of exposure since studies
have shown that some exposure can occur under clothing, especially in the case of vapors
or fine particulates. A default assumption of 25% uncovered is recommended
corresponding to short sleeved shirt, short pants, shoes, and socks. Thus the key
10-28 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.5. Uncertainties associated with the soil ingestion pathway.
Assumption/
Method
Approach
Rationale
Uncertainty
Comments
Children exposed
to surficial
residues
Migrating residues
deposit in a 1-cm
no-till depth and
a 20-cm till depth;
children are exposed
to concentrations
estimated at 1-cm
There is no research
to refine the 1-cm
estimate, although
others have assumed simi-
larly shallow depths for
analogous applications;
evidence suggests a no-
till depth no deeper
than 5 cm.
Increasing the no-till
depth to 5 cm would
reduce concentrations
and exposures by
a factor of 5.
Sections 10.2.1 and
10.3 discuss related
uncertainty issues
concerning migration
of dioxin-like com-
pounds; a 1-cm depth
is a conservative es-
timate that should
be further examined.
Child's inges-
tion rate (2-6
years old).
Ingestion rate assumed
to vary from 0.2-0.8 g/d.
The range selected was
primarily based on the
results of two field
studies of soil ingestion
in children.
Field study methodology
not fully validated.
Data from several sources
indicate this range of
values for small children.
Pica children have been
estimated to ingest higher
quantities (5 g/d).
Ingestion rate
for other ages
Ingestion assumed to
occur only during
ages 2-6.
Mouthing tendencies
strongest and under-
standing of personal
hygiene low during
childhood
Hawley estimates inadver-
tent ingestion may be 60
//g/d for adults, which
would lead to an estimated
exposure pattern similar
to that of children
Adults may inadvertently
ingest soil during gardening
and yard work; farmers may
have a non-trivial soil ingestion
pattern
Overall: Soil ingestion for older children and adult was not considered, which may have underestimated exposures by a factor of two. The
other major area of uncertainty for this pathway is for the scenarios where the source of contamination is located distant from the site of
exposure, including areas of high soil contamination (the off-site and ash landfill source categories) and the stack incinerator. In those cases,
depositions onto the site of exposure are distributed within a thin 1-cm layer. The depth of this layer is uncertain - it is more likely too low than
high. Increasing it to 5-cm would decrease exposure estimates by a factor of five. Pica soil ingestion patterns were not evaluated in this
assessment; the ingestion rates considering this appear reasonable.
uncertainty issue concerns the variability in clothing behavior of the exposed population.
In this document the 25% assumption was adopted for residents and 5% was judged
more reasonable for farmers who are more likely to wear long pants and long sleeved
shirts for field work. Although clothing coverage is likely to vary over the year and with
personal habits, these assumptions are judged to be reasonable averages and unlikely to
introduce more than a factor of two uncertainty.
The potential for soil adherence probably varies little across the population, but few
actual measurements have been made. Thus the uncertainty in these estimates reflect
primarily the lack of measurement data rather than population variability. Site variability is
probably important as well since soil properties such as moisture content, clay content and
10-29
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
particle size distribution are likely to affect adherence. EPA (1992) reports four studies
which estimated soil adherence on hands under both laboratory and field conditions. Data
from these studies were analyzed to obtain a central estimate of 0.2 mg/cm2 and a high
end estimate of 1.0 mg/cm2. The uncertainty in these estimates are derived from
unknown efficiencies in the collection methods, relatively small number of subjects,
assumption that hand measurements apply to other parts of the body and assumption that
child measurements apply to adults as well. The central default value 0.2 mg/cm2 was
adopted here for the residents and the high end value of 1.0 mg/cm2 was adopted for
farmers. The uncertainties in this estimate could combine to produce either under or over
estimates and may vary by as much as a factor of 5 on the basis of the ratio of the high
end to central estimates.
Exposure frequency to soil reflects largely personal habits and thus the uncertainty
is primarily based on population variability. Seasonal and climate conditions can also
affect this behavior introducing site variability as well. EPA (1992) suggests a central
frequency of 40 days/yr corresponding to someone who does yard work, gardens or plays
outdoors on most weekends and a high end estimate of 350 days/yr corresponding to a
serious gardener in a warm climate. These recommendations were based on judgement
rather than actual survey data. In this document, 40 days/year was selected for the
residential scenarios and 350 days/yr for the farmer. The lack of survey data to support
these estimates introduces uncertainty, but the values are judged to be reasonable and to
create relatively little uncertainty.
The dermal absorption fraction of compounds varies widely across chemicals,
whereas skin properties that affect absorption, i.e. thickness and composition vary little
across the population. Thus the uncertainty in this factor is derived primarily from
measurement error rather than population variability. Soil properties, such as organic
carbon content, can also affect the extent of dermal absorption and thus create site
variability as well. EPA (1992) reports two studies which measured dermal absorption of
2,3,7,8-TCDD from soil. Testing included human skin in vitro, rat skin in vitro and rat skin
in vivo. On the basis of these tests, a range of 0.1 - 3.0% was recommended in EPA
(1992). Dermal absorption testing, especially for soils, is a relatively new field and many
uncertainty issues are involved. These include extrapolation of animal tests to humans,
extrapolation of in vitro to in vivo conditions, and extrapolation of experimental conditions
10-30 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
to expected exposure conditions. Extrapolation of the tests on 2,3,7,8-TCDD to the other
dioxin like compounds (which have not been tested) introduces further uncertainties. A
dermal absorption fraction of 3.0% was adopted here for application to all the dioxin like
compounds. Based on the observed range of values for 2,3,7,8-TCDD this assumption
may lead to overestimates of a factor of 30. Considering all possible uncertainties, under
estimates are also possible, though judged less likely.
In summary, dermal exposure estimations rely on a number of parameters whose
values are not well established. Although it is difficult to estimate the overall uncertainty
with this pathway, it is judged to be plus or minus one to two orders of magnitude. A
summary of the uncertainties associated with the dermal absorption pathway is given in
Table 10.6.
10.2.7 Water Ingestion
Surface water concentrations of dioxin-like compounds are driven by concentrations
estimated to occur in suspended sediments. Suspended sediment concentrations are a
function of surface soil concentrations of dioxin-like compounds within the watershed, or
sub-watershed area, draining into the water body. The uncertainty in estimating sediment
concentrations are discussed in Section 10.2.4 above. As water concentrations are
directly and linearly related to suspended sediment concentrations, uncertainties in
estimating concentrations in suspended sediment are also pertinent to estimating surface
water concentrations.
The section begins with evaluating the exposure parameters, which include the
water ingestion rate and then contact fraction. Then, the model results are compared with
limited water quality data. The next subsection considers an alternate approach for
estimating water column concentrations. Finally, the sensitivity and variability of key
model parameters are evaluated. Table 10.7 summarizes the uncertainties associated with
water ingestion.
10.2.7.1. Exposure Parameters
The classically assumed water ingestion rate of 2.0 L/day was examined in EPA
(1989). The conclusion was that this estimate is more appropriately described as an upper
percentile consumption rate for adults, and recommended 1.4 L/day for use as an average.
10-31 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.6. Uncertainties associated with the dermal exposure pathway.
Assumption/
Method
Approach
Rationale
Uncertainty
Comments
Soil
Concentrations
Contact
rate
Contact
frequency
Surface
area
Absorption
fraction
same for on-site source
category; used "tilled"
concentrations for high
end farming scenarios,
"non-tilled" for non-
farmers in central
scenarios for all
other source categories
0.2 mg/cm2-event for non-
farming residents;
1 mg/cm2-event for
farmers
40 for residents;
350 for farm families
behavior parameters
for farmers assume
dermal exposure results
from farming in tilled
soils; for non-farmers,
yard work during summer
assumes dermal contact
occurs in non-tilled
soils
corresponds to central
and high end values of
measurement data;
supported by EPA, 1992
Based on judgement;
supported by EPA
(1992)
5% (1000 cm2) for farming Based on total body
adults and 25% (5000 cm2) surface area data and
for non-farming adults clothing assumptions;
supported by EPA (1992)
0.03 for all dioxin-
like compounds
Based on EPA (1992)
which gave a range of
0.001 to 0.03; range
supported by 2 different
studies in rat and human
tilled concentrations are
20 times less than non-tilled
concentrations; much un-
certainty associated with
difference between tilled
and unfilled mixing zone
depths; non-farmers working
in home gardens also are
exposed to tilled
concentrations
measurement data may have
experimental error or not be
representative; uncertainty
plus/minus factor of 5
personal behavior patterns
could differ but uncertainty
judged to be small
Clothing assumptions
based on judgement
rather than survey data;
uncertainty judged to be
plus or minus factor of
2.
Experimental procedures
very uncertain, may over
estimate by factor of 30
For real sites, monitoring
is best way to resolve this
uncertainty
soil properties could
affect adherence
climatic conditions intro-
duce site variability
Studies have shown that
fine particulates can
deposit under clothing
Soil properties may also
affect absorption
Overall: The high uncertainty in estimates of soil adherence and absorption fraction make the overall uncertainty in the exposure estimates
highly uncertain, judged to be plus or minus 1 to 2 orders of magnitude.
However, EPA (1989) cautions that data on consumption rate for sensitive subpopulations
such as manual laborers are unavailable. As such, the 1.4 L/day rate for individuals in
farming families who work the field may be low.
The contact fraction is defined as fraction of total contact with an exposure media
that is contact with contaminated media. For drinking water, this translates to the fraction
of water ingestion that comes from the contaminated water source. In the example
scenarios, it was assumed that the impacted water was a river which supplied water to
10-32
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.7. Uncertainties associated with the water ingestion pathway.
Assumption/
Method
Approach
Rationale
Uncertainty
Comments
Water ingestion 1.4 L/day
rate
Contact rate
0.75 & 1.00
Sediment concentrations
Model
Parameters
Koc, TSS,
foe
The classically assumed
2.0 L/day was evaluated
in EPA (1989) and found
to be high; recommended
1.4 L/day in general
assessments
0.75 recommended in EPA
(1989) for general uses
based on time spent at
home; recommended value
of 1.00 was felt to be
too high for high end
scenarios
SEE Section 10.2.2.
organic carbon partition
coefficient (Koc) estimated
from octanol water partition
coefficient (Kow); total
suspended solids (TSS)
appropriate for river
which supports fish;
fraction organic carbon
(foe) may be low but
reasonable
EPA (1989) also noted that
information on sensitive
subpopulations such as
laborers was unavailable;
estimate might be low,
therefore, for farmers
Not expected to be widely
variable for rural settings.
Koc an estimated rather
than measured value; selected
value for 2,3,7,8-TCDD may
be low; other parameters
are site-specific; overall,
estimations in example
scenarios may have led
to overpredictions by a
factor of 2
Not expected to be
a critical factor for
uncertainty
Same comment as above
Relatively narrow range
of possible values for
all noted parameters;
uncertainty would be
evaluated as low
EVALUATION: Data was unavailable in the literature which linked soil contamination with surface water concentrations. Estimated
concentrations were in the low to sub pg/L (ppq) range, which was consistent with literature values. Modeling exercise using complex WASP4
model applied to Lake Ontario (EPA, 1990a) was duplicated using simplistic approach of this assessment, and results from both exercises were
comparable. Model parameters were judged as reasonable values with narrow range of possible values. Possibly low values of 2,3,7,8-TCDD
Koc and foe may have led to overpredictions of concentrations by a factor of 2.
the exposed individuals, perhaps through a public water system. The contact fraction of
0.75 for central scenarios is based on time use surveys which showed roughly this
fraction of time spent in and around the home environment on the average. The upper
recommended limit in EPA (1989) was 1.00; this was felt to be unrealistic for the example
scenarios which involved relatively small sources and consequently the likelihood that
contamination would not be widespread. Thus, a farmer would likely obtain some water
from outside his home where the water supply was not contaminated. An assumption of
10-33
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
0.90 for farming families was selected for the high end scenarios of this assessment.
In summary, the ingestion rate and contact fractions are justifiable for use in central
and high end scenarios. Variations outside noted ranges are not expected to be great and
any reasonable variation would have limited impact to long term exposure estimates.
10.2.7.2. Comparison of Estimated Water Concentrations With Observed Data
Tables 3-3 and 3-4 (Chapter 3) summarize available data on surface water
concentrations of the PCDDs and PCDFs. The results on this table are not directly
amenable to comparison because the sources of contamination were unspecified except to
note that, in some studies, a portion of the sampling occurred for water bodies known to
be impacted by industrial discharges. No studies were found which both measured and
found surface soil concentrations, suspended sediment concentrations, and water column
concentrations of dioxin-like compounds. Such studies would be necessary to validate the
modeling approach in this assessment. The impact to surface water from industrial
discharges is not discussed in this assessment. Assessments of exposure to 2,3,7,8-
TCDD resulting from paper and pulp mill discharges can be found in EPA (1990b) and EPA
(1991).
Nonetheless, this data does indicate that occurrences of PCDDs and PCDFs are
generally not-detected or in the low pg/L (ppq) range; detection limits were generally at or
near 1 pg/L. The one exception to this is occurrences in tens to hundreds of pg/L range
for PCDFs in one of twenty community water systems sampled in New York (Meyer, et
al., 1989). Concentrations exceeding 200 pg/L were found in the hepta- and octa-CDFs;
concentrations between 2 and 85 pg/L were found in the tetra to hexa-CDFs for this
impacted water system.
The highest water concentration estimated in the demonstration scenarios in this
assessment was 8 pg/L for the ash landfill demonstration scenario. Scenario 6, for
2,3,4,7,8-PCDF. Soil concentrations of this congener for the 27 ha ash landfill were 8
ng/kg (ppb). Water concentrations for 2,3,7,8-TCDD and 2,3,3',4,4',5,5'-HPCB were
both 0.9 pg/L for this scenario, and ash landfill concentrations were 0.7 and 2.0 ppb for
each congener, respectively. Water concentrations of the three demonstration congeners
were 0.2 (2,3,7,8-TCDD), 0.2 (2,3,4,7,8-PCDF), and 0.07 (2,3,3',4,4',5,5'-HPCB) pg/L
range for the example scenario demonstrating off-site soil contamination, Scenario 3; soil
10-34 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
concentrations of the three contaminants were set at 1 ppb for the 4 ha area of off-site
contamination. For Scenarios 1 and 2, where watershed soil concentrations were set at 1
ng/kg (ppt), surface water concentrations were in the 10~3 pg/L (ppq) range. For
Scenarios 4 and 5 demonstrating stack emission depositions and where watershed soil
concentrations were estimated to be in the 10~2 to 10~1 ppt range, surface water
concentrations were the in the 10"4 to 10~3 ppq range.
10.2.7.3. An Alternate Modeling Approach for Estimating Water Concentrations
A study to evaluate bioaccumulation of 2,3,7,8-TCDD in fish in Lake Ontario
included an extensive modeling exercise (EPA, 1990a). The model used was WASP4
(Ambrose, et al., 1988). This is a substantially more complicated model than used in this
assessment. The underlying principal for this model is a conservation of mass.
Contaminant input to a water body, described in mass/time units, enters what are termed
control volumes, or segments. The contaminant entering partitions between sorbed,
bound, and dissolved phases once entering; it is not required to specify whether the
contaminant enters via soil erosion, water runoff, surface deposition, or otherwise.
Contaminants are, however, assumed to enter via the surface or as part of inflows to the
water body, in contrast to ground water recharge. The mass transported into a segment is
either transported out of the segment, accumulates in the segment, or is transformed by
chemical reaction or biological degradation.
As noted, 2,3,7,8-TCDD input into the Lake Ontario application partitions within the
water column into a sorbed compartment, a dissolved compartment, and a bound
compartment. This bound compartment is further described as non-settling organic
matter. Three analogous compartments receive 2,3,7,8-TCDD in the bottom sediment
layer. Several exchanges between the six compartments and contaminant losses within
each compartment are modeled. For example, losses from water column compartments
include downstream transport, volatilization and photolysis; the loss mechanism from the
bottom sediment layer is sedimentation. Exchanges between compartments consider
partitioning, diffusion, and sediment settling and resuspension.
This model requires substantial parameterization. Once values were selected for
the Lake Ontario application, an evaluation was made on the impact of different levels of
2,3,7,8-TCDD input. Dynamic and steady state results were discussed. Principally
10-35 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
examined for the steady state results were the concentrations of bottom sediment sorbed
2,3,7,8-TCDD and water column dissolved (soluble phase) 2,3,7,8-TCDD. A given level
of steady 2,3,7,8-TCDD input, in kg/yr, resulted in a steady state concentration sorbed to
bottom sediment and dissolved in the water column.
The premise in both the Lake Ontario steady state application of WASP4 and the
water concentration algorithms in this assessment is that contaminants continue to enter
water bodies over time unabated. Ground water entry of contaminants is not considered
in either approach. Although a direct modeling comparison cannot be done, it is possible
to slightly adjust the algorithms of this assessment to evaluate how results from a simple
partitioning approach would compare with results from the complex fate and transport
approach of the WASP4 steady state application.
Assume a surface water body is initially free of contaminant and at time t equals 1
day, a strongly hydrophobic contaminant, such as the dioxin-like compounds of this
assessment, begins to enter a lake. Assuming the contaminant enters via soil particles, as
in the approach of this assessment, it will then partition between those soil particles and
surrounding water. The soil particles will slowly move toward the bottom of the lake at a
rate described by a particle settling velocity. A settling velocity of 1 m/day is assumed in
the Lake Ontario simulations. The amount of time it takes to settle to the bottom once
entering from the surface equals the lake depth divided by this settling time. The Lake
Ontario depth was 86 m. Therefore, it might take 86 days to settle. This, of course,
neglects resuspension of settled particulates. With this simplistic framework, a steady
state amount coming into the lake after 86 days is matched by an amount depositing onto
the lake bottom; the amount of contaminant within the water column has reached steady
state. Water concentrations can then be estimated assuming equilibrium partitioning.
Results of sediment and water column steady state concentrations are described for
any loading of 2,3,7,8-TCDD in the WASP4 steady state application; those loadings are
described in kg/yr. Loadings in kg/yr are easily correlated to a steady state water column
amount, given the above analysis. For example, a loading of 1.0 kg/yr could translate to a
within water column steady state amount of 0.24 kg (1.0 kg/yr * (86 d)/(365 d/yr)).
This steady water column amount partitions between suspended sediment and
surrounding water. First, the total concentration (sorbed + soluble) simply equals:
where:
10-36 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
cm = 1000 ^L (lo-i)
Ctot = total concentration, mg/L
LD = water column steady state amount of contaminant, kg
VOL = lake volume, m3
1000 = converts kg to mg and m3 to L
The dissolved phase portion of total is given, as in Equation (5-3), by:
Cwat = - -
TSS 10
where:
Cwat = soluble phase water concentration, mg/L
Ctot = tota' concentration, mg/L
Kdssed = partition coefficient between suspended sediment and surrounding
water, L/kg
Koc*OCssed
Koc = organic carbon partition coefficient, L/kg
OCSsed = fraction organic carbon of suspended sediments
TSS = total suspended sediments, mg/L
10"6 = converts mg/kg to mg/mg
Parameters in this equation for the Lake Ontario WASP4 application include VOL,
Koc, OCssed, and TSS. Lake Ontario volume was given as 1.68 x 1012 m3, Koc was
estimated for the WASP4 application as 3,162,000, OCssed was estimated at 0.03, and
TSS was estimated 1.2 mg/L. For a steady load of 1 kg/yr and a resulting LD of 0.24 kg,
the steady state water column 2,3,7,8-TCDD concentration, using the simplistic approach
described above, is estimated as 0.13 pg/L (ppq). The steady state water column
concentration estimated by WASP4 given the same parameters and a load of 1 kg/yr is
roughly 0.20 pg/L. An uncertainty analysis done with these WASP4 results concluded
10-37 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
that 95% confidence limits around this prediction are 0.03 and 0.40 pg/L.
This would seem to imply that the simple partitioning approach used in this
assessment compares favorably with the more complex fate and transport modeling
assessment using WASP4, for Lake Ontario.
10.2.7.4. Sensitivity to Model Parameters
As noted in the introduction to this section, a key quantity of uncertainty is the
concentration of contaminants in suspended sediments. An uncertainty discussion on this
quantity is given in Section 10.2.2. In summary, that section notes the following: 1) the
premise that suspended sediment concentrations are related to concentrations in soil
within the watershed is a sound one, 2) most data in the literature including sediment
concentrations cited industrial discharges (direct discharges such as from paper or pulp
mills, or air emission discharges) as the source of contamination; there were no data
linking concentrations found to soil concentrations, and 3) sediment concentration
predictions would change by no more than a factor of five, either higher or lower, with
changes to single parameters within a range of reasonable values.
Parameters which are required given suspended sediment contaminant
concentrations include the concentration of suspended sediment, TSS, the organic carbon
partition coefficient, Koc, and the fraction organic carbon in suspended sediments.
All example scenarios assumed that the site of contamination was in a watershed
draining into a river. This justified a TSS of 10 mg/L. Suspended sediment concentrations
in a standing water body such as a lake or a reservoir would likely be lower, around 2
mg/L. River suspended sediments could be higher, which would be the result of more
erosion, turbulence, or bioturbation. The discussion in Section 5.3.1. indicated that a
suspended sediment concentration of 10 mg/L or less indicated no problem with
sustainability of aquatic life. The rivers of the example scenarios were assumed to support
fish for recreational fishing and water for consumption. Water column concentrations are
linearly related to suspended sediment concentrations. Reducing the concentration by 5 to
2 mg/L would reduce water concentrations by 5; increasing it to 20 or 30 mg/L would
increase water concentrations 2-3 times.
The OCssed and Koc parameters measure the sorption of contaminant to suspended
sediments. Increasing the value of these terms tends to decrease water concentrations.
10-38 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
The organic carbon fraction, OCssed, was set at 0.01 (1% organic carbon). This was
lower than the organic carbon fraction measured for sediments of Lake Ontario in the
study evaluating the bioaccumulation of 2,3,7,8-TCDD (EPA, 1990a). Increasing OCssed
to 0.03 decreases concentrations by approximately 30%. Increasing OCssed to 0.05,
which is a reasonable upper estimate for this parameter, reduces concentrations by 46%.
The Koc for 2,3,7,8-TCDD was set at 2,700,000. Evaluation of Koc for sediments of
Lake Ontario led researchers to conclude that it was greater than 106'3 (2,000,000) and
may be as high as 107-3 (20,000,000). Changing Koc to 24,500,000, evaluated as a high
estimate for 2,3,7,8-TCDD Koc, reduces estimated concentrations by 63%. Keeping this
high Koc and increasing OCssed from 0.01 to 0.05 decreases concentrations by an order of
magnitude. This analysis shows that selections of Koc and OCssed for suspended
sediments might result in an overprediction of water column soluble phase concentrations,
but likely by no more than a factor of 2 (i.e., higher Koc and OCssed might reduce
concentrations by no more than 50%).
10.2.8. Fish Ingestion Exposure
Fish lipid concentrations are estimated as the multiplication of a Biota Sediment
Accumulation Factor, BSAF, and an organic carbon normalized concentration of
contaminant on bottom sediment. Whole fish tissue concentrations equals this lipid
concentration times the lipid content of fish. The first section evaluates the rate of fish
ingestion. Then, estimates of fish tissue concentrations made in this assessment are
compared with fish tissue concentrations given in the literature. The third section
evaluates alternate approaches to estimating fish concentrations, and the final section
evaluates the sensitivity and uncertainty in key model parameters. A summary of key
uncertainty issues associated with the fish ingestion pathway is given in Table 10.8.
10.2.8.1. Fish Ingestion Ra tes
Although fish consumption surveys are available and are discussed in EPA (1989),
this assessment uses a different approach to estimate the consumption of fish from an
impacted water body. The approach is recommended for use when site-specific survey or
other information is unavailable (EPA, 1989). Briefly, assume a meal size of between 100
and 200 g/meal - this assessment assumed 1 50 g/meals - and estimate the number of fish
10-39 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.8. Uncertainties associated with the fish ingestion pathway.
Assumption/
Method
Approach
Rationale
Uncertainty
Comments
Fish ingestion 1.2 and 4.1 Based on 3 and 10 meals
rate g/day per year caught from the
impacted water, and
150 g/meal
A low estimate compared
to surveys of recreational
or subsistence fisherman
near large water bodies
Example scenarios were for
"rural",agricultural settings;
higher rates of ingestion
appropriate for other purposes
was not deemed appropriate
for such settings.
Model
Parameters
OC
TSS
-fl
ssed>
llpld,
Koc,
Biota Sediment Accumula-
tion Factors, BSAFs,
based on field measured
values; fish lipid
content, fhpld, of
0.07 used by Cook,
et al. (1991) for
2,3,7,8-TCDDexposure
assessment; fraction
organic carbon in
sediment, OC ssed, may be
low but reasonable. TSS
reasonable for river
setting, and Koc
based on literature
BSAFs for dioxin-like compounds
field measured to be in 0.01-
0.30 except one study measuring
2,3,7,8-TCDDBSAFto be 1.60;
f | |cj variable even for same
species, but values unlikely
to exceed 0.20; OC ssed no
higher than 0.20; TSS
lower for lake setting;
Koc may be higher for
aquatic settings
Relatively narrow range
of possible values for
all noted parameters;
High BSAF from one study
and variable BSAF for PCBs
suggests further refinement
of BSAF assignment; sensitivity
showed narrow range for
fish tissue concentrations
for individual changes
in OC ssed, TSS, and Koc
EVALUATION: Data was unavailable in the literature which linked soil contamination with sediment and fish concentrations; comparison with
literature values of fish concentrations of dioxin-like compounds suggests that concentrations predicted for background and high levels of soil
contamination may be slightly low for dioxins, although predictions are clearly within range, and relative increases from background to high soil
contamination as noted in the literature are mirrored by the model. Much higher sediment concentrations of PCBs and hence fish concentrations, in
comparison to the estimates made in this assessment, were noted. The BSAFs were assigned based on field measurements for dioxins, furans, and
PCBs. PCB BSAFs were an order of magnitude and more higher than dioxin and furan BSAFs; one literature reference for dioxin/fiiran BSAFs is an
order of magnitude higher than others -this should be further investigated. Alternate modeling approaches based on water column concentrations
show higher but comparable fish concentration estimations. Fish concentrations estimations vary by less than an order of magnitude with changes
in model parameters.
meals that may be recreationally caught from the impacted water body. An estimate of 3
meals/year was made for central exposure scenarios, and 10 meals/year was made for
high end exposures. Ingestion of contaminated fish is therefore, estimated as 1.2 and 4.1
g/day, respectively (150 g/meal * 3 meals/yr * 1/(365 d/yr) = 1.2 g/d).
Surveys of recreational fisherman near large water indicates that these estimates
are low for this subgroup. As noted in Chapter 7, EPA (1989) estimates that a typical rate
of ingestion of recreationally caught fish for this subgroup is 30 g/day, with a 90%
estimate of 140 g/day. The range of 30-140 g/day may be more appropriate, therefore, if
estimating fish ingestion exposure for recreational fisherman near a large, impacted water
10-40
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
body. EPA (1989) also reports on two surveys of per capita fish consumption, on a
national basis and including all consumption, not just recreationally caught consumption.
The mean and 95% consumption rates determined by Javitz (1980) were 14.3 and 41.7
g/day. The median and 95% consumption rates determined by Pao, et al. (1982) were 37
and 128 g/day. If using either of these estimates in exposure exercises, assumptions on
percent of total consumption which is recreationally caught and/or impacted by dioxin-like
compounds needs to be made.
A key trend noted for the example scenarios in Chapter 9 is that fish, along with
beef and milk ingestion, led to the highest exposure estimates for the dioxin-like
compounds. Obtaining site-specific information for fish ingestion is critical for this
pathway. The ingestion rates made in this assessment are very likely low by an order of
magnitude or more for use to a subgroup of recreational fisherman obtaining fish from a
nearby large water body.
10.2.8.2. Comparison of Estimated Fish Concentration With Literature Values
This assessment estimated fish concentrations of 2,3,7,8-TCDD for the various
source categories to be: 1) ash landfill - 5 ppt, 2) off-site soil contamination - 1 ppt, 3) on-
site soil concentration - 0.07 ppt, and 4) stack emissions - 0.002 ppt. This section will
examine some available data on fish concentrations in order to compare these results with
measured results.
The most appropriate study with which to make comparisons is the National
Bioaccumulation Study (see Chapter 3, Section 3.5 for reference and more detail;
abbreviated NBS). Fish tissue data on a variety of species and contaminants of concern in
aquatic environments and fish from around the country were developed. Most important
for current purposes, the sites were carefully characterized in terms of potential sources of
fish contamination. There were 353 sites from which fish tissue data was available, of
which 347 had data on 2,3,7,8-TCDD. Results from four site categories might be
appropriate for comparison with concentrations estimated to occur from low, possibly
background, soil concentrations of 2,3,7,8-TCDD. The four categories and number of
sites per category were: the USGS water quality network NASQAN - 40 sites; Background
(B) - 34 sites. Agricultural (A) - 1 7 sites, and Publicly Owned Treatment Works (POTW) - 8
sites. The average 2,3,7,8-TCDD concentrations measured for these four categories were:
10-41 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
NASQAN - 1.02 ppt; B - 0.56 ppt; A - 0.75 ppt, and POTW - 0.90 ppt. The "on-site"
source category was demonstrated in Chapter 9 using a soil concentration of 2,3,7,8-
TCDD that might be typical of background conditions - 1 ng/kg (ppt). The resulting fish
tissue concentrations estimated for this soil concentration was 0.07 ppt. Four of the site
categories of the NBS might be considered representative of sources characterized as land
areas of high soil concentrations of 2,3,7,8-TCDD. These were: Industrial/Urban site
(IND/URB) - 105 sites, Refinery/Other Industry (R/l) - 20 sites, Wood Preservers (WP) - 11
sites, and Superfund Sites (NPL) - 7 sites. Average fish tissue concentrations measured
for these site categories were: IND/URB - 4.04 ppt, R/l - 4.38 ppt, WP - 1.40 ppt, and NPL
- 30.02 ppt. Two of the source categories of this assessment are similarly characterized -
land areas with elevated levels of 2,3,7,8-TCDD. These were the off-site category,
demonstrated with 1 //g/kg (ppb) concentrations of 2,3,7,8-TCDD, and the ash landfill
source category, demonstrated assuming 0.7 ppb 2,3,7,8-TCDD. As noted above, the fish
tissue concentrations estimated with these source categories were 1.0 (off-site) and 5.0
ppt (ash landfill). The two remaining site categories of the NBS were Paper Mills Using
Chlorine (PPC), and Other Paper Mills (PPNC). Fish tissue concentrations with these
sources were 19.02 and 2.17 ppt, respectively. A discussion of effluent discharge
modeling is given in Appendix D of this assessment.
In general, the range of fish tissue concentrations measured for (perhaps)
background conditions, 0.56 - 1.02 ppt, were somewhat higher than the fish tissue
concentration estimating assuming the low (perhaps) background soil concentration of 1
ppt soil concentration, 0.07 ppt. It may also be appropriate to make the same observation
for the source categories assuming higher soil concentrations as compared to measured
concentrations noted above. In this case, the range of average measured concentrations -
1.4 - 30.02 ppt compares with the modeled 1.0 - 5.0 ppt. Obviously, it would be
inappropriate to conclude that the model underestimates fish tissue concentrations since
there is no information on the source strengths (or other information for environmental fate
modeling) for the measured tissue concentrations. It is probably more appropriate to view
this comparison as one that tends to lend some credibility to perhaps both the selection of
source strengths and modeling approaches used in this assessment. The difference in
measured background concentrations versus those measured for industrial-like sites is
mirrored by the model. The actual concentrations predicted are within range of the
10-42 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
measured concentrations. Different assumptions on the source strength terms (soil
concentrations, erosion amounts) and different assumptions on water column partitioning
(as modeled with partition coefficients, suspended solids concentrations, and fractions of
organic carbon in suspended solids and bottom sediments) would increase modeled fish
concentrations. The impact of the partition coefficient is discussed in Section 10.2.8.4
below.
Another comprehensive data base of fish concentrations of 2,3,7,8-TCDD is from
EPA's National Dioxin Study (1987; abbreviated the NDS). Fish concentrations from that
study are also listed and discussed in Kuehl, et al. (1989). Travis and Hattermer-Frey
(1991) summarized the fish data from the NDS, and that summary is reiterated in Chapter
3, Section 3.5. Briefly, data collected from 304 urban sites in the vicinity of population
centers or areas with known commercial fishing activity, including the Great Lakes Region,
showed concentrations to range from non-detected to 85 ng/kg (ppt). The geometric
mean concentration was 0.3 ppt, and only 29% had detectable levels of 2,3,7,8-TCDD.
The Great Lakes data had more contamination, with 80% detection rate and a geometric
mean concentration of 3.8 ppt. The fish tissue concentrations estimated for 2,3,7,8-
TCDD for the two categories of off-site soil contamination - the ash landfill at 5 ppt and
off-site soil at 1 ppt - are comparable to the two geometric means for the NDS, 0.3 and
3.8 ppt.
While the data from the NDS targeted urban areas where higher fish concentrations
can be expected, other data with known point source inputs had higher concentrations of
2,3,7,8-TCDD and related compounds. In evaluating the impact of paper and pulp mills,
EPA (1991) used specific data from the NDS. This data was for four species of fish
(composites of each species) from the Wisconsin River, and samples were taken
approximately two miles from a mill discharging into the river. The estimated average
whole body concentration of 2,3,7,8-TCDD for these fish was 33 ppt (estimated because
3 of the 4 fish species were sampled for fillet only; whole fish concentrations estimated at
twice the fillet concentration). Pike sampled from Lake Vanern in Sweden had whole fish
average concentrations of 41 ppt of 2,3,7,8-TCDD with a high concentration of 150 ppt
(Kjeller, et al., 1990).
The National Bioaccumulation Study (NBS) also collected data on 2,3,4,7,8-PCDF,
the second example compound demonstrated. Briefly, the range of average fish tissue
10-43 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
concentrations noted for the site categories evaluated as background above is 0.42-0.78
ppt, very similar to the 2,3,7,8-TCDD range of 0.56-1.02 ppt. The modeled fish tissue
concentration of 2,3,4,7,8-PCDF for background conditions was 0.09 ppt, slightly higher
than the 2,3,7,8-TCDD modeled concentration, 0.07 ppt, because of different partitioning
in the water body. The range of 2,3,4,7,8-PCDF average fish concentrations for the NBS
sites of elevated soil concentration was 1.86-5.44 ppt, which compares to the modeled
2,3,4,7,8-PCDF concentrations of 2 (off-site source category with initial soil
concentrations of 1.0 ppt 2,3,4,7,8-PCDF) and 80 ppt (ash landfill with initial soil
concentrations of 8.2 ppt).
The NBS also collected data on PCB concentrations in fish, although the results
were expressed in terms of total tetra-, hepta-, and so on. The data indicates
concentrations well into the part per billion range for this breakout, and even higher
considering total PCBs. The average concentration of total heptachlorobiphenyls over all
NBS sites was 96.7 /vg/kg (ppb). The average concentration of total PCBs over all sites
was estimated as 1897.88 ppb, and the average concentration of total PCBs for
background sites was 46.9 ppb. The modeled concentration of the example
heptachlorobiphenyl, 2,3,3',4,4',5,5'-HPCB for the background scenario (soil
concentration = 1.0 ppt) was 3 ppt. The modeled concentration of 2,3,3',4,4',5,5'-HPCB
for the off-site scenario (soil concentration = 1 ppb) and the ash landfill (soil concentration
= 2.2 ppb) was 40 and 600 ppt, respectively.
Data from the Great Lakes region indicate that PCB concentrations are significantly
higher than PCDD/PCDF concentrations in this area. PCB concentrations from fish in Lake
Ontario are in the tens to hundreds of ppb level (Niimi and Oliver, 1989), while 2,3,7,8-
TCDD contamination in Lake Ontario was in the tens of ppt level (EPA, 1990a) - a three
order of magnitude difference. Other data in Table B.10 (Appendix B), where
concentrations were similarly in the tens to hundreds of ppb level were from Lake
Michigan (Smith, et at. 1990) and Waukegan Harbor in Illinois (Huckins, et al. 1988). The
single data point from that table for 2,3,3',4,4',5,5'-HPCB, the example PCB congener in
Chapter 9, was for carp in Lake Michigan, and was 29 ppb (29,000 ppt).
While the modeled PCDD/PCDF fish concentrations seem reasonably in line with
measured concentrations, this assessment may have underestimated concentrations of
2,3,3',4,4',5,5'-HPCB. As noted, concentrations for fish in the Great Lakes Region were
10-44 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
in the tens to hundreds of ppb range, while this assessment derived estimates all under 1
ppb. It is inappropriate to make direct comparisons without also comparing source
strengths. Concentrations of PCBs in bottom sediments ranged from the low ppb for the
tri-PCBs, to the tens of ppb for the tetra through hexa-PCBs, back to the low ppb for the
hepta and octa-PCBs, in Lake Ontario (Oliver and Niimi, 1988). Another literature source
showing fish concentrations in Waukegan Harbor, IL, in the hundreds of ppb range, had
sediment concentrations of specific congeners as low as 5 ppb to as high as 131 ppm.
The concentration of 2,3,3',4,4',5,5'-HPCB in bottom sediments never reached 1 ppb.
Therefore, one reason PCB concentrations in fish estimated in this assessment are as
much as three orders of magnitude lower than noted in the literature is because sediment
concentrations estimated for the source categories in this assessment are three orders of
magnitude lower. The BSAF for PCBs also was noted to be variable, with values below
1.0 to values over 20.0 (see Section 5.4.2). The BSAF for the example PCB congener in
this assessment was 2.0. Higher BSAFs would also increase PCB concentrations
estimated for fish.
10.2.8.3. Alternate Modeling Approaches
EPA is currently preparing a document titled, "Interim Aquatic Ecological Risk
Characterization for 2,3,7,8-TCDD" (P. Cook, personal communication, Duluth
Environmental Research Laboratory, US EPA, Duluth, MN). A draft version of an exposure
and bioaccumulation chapter was obtained for purposes of this section.
Alternate measures of the potential for accumulation of contaminants in fish based
on water column and bottom sediment (i.e., the BSAF approach) contaminant
concentrations were discussed in that chapter. One water column measure which has
been classically used is termed the Bioconcentration Factor, or BCF. Bioconcentration
refers to the net accumulation of a chemical from exposure via water only, and BCFs are
most often obtained in laboratory conditions. BCFs are defined as the ratio of the chemical
concentration in organism (mass of chemical divided by wet weight of organism tissue) to
that in water.
Another water column measure of the potential for a contaminant to accumulate in
fish tissue is termed the Bioaccumulation Factor, or BAF. Bioaccumulation refers to the
net accumulation of a chemical from exposure via food and sediments as well as water.
10-45 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Similar to the BCF, BAFs are defined as the ratio of the chemical concentration in the
organism to that in the water.
For chemicals that are not strongly hydrophobic (unlike the dioxin-like compounds),
the distinction between bioconcentration and bioaccumulation is small. Whereas food
intake is generally a few percent of body weight per day, water passing over gills will
equal hundreds to thousands times the organism weight per day, depending on species,
activity, temperature, and other factors. Given this, the concentration of chemical in food
must be 3 or more orders of magnitude greater in food than in water before food can
substantially contribute to uptake. Cook (reference above) estimates food intake becomes
a critical contributor to the accumulation of contaminants in fish tissue for contaminants
with log Kow of 5 and greater.
Since the dioxin-like compounds fall into this category, the remainder of this section
will focus on the Bioaccumulation Factor. Cook defines steady-state lipid-based BAFs for
total chemical in water and freely dissolved chemical in water (i.e., chemical which is truly
in a dissolved phase and not bound to dissolved organic materials) as:
(10-33)
(10-3b)
where:
ssBAFi* = steady-state lipid-based BAF for total chemical in water, unitless
CH jd = the mass of contaminant in fish lipid tissue divided by the mass of
fish lipid tissue, mg/kg
Cw* = the mass of total contaminant in water divided by the mass of water
in the water body, mg/kg (note: 1 L water = 1 kg)
ssBAF|d = steady-state lipid-based BAF for freely dissolved chemical in water,
unitless
Cwd = the mass of freely dissolved contaminant in water divided by the
mass of water in the water body, mg/kg
10-46 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Cook then develops these steady-state lipid-based BAFs for lake trout, 2,3,7,8-
TCDD, and for Lake Ontario 1987 contamination conditions. The WASP4 model
(Ambrose, et al., 1988) was used to model two loading conditions to the lake: steady
state loading and no loading of 2,3,7,8-TCDD. The BSAF for lake trout estimated for
1987 conditions is given in EPA (1990a) as 0.07. Details of the Lake Ontario study can
be found in EPA (1990a). The BAFs Cook determined will be tested using the models of
this assessment.
In order to do this exercise, all critical model parameters used to develop the BAFs
for Lake Ontario will be used in the model framework of this assessment. The most
critical parameter is the organic carbon partition coefficients, Koc, assumed for 2,3,7,8-
TCDD. BAFs were determined assuming Koc of 107 and 108. Since the models of this
assessment assume steady loading into water bodies (via erosion of soil from the
watershed), only the BAFs developed under "steady state" loading conditions will be used.
Two compartments onto which 2,3,7,8-TCDD sorbs in the water column were simulated
using the WASP4 model - these include the suspended solids and what was termed the
water column non-settling organic matter (NSOM). The concentration of suspended solids
was 1.2 mg/L, and the concentration of the NSOM was 3 mg/L. The framework of this
assessment has only one compartment termed suspended solids; the value given this term
for this exercise will be 4.2 mg/L, the sum of concentrations of the two compartments of
the WASP4 exercise. The other critical parameters are the fraction organic carbon
contents of the suspended solids and the bottom sediments, OCssed and OCsed,
respectively. Assigned values to these parameters in the WASP4 exercise (and in this
exercise) were 0.12 (12%) and 0.03 (3%), respectively. One final note is that there is no
analogous NSOM compartment in this model. This compartment effectively removes
contaminant from the water column - it cannot partition into the dissolved phase. EPA
(1990a) estimated that the fraction of total water column 2,3,7,8-TCDD to be unavailable
by this binding is 6.7% of total water column 2,3,7,8-TCDD. Because the model of this
assessment does not have a similar binding compartment, it may generally tend to predict
higher dissolved phase concentrations than the WASP4 model.
Since the purpose of this exercise is to evaluate how the modeling approaches of
this document perform using the BSAF or the alternate BAF approach, it is not critical
which source category is demonstrated. Any source category will estimate a given load to
10-47 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
a water body - the question is how the BSAF approach predicts fish tissue concentrations
as compared to BAF approach, whatever load enters the water body. For simplicity, the
on-site source category as demonstrated in Chapter 9 will be used. In this scenario, the
soil within the watershed is assumed to uniformly be 1.0 ppt.
In summary, the parameters for this exercise including the BAFs are:
Test 1: Koc = 10,000,000; ssBAF,d = 4,040,000; ssBAF,' = 866,000; BSAF = 0.07
Test 2: Koc = 100,000,000; ssBAF,d = 40,400,000; ssBAF,' = 1,080,000; BSAF =
0.07
For both tests: soil concentration of 2,3,7,8-TCDD = 1.0 ng/kg (ppt), total suspended
solids (TSS) = 4.2 mg/L, the organic carbon content of suspended sediments (OCssed) =
0.12, and the organic carbon content of bottom sediments (OCsed) = 0.03
The whole fish tissue concentration (i.e., C|ipid multiplied by the fraction lipid, flipid,
assumed to be 0.07) for the BSAF approach using Koc = 107 was estimated to be 0.03
ppt. Using the ssBAF^ and ssBAF|d, the whole fish tissue concentrations were estimated
to be 0.25 and 0.20 ppt, respectively. The test results for Koc = 108 were similar; the
BSAF estimated fish tissue concentrations to be 0.04 ppt, and the ssBAF,* and ssBAF|d
predicted concentrations of 0.32 and 0.23 ppt.
While it appears that the water column based approaches estimate fish tissue
concentrations within an order of magnitude higher than the sediment approach, it would
not be appropriate to conclude that as a generalization. Rather, the predictions are
reasonably close, given the uncertainties in the bioaccumulation and water modeling
parameters. One important consideration in using the water column based approaches is
that the BAFs developed by Cook (or that could be developed otherwise) are based on
modeled rather than measured water column concentrations, and measured lake trout
tissue concentrations. In that sense, the BAFs were calibrated for Lake Ontario conditions
and specific to the WASP4 modeling exercise. The BSAF developed for lake trout for Lake
Ontario was developed using measurements of both fish tissue and bottom sediment
concentrations.
Both the BSAF and BAF are most appropriately developed using site specific data
(coupled with a modeling exercise for BAF). Inasmuch as that can be impractical or
difficult for many sites, efforts are underway to determine the general applicability of
BSAFs and BAFs determined for one site to other sites. Cook (reference given at
10-48 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
beginning of this section) writes that BAF|S for different congeners can be roughly
estimated as the BAP, for 2,3,7,8-TCDD multiplied by the ratio of the BSAF for the
congener and the BSAF for 2,3,7,8-TCDD. Such an estimate will incorporate differences
in uptake, metabolism and chemical partitioning but not differences caused by chemical
loss processes such as volatilization and photolysis. This approach for estimating BAF|S
for other congeners does allow for some generality since sediment and fish tissue data for
other congeners and water bodies is available. Cook also suggests that the ssBAF|d may
be the best predictor of 2,3,7,8-TCDD residues in other systems ]f (sic) the dissolved
phase concentrations can be estimated accurately. A comprehensive discussion of the
many issues involved in exposure and bioaccumulation will be available in the document
cited at the beginning of this section.
10.2.8.4. Sensitivity to Key Model Parameters
Fish tissue concentrations are based on organic carbon normalized concentrations in
sediments. Strictly speaking, they are a function of the organic carbon normalized
concentration on bottom sediments. However, one key assumption made for the solution
for bottom sediment concentrations is that the organic carbon normalized concentrations
of bottom sediments and suspended solids are equal. Soil eroding into the water body
from the watershed is assumed to equilibrate with the water in the water body while
suspended. Therefore, since the organic carbon normalized concentrations of bottom
sediments and suspended solids are assumed to be equal, fish tissue concentrations are a
function of partitioning in the water column.
Three model parameters determine water column partitioning: the total suspended
solids concentration, TSS, the organic carbon partition coefficient, Koc, and the fraction
organic carbon fraction of suspended solids, OCssed. Sensitivity analysis exercises on
these parameters individually showed a reasonably narrow range of predicted fish
concentrations.
The value of Koc used in this assessment for 2,3,7,8-TCDD was 2.7 x 106. Cook
(P. Cook, Duluth Environmental Research Laboratory, US EPA, Duluth, MN) suggests that
the apparent Koc in aquatic systems may be higher at 107 or even 108. The 2,3,7,8-
TCDD was increased to these two values and the predicted fish concentrations increased
46 and 72% respectively. Some literature places Koc for 2,3,7,8-TCDD lower at 106; this
10-49 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
value results in a 42% reduction from the value of Koc used. There is less than an order
of magnitude difference in fish tissue concentration predictions between 106 and 108,
indicating little sensitivity to this parameter.
The same trend of little sensitivity was noted for the other two parameters, TSS
and OCssed. An earlier discussion on total suspended solids, TSS, indicated a reasonable
range of 2 to 20 mg/L (see section 10.2.7.4); the value in this assessment was described
as reasonable for a river setting at 10 mg/L. Changing the value to 2 and up to 20
resulted in a 63% decrease and a 28% increase in estimated fish tissue concentration.
Like Koc, there was less than an order of magnitude difference between 2 and 20 mg/L
estimates. The organic carbon content of suspended sediments, OCssed, should be higher
than bottom sediment organic carbon content, OCsed; OCssed was set at 0.05. It will
rarely exceed 0.15, and setting it to this value decreased concentrations by 58%.
The reason that fish concentrations did not vary with changes in these parameters
is that the Koc is very high to begin with. Changes to any of the three parameters noted
above will only result in slightly more or slightly less 2,3,7,8-TCDD partitioning into the
water column from suspended solids. More sensitivity would undoubtably be shown with
this modeling framework and Koc below 105.
The two remaining parameters used to estimate fish tissue concentrations are the
Biota Sediment Accumulation Factor, BSAF, and the lipid content of fish, fl|pld. Fish
concentrations are linearly related to these two parameters, so there is a linear response to
predictions with changes to these parameters.
The BSAF was assigned to the example compounds on the basis of literature
values. Available values are summarized on Table 5.2 and discussed in Section 5.3.4.1.
Available BSAFs for CDDs and CDFs from 3 of 4 studies were consistent and were in the
0.01 to 0.30 range, with some indications that BSAFs decreased with increasing
chlorination. One study on lake and river contamination in Sweden, including data on
sediment and lipid-based fish tissue concentrations for Pike, allowed for estimation of a
BSAF for 2,3,7,8-TCDD to be 1.6 (see further discussion in Section 5.3.4.1. on this data
point). Changes in BSAF lead to direct changes in fish tissue concentrations. The value
selected for 2,3,7,8-TCDD and 2,3,4,7,8-PCDF was 0.09; increasing this value three-fold,
which is near the maximum reported for all but this one data point, would increase fish
tissue concentrations three-fold.
10-50 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 5.2. and the discussion in Section 5.3.4.1. conclude that the data on BSAFs
for PCBs are more variable. Data from a remote lake in Lake Superior (Swackhamer, et al.,
1988) had no BSAFs less than 2.2 for lake trout and whitefish for penta- through octa-
PCBs, with one BSAF estimated at 20.8. The one BSAF for hepta-PCBs from this study
was 12.5. In contrast, data from New Bedford Harbor showed a lower range for hepta-
PCBs at 0.84 to 2.74 for flounder, lobster and crab. Total PCB BSAFs for assorted Lake
Ontario data, developed and summarized in EPA (1990a), showed a range of 0.58-4.06.
The BSAF selected for 2,3,3',4,4',5,5'-PCB was 2.0. Increasing it by 5, which might
mirror the highest found BSAFs for PCBs, would increase fish concentrations five-fold.
Fish lipid concentrations are also variable, even within the same fish species. EPA
(1990a) found that fish from Lake Ontario had three times the lipid content as the same
species of fish brought into the laboratory from Lake Ontario for microcosm study. A brief
summary of some available lipid contents of fish given in Section 5.4.2 showed that f|ipid
would not be much larger than 0.20. The default value assigned was 0.07. Whole fish
concentrations, which are used in exposure estimates, would roughly triple if using 0.20
instead of 0.07 for f|ipid
10.2.9. Vapor Phase Inhalation Exposures
This section will address the uncertainty associated with vapor phase inhalation
exposures. Sources addressed in this assessment include incinerator stack emissions and
contaminated soils; this section will only address contaminated soils. Uncertainties
associated with vapor phase air concentrations resulting from stack emissions are
discussed in Section 10.3.
Vapor-phase emissions are estimated with a volatilization flux algorithm. The
procedures were developed in Hwang, et al. (1986). A near-field dispersion model
estimates air concentrations for the on-site source category - the category addressing soil
contamination at the site of exposure. For the off-site and ash landfill source categories,
where the site of contamination is located distant from the site of exposure, the same
volatilization flux model is used. Exposure site concentrations for these sources are
estimated using a far-field dispersion model.
The respiration rate of 20 m3/day used for inhalation exposures is within the
standard range of 20-23 m3/day. The contact fraction is 0.75 for central scenarios and
10-51 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
0.90 for high end scenarios. Like the water ingestion contact fractions, these were based
on time at home surveys. The inhalation rate and contact fractions are not expected to
introduce much uncertainty into inhalation exposure estimates. Another exposure
parameter critical for the inhalation pathway is exposure durations, which is 9 years for
central and 20 years for high end exposures. The uncertainties associated with this
parameter are discussed above in Section 10.2.3. However, exposure duration is
additionally critical for the inhalation pathway, as estimated volatilization flux is a function
of the time during which volatilization is occurring; this is further explained below.
The first section below compares measured air concentrations to concentrations
predicted in this assessment. The next section looks at alternate modeling approaches and
considerations. The final section evaluates the impact and uncertainty of key factors that
estimate vapor phase air concentrations. Uncertainty and model sensitivity issues are
summarized in Table 10.9.
10.2.9.1. Comparison of Model Results with Measured Concentrations
The volatilization and near-field dispersion models were developed from well
established theoretical principals, and were developed as part of an effort to assess the
impact of soils contaminated with PCBs (Hwang, et al., 1986). The virtual point source
dispersion model for far-field dispersion estimates is also based on well established theory
(Turner, 1970). However, these models have not been field validated for soils
contaminated with dioxin-like compounds. Ideally, the overall process would be validated
using concentrations of dioxin-like compounds in soil and concurrently measured
concentrations in air above and downwind of the contaminated soil.
However, some sense of the reasonableness of model values can be made by
comparing predictions to measured concentrations in ambient air in urban environments.
Reports of such concentrations are summarized in Section 3.7 (Chapter 3) and Table B.13
«
(Appendix B). Sources other than soil are likely to be the cause of levels measured in urban
air environments; sources such as industrial and automobile emissions. Also, the
dispersion model is designed for situations where the contaminated soil is surrounded by
relatively clean soil. As discussed in Section 10.2.1. above, off-site impacts have been
noted in several sites of 2,3,7,8-TCDD contamination. These two points are made in order
to establish a basis for comparing urban air concentration to concentrations predicted to
10-52 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.9. Uncertainties and sensitivities associated with estimating vapor-phase air concentrations
from contaminated soils.
Assumption/
Method
Approach
Rationale
Uncertainty
Comments
Site-specific
physical
parameters
Chemical parameters
porosity, wind speed
areas of contamination,
distance to receptor,
frequencey wind blows
to receptor; height of
individual
Organic carbon
partition coefficient,
Koc, Henry's Constant,
H
All such parameters
assigned reasonable
midrange values.
Koc empirically esti-
mated from Kow; H
estimated from water
solubility and vapor
pressure
With the exception
of distance to recep-
tor, all parameters
show no more than 50%
differences in estimated
air concentrations with
changes within a range
of expected values;
Field data implies
Koc could be an order
of magnitude higher
than assumed in this
assessment for 2,3,7,8-
TCDD; H consistent
with literature values;
alternate values tend
to decrease estimated
air concentrations by
up to 75%.
Physical parameters
do not appear to
limit certainty of
of estimations.
Consideration should be
given to Koc selected
for 2,3,7,8-TCDD;
alternate value may be
higher but not lower.
Hwang vs. Jury
volatilization
model
selected Hwang model
solutions; Hwang model
includes development
of average flux term
based on time; both
Both analytical
comparable volatilization
flux; Hwang model based
on tentative assumption
of an infinite reser-
models require similar
data; assumptions and
boundary conditions
different
Both models estimate
be appropriate.
voir; but analysis showed
this to be tenable for
short exposure durations
Either model would
Near field vs.
box-model dis-
persion.
selected near-field
dispersion model
required parameters
are less uncertain for
near-field model
Both models estimated Selected model is
comparable concentrations; preferable.
box model showed particular
sensitivity to area and
height of mixing.
Volatilization
or eroded
contaminants
contaminants eroding
to exposure site
assumed not to
volatilize and contri-
bute to exposure site
air concentrations
Uncertainty in depth
of mixing at site of
exposure implies that
volatilization of
delivered contaminated
soil could an order of
higher to an order of
magnitude lower than
volatilization origi-
nating at the site
of contamination.
As noted, the key uncer-
tainty is in the fate
of delivered residues
No change in approach
is warranted.
(continued on next page)
10-53
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.9. (cont'd)
Evaluation: Comparison of model estimations of vapor-phase 2,3,7,8-TCDD and 2,3,4,7,8-PCDF with observed urban air concentrations of
these contaminants supports but does not validate the approaches taken. Sensitivity analysis showed estimations to be relatively insensitive to
physical parameters but sensitive to Koc and H. Data suggests that Koc may be underestimated with selected empirical approach. Alternate
approaches estimate comparable air concentrations, which lends credibility (but not validity) to models. Approaches to estimate particulate
phase concentrations are empirical and based on field data. Evaluation of parameters indicates that predictions could change by up to an order
of magnitude with reasonable parameter changes.
occur from soils: one would expect urban air concentrations to be at least higher, if not
higher by orders of magnitude, than soil emissions.
Tables B.13 and B.14 summarize average PCDD/PCDF congener-specific
concentrations in urban air in the United States and in Europe. Results for two example
compounds demonstrated in Chapter 9, 2,3,7,8-TCDD and 2,3,4,7,8-PCDF, are examined
in this section. Observed concentrations of 2,3,7,8-TCDD were mostly non-detects with
detection limits listed at 0.002-0.03 pg/m3, with two occurrences at 0.01 and 0.05
pg/m3. Concentrations of 2,3,4,7,8-PCDF were detected in all reported studies. The
range of 2,3,4,7,8-PCDF concentrations was 0.01-1.03 pg/m3. In the two reports where
both compounds were detected, 2,3,4,7,8-PCDF was detected at 5 and 10 times higher
concentration than 2,3,7,8-TCDD.
The on-site source category was demonstrated using concentrations of 1.0 ng/kg
(ppt) for each example compound. This low concentration was assigned based on reports
by researchers who measured concentrations of dioxin-like compounds in what they
described as "background" and "rural" soils - they found non-detects to concentrations in
the low ppt level. Modeled air concentrations of the example compounds 2,3,7,8-TCDD
and 2,3,4,7,8-PCDF resulting from this level in soil were in the 10"5 pg/m3 range, or over
three orders of magnitude lower than observed urban concentrations (5 orders of
magnitude lower than the high noted observation of 1 pg/m3 2,3,4,7,8-PCDF). Two
source categories, the off-site and ash landfill categories, evaluated the impact of elevated
soil concentrations in areas that were located distant from the from the site of
contamination. The example scenarios demonstrating these source categories had
concentrations of this dioxin and furan congener in the ppb range, or three orders of
magnitude higher than the on-site source category concentrations. Air concentrations
10-54 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
predicted in these example scenarios were in the 10~3 to 10~2 pg/m3 range, which is
generally the same order of magnitude as these urban findings.
As noted, this was not meant to be a validation exercise, merely an attempt to put
the predicted concentrations into perspective. The fact that concentrations predicted to
occur from very low soil concentrations are 3 orders of magnitude lower than measured
urban concentrations supports the models approach. It is at least plausible that elevated
concentrations in soil would result in air concentrations that are in the same range as
found in urban environments. A model result that would have questioned the model
validity would have been, for example, that air concentrations resulting from soils of very
low concentrations would match urban air concentrations, and that concentrations
resulting from soils with elevated concentrations exceed those found in urban
environments.
10.2.9.2. Other Modeling Approaches and Considerations for Vapor-Phase
Emissions
Volatilization flux was modeled using an approach given in Hwang, et al. (1986),
developed for PCB flux from soils. Principal assumptions for their derivation were that
contamination extended indefinitely, biodegradation or other degradation processes were
not considered, residues were in equilibrium between soil and soil air, and vertical
movement was through vapor phase diffusion. Their analytical solution was integrated
over time and a solution was presented which gave average unit flux as a function of time
during which volatilization occurs. PCBs and other dioxin-like compounds resist
degradation, although there is evidence of photodegradation, which may influence surficial
residues. These compounds sorb tightly to soil, so that an assumption of vertical
movement primarily through vapor phase diffusion (rather than in a soluble phase with
leaching, runoff, or evaporating water) is a tenable one. Also, presentation of an average
flux rate solution made Hwang's approach amenable to spreadsheet analysis, the computer
software tool used in this assessment.
An alternate model for estimating volatilization flux was presented in Jury, et al.
(1983). It is a generalized analytical solution which assumes equilibrium between the
sorbed, soluble, and vapor phases. It incorporates considerations of steady state water
fluxes and degradation mechanisms. A depth over which contamination occurs is
10-55 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
specified. A computer code of this model was obtained from the author (William A. Jury,
Professor and Chair, Department of Soil and Environmental Sciences, University of
California, Riverside, 92521-0424). Tests were run holding all pertinent quantities the
same with both models including initial concentrations, organic carbon partition
coefficients, Henry's Constant, molecular diffusivity, fraction organic carbon in soil, soil
bulk density, porosity, and an assumption of contaminant non-degradation. All of these
parameters, the contaminant as well as the physical parameters, were the ones assumed
for 2,3,7,8-TCDD and the surface soils of this assessment. In applying Jury's model, the
depth of contamination was assumed to be 10 cm. Also, Jury's model allowed for a
selection of water flux to be 0.5 cm/day (heavy leaching), -0.5 cm/day (heavy
evapotranspiration), or 0.0 cm/day (no water flux). The latter selection of no water flux
was chosen. This model comparison test showed that the Hwang model predicted an
average flux over 10 years roughly three times higher than the average flux predicted by
the Jury model over the same time period. Running both models over 50 years showed
similar results. The average flux over that time dropped by about 50% for both models
and there was still a three-fold difference in predicted volatilization fluxes. The exact
reason for this three-fold difference was not investigated, and could lie in differences in
assumed boundary conditions (Hwang, et al. (1986) discusses differences in boundary
conditions between his and Jury's models). In any case, it is judged that both models
predict comparable volatilization fluxes. The Hwang model might be considered
conservative in that it predicts 3 times higher volatilization flux (with 2,3,7,8-TCDD
parameters, etc.).
The Jury model also provides other informative results. It provides a mass balance
which, for the 50-year test, showed that only 2.6% percent of the original mass within
the 10-cm layer had volatilized. By implication, the Hwang model predicts a 7.3% loss by
volatilization over that time period. With the other parameters and assumptions - no
degradation and tight sorption to soil - the Jury model showed that 97.4% remained in the
profile and that only a minute quantity diffused below 10 cm. Also, the Jury model gives
a concentration profile over time. After 50 years, it showed that all volatilization loss was
contained within the upper 2 cm of soil profile. This implies that the boundary condition
assumption for the Hwang model, that contamination extends indefinitely, is not
consequential for the dioxin-like compounds.
10-56 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
A near-field dispersion model is used to estimate air concentrations resulting from
soil volatilization, for the on-site source category (where contamination and exposure occur
at the same site). An alternate approach to estimating on-site dispersion given a
volatilization flux is the "box-model" approach. This simple approach can be visualized as
follows: air above soil is contained within a structure which has two walls, say a north and
south wall, and a ceiling - wind blows through the building in an east-west direction mixing
the volatilized flux. This is expressed mathematically (using similar parameter naming as in
Equation (5-11), Section 5.3.2 in Chapter 5) as:
FLUX AREA 106
•va
*> "mix
(10-4)
where:
va
FLUX
AREA
b
Umix
z
106
vapor-phase concentration of contaminant in air, //g/m3
average volatilization flux rate of contaminant from soil, g/cm2-sec
area over which flux occurs, cm2
side length perpendicular to wind direction, m
mean annual wind speed corresponding to mixing zone height, m/sec;
estimated as 1/2*Um, where Um is average wind speed
mixing zone height, m
converts g to JJQ
Before testing the box-model equation, results for the approach used in this
assessment are summarized. The key factors impacting air concentration calculations in
Scenarios 1 and 2 is the duration of exposure and area over which contamination occurs.
In the central scenario. Scenario 1, the area was 4,000 m2 (1 acre) and in the high end
scenario. Scenario 3, the area was 40,000 m2 (10 acres). The exposure duration was 9
years in Scenario 1 and 20 years in Scenario 2. The volatilization flux was different for
both scenarios, but not because of area considerations, but because of exposure duration
assumptions; the average flux of 2,3,7,8-TCDD for the high end scenario was 1.2x10~21
g/cm2-sec, whereas the average flux for the central scenario was 1.7x10"21 g/cm2-sec.
10-57
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
The air concentration estimated for both the central and high end scenario was the same
at 4.4x10~11 /yg/m3. Larger areas tend to increase air concentration prediction; the larger
area of the high end scenario countered the effect of having a lower average volatilization
flux; hence similar air concentrations were predicted for the central and high end
scenarios.
The values used to evaluate the box model approach were the fluxes, as given
above, the mixing zone wind speed, 2 m/sec, which is half the average wind speed
assumed in this assessment, the areas noted above, the side length, estimated as the
square root of the area, and a mixing zone height estimated initially at 2 m. The box-
model air concentration for the central scenario with these parameters is 2.7x10~10//g/m3.
This is 6 times higher than the concentrations predicted in this assessment. The box-
model concentration estimation for the high end scenario, given slightly lower flux as
noted above and the larger land area, was 8.5x10~10//g/m3, which is over an order of
magnitude higher than the concentration estimated for this assessment.
These box-model estimations are higher than the ones made for this assessment.
An uncertain parameter for both modeling approaches is the area of soil contamination
(see next section which discusses area considerations for the approach used in this
assessment). The mixing zone height for the box model is also a parameter of uncertainty.
Users of the box model approach have often assumed a conservative 2 m height
approximating the height of exposed individuals. However, others have claimed this is far
too low a mixing height, suggesting 10 meters or even an atmospheric height closer to
100 meters. Higher mixing zone heights would have brought the box model estimations
more in line with estimations made in this assessment. The closest analogous parameter
in the dispersion model to the mixing zone height is the height of exposed individual,
which is more unambiguously the breathing zone height of 2 m.
One key assumption concerning the exposure site air concentrations resulting from
an off-site area of soil contamination should be questioned. The current approach
assumes that air-borne contaminates originate at the site of contamination and are
transported to the site of exposure. On the other hand, this assessment also assumes that
exposure site soil becomes contaminated over time due to erosion. Also, some of the
example scenarios have tested the impact of very low, perhaps "background", levels of
dioxin-like compounds, which would occur surrounding a site of exposure. It is at least
10-58 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
plausible that volatilization from soils other than the area of elevated contamination would
contribute to air-borne contamination, and concentrations to which individuals are exposed
to.
This was tested by using the on-site algorithms starting with the concentration
estimated to occur at exposure sites resulting from erosion from the nearby off-site
contaminated area. Two exposure site concentrations are considered: one, the untilled or
higher concentration assumed to occur in a 1-cm soil depth at the exposure site, and a
lower tilled concentration relevant for a 20-cm depth. The air concentration estimated to
occur from untilled soil is nearly an order of magnitude higher than that estimated to occur
from the off-site area and transported; the air concentration estimated to occur from tilled
soil is nearly an order of magnitude lower. If the untilled depth were estimated at 5-cm
instead of 1-cm, resulting air concentrations would be similar to concentrations estimated
to originate from the off-site location.
This might imply that exposure site air concentrations are being underestimated if
air concentrations at the site of exposure are assumed to only originate at the site of
contamination, and not also at the site of exposure, or even from other areas. This
implication, as discussed above, is dependent on the assumption made for a depth of
mixing for transported residues in untilled situations. If the exposure site were a farm,
which might be dominated by tilled soil (distributing residues to a 20 cm depth, or
thereabouts), than this exercise implies that exposure site air concentrations might be
dominated by residues which have been transported from a nearby site of contamination.
If the exposure site and surrounding land is residential, which might not be dominated by
tilled soil, than the procedures might be estimating air concentrations up to an order of
magnitude too low.
10.2.9.4. Uncertainty of Key Model Parameters
The two chemical parameters which impact air concentrations are the soil organic
carbon partition coefficient, Koc, and the Henry's Constant, H. Section 10.2.1 above
indicated that Koc for 2,3,7,8-TCDD might be as high as 24,500,000, although it appears
unlikely to be much lower than the value assumed in this assessment, 2,700,000.
Increasing Koc for 2,3,7,8-TCDD to this higher value would decrease exposure site
concentrations by approximately 70% for the various scenarios. Section 10.2.1. above
10-59 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
also notes the H values were in the 10"6 to 10~5 atm-m3/mol range for PCDDs and PCDFs.
The value selected for 2,3,7,8-TCDD was 1.65 x 10~5 atm-m3/mol. Reducing this value to
1x10"6 atm-m3/mol reduces air concentrations by approximately 75% over the various
scenarios. While the Koc and H assignments to 2,3,7,8-TCDD are evaluated as
appropriate, alternate selections would tend to decrease predicted air concentrations,
perhaps by as much as 75%.
Physical parameters associated with estimating volatilization flux are soil porosity
and soil organic carbon content. The value used for soil porosity was 0.50. Porosities for
sandy surface soils show a range of 0.35-0.50. Medium to fine-textured soils (loams,
clays, etc.) show a range of 0.40-0.60 (Brady, 1984). Varying porosity between 0.3 and
0.6 showed little impact to model results. Reducing the porosity to 0.30 only reduced
estimated air concentrations by 17%; increasing to 0.60 increased air concentrations by
1.3%. The soil organic carbon content, OCS,, was 0.01. This might be low and more
typical of sandy soils rather than soils with high silt or clay content. Most soils would not
have organic carbon contents exceeding 0.05. Increasing the OCS| to 0.05 decreases air
concentrations by approximately 55%; increasing OCS, to 0.03 decreases concentrations
by 42%.
Physical parameters associated with dispersion modeling for the on-site source
category include wind speed, area of contamination, and height of exposed individual.
None of these parameters have a significant impact on estimating air concentrations. The
wind speed was set at 4.0 m/s, which is mid-range between observed windspeeds around
the United States of 2.8 to 6.0 m/s. Decreasing the windspeed to 2.8 m/sec increased air
concentrations by 22%, increasing the wind speed to 6.0 m/s decreased air concentrations
by 35%. The area was set at 4,000 and 40,000 m2 for the on-site central example
scenario, #1, and the on-site high end scenario, #2. Increasing these areas by a factor of
ten increases air concentrations 39 (Scenario #1) and 48% (#2). Decreasing them by a
factor of ten reduces air concentrations 22 (#1) and 30% (#2). For both Scenarios #1 and
#2, soil concentrations were set at very low levels which researchers described as
"background" or "rural". The concept for example Scenarios 1 and 2 was that there may
be a low level extensive concentration typical of large areas, such as a background level.
If so, than much more area than just the area of the exposure site would be characterized
by this low concentrations. Given this framework, the exposure sites should probably be
10-60 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
considered small land areas within larger areas that contribute to emissions impacting
exposed individuals. The model's estimations, which limited land area, may be
underpredictions, although more appropriate predictions might only be 30-40% higher than
model results (i.e., a 10-fold larger land area impacting the breathing areas of exposed
individuals is probably too large). Finally, the height of the exposed individual was set at
2.0 meters. The dispersion model includes an exponential term which results in a
logarithmic decrease in estimated air concentration as the height parameter increases. The
height term is not expected to be variable, and the estimate of 2 meters as the breathing
zone height is a reasonable parameter assignment. Regardless, changing this parameter to
1 or 3 m only results in a 5% increase (at 1 m) or decrease (at 3 m) in estimated air
concentrations.
Physical parameters associated with dispersion modeling for the source categories
where soil contamination is distant from the site of exposure, the off-site and ash landfill
source categories, include the frequency wind blows from source to receptor and the
distance from source to receptor. The 15% frequency used generally assumes wind blows
in all directions equally. Assuming 30% roughly doubles exposure site air concentrations
(100% increase), and reducing frequency to 5% reduces air concentrations by 67%.
These are more likely for real world settings, where wind normally blows in a prevailing
direction, so it may be concluded that different frequencies for specific sites would change
predictions by less than 50%. The distance from the sites of soil contamination and the
exposure sites was 150 meters for the two source categories. Increasing this distance to
500 meters decreases concentrations by 66%, increasing it to 1000 meters results in
nearly an order of magnitude reduction (85% reduction), and increasing the distance to
1500 meters results in an more than an order of magnitude reduction in air concentrations.
At 15,000 meters, air concentrations are three orders of magnitude lower than at 150
meters. The virtual point source model used to calculate far field dispersion is not
considered reliable for small distances, i.e., less than 100 meters. It is more appropriate to
use the near field dispersion model used to estimate on-site air concentrations if an
exposed individual is very near a site of contaminated soil. A test was run using the on-
site source near-field dispersion model and the concentration (1 ppb) assumed for the off-
site example scenario. The air concentration predicted was one order of magnitude higher
than calculated for the exposure site 150 meters from the contaminated site.
10-61 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Another parameter which impacts air concentrations is the duration of exposure,
although of course this is not a physical parameter. It is seen in Equation (5-18) that
volatilization flux is a function of the inverse square root of the exposure duration (in
seconds) - the longer the duration of exposure, the lower will be estimated average flux
rate (see Section 5.3.2. for further explanation). The impact of the exposure duration
assumption is small, however. Starting with the 20-year duration assumed for high end
scenarios, average air concentrations are doubled when duration is decreased to 5 years,
and reduced just over one-half when duration is increased to 70 years.
In summary, this section has evaluated the sensitivity of physical and chemical
parameters, and the exposure duration, to estimates of air concentrations. It is important
to note that all physical parameters discussed above can be known with certainty for a
specific site. The same cannot be said for the chemical parameters, Koc and H. As such,
these chemical parameters might be judged as the most uncertain. Possible alternate
values for these two parameters might decrease air concentrations by as much as 75%.
Most physical parameters, when changed within a range of reasonable values, changed
concentrations by 50% or less. Parameters with this small an impact include soil porosity,
wind speed, area of contamination, height of the exposed individual, frequency wind blows
from contaminated source to exposed individual, and exposure duration. Distance from a
contaminated site to exposed individuals could have more of a pronounced impact, with
differences orders of impact or more with similar large changes in distance from the 150
meters assumed for the example scenarios in Chapter 9.
10.2.10. Particulate Phase Inhalation
This section will summarize the uncertainties associated with estimating particulate-
phase inhalation exposures. All exposure parameters for this pathway, including the
inhalation rate, contact fractions, and exposure durations, were discussed in the previous
section on vapor-phase inhalation exposures. The same discussion is pertinent to this
pathway, and will not be further discussed. This section will focus on the algorithms
estimating particulate-phase concentrations at the site of exposure.
Dust emissions from contaminated soil occur via wind erosion. An empirical
equation estimates the emission due to wind erosion. Other sources of emissions
estimated for the ash landfill source category include: suspension of contaminated
10-62 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
roadway dust due to vehicular traffic, fugitive emissions from trucks, emissions from truck
unloading, and emissions from spreading and compacting ash at the landfill. The near- and
far-field dispersion models as described above are used to estimate dispersion and
transport for dust emissions; their uncertainties are discussed above in the section on
vapor-phase inhalation exposure. No data was found on air-borne particle-phase
concentrations of dioxin-like compounds originating from soil contamination. This section
focuses only on sensitivity and uncertainty of model parameters.
A summary of uncertainty issues with this pathway is given in Table 10.10.
10.2.10.1 Wind Erosion Modeling
The empirical relationship used to estimate wind erosion assumes that the soil
surface is uncrusted and composed of finely divided particulates, a situation where dust
emissions will be maximized. The procedures are presented in EPA (1985) and were
developed using field measurements presented in Gillette (1981) for highly erodible soils.
It is presumed that estimations made with these procedures would correlate with data in
Gillette (1981); other means to validate the approach were not presented in EPA (1985),
and additional validation is not attempted here.
The model does show some sensitivity to required parameters. The wind speed
generally varies from 2.8 to 6.0 m/sec, with 4.0 m/sec used in model estimations.
Reducing this speed to 2.8 m/sec reduces air-borne concentrations by about one-half,
while increasing wind speed to 6.0 m/sec roughly doubles air-borne concentrations. Other
parameters include the fraction of ground covered by vegetation, the threshold wind
velocity, and the model specific function, F(x). These are all related, as assumptions of
ground cover influence all three. The example scenarios include three sets of assumptions
demonstrate the range of predictions with different ground cover assumptions. Scenario
1, the 1-acre residential scenario, assumes grass cover and a 90% ground cover
assumption. Scenario 2, the 10-acre farm scenario, assumes 50% ground cover, and
Scenario 3, the off-site 10-acre contaminated site, is assumed to be bare with 0% ground
cover. There is a 5-fold difference in wind erosion flux from the set of assumptions
leading to the most wind erosion flux, those for the farm with 50% cover, to the set of
assumptions leading to the least wind erosion, those for the 1-acre residence with 90%
ground cover. Interestingly, the wind erosion flux from the off-site area with bare soil
10-63 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.10. Uncertainties and sensitivities associated with estimating particulate-phase air
concentrations from contaminated soils.
Assumption/
Method
Approach
Rationale
Uncertainty
Comments
Exposure parameters
See Table 10.7. and Section 10.2.9.1. for discussion of inhalation rate and contact fraction
Wind Erosion
Empirical equation
assumes uncrusted soil
surface and finely
divided particulates
Equation based on
field data and given
assumptions, would
tend to maximiz«
predictions
Data was unavailable
for comparison with
model results; approach
assumed reasonable
since it was based on
field data; changes in
parameters could lead
to as much as order of
magnitude change in
predictions as compared
to estimations in this
assessment.
Particulate phase exposures
are 2-3 orders of magnitude
lower than vapor phase
exposures; given estimation
sensitivity within an
order of magnitude,
uncertainty is less an
issue for particulate
phase concentrations
Fugitive Ash
Emissions
Used AP-42 emission
factor equations
EPA (1988)
Like wind erosion,
these are empirical
equations based on
data
Same comment as made
for wind erosion applies
here
Same comment as made for
wind erosion
Overall: Sensitivity and uncertainty do not appear to be critical issues for particulate phase concentration estimation as exposures from
contaminants on particulates are 2 to 3 orders of magnitude lower than exposure to vapor-phase contaminant concentrations, and lower still
compared to all other exposures.
contamination was half as much as from the farm with 50% ground cover. This is
counterintuitive and would question the procedure.
In summary, the validity and hence certainty of the approach to estimate wind
erosion flux cannot be fully evaluated, although it was developed from field data. The
underlying assumption that the soil surface is uncrusted and composed of fine particulates
is a maximizing assumption. In Chapter 9 demonstrating the methodologies, it was
observed that particulate phase exposures were 2 to 3 orders of magnitude lower than
vapor-phase exposures, and were lower than all exposures except water and fish
exposures from a stream setting. The fact that the approach would tend to maximize dust
emissions and yet particulate phase exposures were still significantly lower than vapor
phase exposures tends to minimize the need for certainty in this algorithm. Air
10-64
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
concentrations were relatively insensitive to changes in required parameters: alternate
parameter selections from those selected for demonstration in Chapter 9 resulted in less
than an order of magnitude difference.
10.2.10.2 Fugitive A sh Emissions
Fugitive emissions of contaminated ash occur due to ash handling and management
at the landfill site. Emission factor equations for these processes are known as AP-42
emission factors and are described in EPA (1988). These equations are empirical and are
developed from data. Like the wind erosion algorithm, further validation of these
equations was not undertaken for this assessment.
There are numerous parameters and assumptions necessary to apply these
equations. All such selections made for the demonstration in Chapter 9 considered the
following: 1) particle size multipliers appropriate to estimate emissions of inhalable size
particulates, < 10;/m, 2) survey information on ash landfill operations described in MRI
(1990), and collected with the intent of applying similar AP-42 emission factor equations,
3) mid-range values for AP-42 parameters such as number of truck wheels, number of
wheels, number of days with rainfall, and so on, and 4) control efficiencies of 90% (only
10% of potential emissions occur) associated with 2 of 4 sources of emissions for which
controls are common and expected - these include emissions resulting from vehicular
traffic over unpaved roadways (chemical suppression or roadway wetting) and emissions
off trucks in transit (wetting or tarpaulin). The observation was made in Chapter 9 that
fugitive ash emissions were equal to the emissions that result from wind erosion of the
ash landfill surface.
An in-depth analysis of the several parameters and assumptions is not warranted.
They are all site-specific, and guidance to select parameter values is provided in EPA
(1988). However, it is easily shown that estimates of fugitive emissions can be larger if a
set of plausible alternate assumptions are made. The following are made in a single
demonstration: 1) 90% control efficiencies will be reduced to 30% indicating some but
little control of emissions, 2) 60 instead of 121 days of rainfall (not atypical of western
United States), 3) larger (more wheels) and heavier trucks involved in ash delivery, and 4)
slightly windier at 5 m/s instead of 4 m/sec. No assumptions on increased deliveries of
ash were made since the 2700 TPD incinerator is relatively large to start with.
10-65 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
This exercise showed wind erosion doubled because of the increase in wind speed,
and fugitive emissions increased by an order of magnitude with the several changes.
Particulate phase air concentrations at the site of exposure 150 meters away only
increased by a factor of five. Examination of the results also showed that emissions off
trucks, from unloading of trucks, and from spreading of ash onto the landfill were 4-6
orders of magnitude lower than emissions resulting from vehicular traffic over unpaved
roads and wind erosion. Any changes to parameters with those less significant
contributors is irrelevant.
As noted above in the wind erosion section, exposures due to particulate phase
emissions are 2 to 3 orders of magnitude lower than vapor-phase exposures, and lower
than nearly all other exposures. This observation was made including the estimations from
the ash landfill. While there are several parameters and assumptions required for fugitive
ash emissions, the above analysis showed that only emissions from vehicular traffic over
unpaved roadways is significant. It also showed that with a set of plausible assumptions
that would lead to more particulate emissions, exposure site particulate phase
concentrations would only increased by a factor of five.
10.2.11. Uncertainty in Fruit and Vegetable Ingestion
This section will evaluate the algorithms used to estimate concentrations of dioxin-
like compounds in vegetation, including fruits and vegetables as well as pasture grass and
fodder. The latter two vegetations are used in the food chain algorithm estimating
concentrations in beef and milk. Discussions here are therefore pertinent to the next
section below.
The vegetation algorithm separates above and below ground vegetations. Below
ground vegetation includes vegetables such as potatoes, carrots, or onions. Above ground
vegetation includes other fruits and vegetables, as well as pasture grass and fodder.
Below ground vegetation concentration are estimated solely on the basis of soil to root
transfers. Above ground vegetations are estimated based on vapor phase transfers and
particulate depositions.
The first section below discusses the rates of fruit and vegetable ingestion.
Following that is an indepth evaluation of the algorithms estimating vegetation
concentrations. First, studies measuring soil and plant concentrations are evaluated and
10-66 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
compared to model behavior. Then, several issues of uncertainty are discussed, including
assumptions and model parameters. A summary of uncertainty issues associated with
the fruit and vegetable ingestion pathway is given in Table 10.11.
10.2.11.1 Fruit and Vegetable Ingestion Rates
Consumption rates of 200 g/day for vegetables and 140 g/day for fruit were
derived in EPA (1989) and recommended for general assessment purposes. They include
all fruits and vegetables and were derived from two principal sources: Foods Commonly
Eaten by Individuals: Amount Per Day and Per Eating Occasion (Pao, et al. 1982), and 2)
Food Consumption: Households in the United States. Seasons and Year 1977-1978
(USDA, 1983). Pao, et al. (1982) used the data from the USDA survey, which included
interview responses from 37,874 individuals, to estimate total consumption and
percentiles of home-grown fruits and vegetables. EPA (1989) identifies two principal
sources of uncertainty with Pao's estimates:
• These data are from all consumers, only a small percentage of whom are also
home gardeners. Those who home garden may have higher total rates of consumption.
• USDA's survey only included information for 3 days from each respondent:
products eaten the day before, the day of, and the day after the interview. Therefore, the
results on a per day basis only include information for three days from respondents; what
is required for long term exposure assessments is an amount eaten per average day over
the course of a long time period, such as a year or a duration of exposure. EPA (1989) did
not discuss whether this aspect of uncertainty might render the 200 and 140 g/day
estimates over- or underestimates.
These total consumption rates were reduced considering fruit and vegetables which
are "protected" and "unprotected". Protected fruit, for example, included citrus and
cantaloupe, whereas unprotected fruit included peaches or apples. This distinction was
made because evidence indicates very little translocation or residues to within the plant. It
was assumed that there would be no exposure when the produce was protected. Again
using data from Pao, et al. (1982) as summarized in EPA (1989), it was estimated that
44% of total fruit ingestion was ingestion of unprotected fruit and 74% of total vegetable
ingestion was unprotected vegetables.
10-67 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.11. Uncertainties associated with vegetable and fruit ingestion exposure algorithms.
Assumption/
Method
Approach
Rationale
Uncertainty
Comments
Rates of fruits: 88 g/day above
ingestion ground unprotected, 0
g/day below ground unp.,
0.20-0.30 home grown;
veg: 76 g/day abv grd
unp. 28 g/day bel. grd.
unp.; 0.25-0.40 home
grown.
Protected/unprotected
distinction because resi-
dues not expected to
tranlocate; above/below
ground because procedures
for soil transfers are
different
Much variability expected
when using approach for a
specific site when actual
home gardening can be
ascertained.
All parameters assumed are
evaluated as reasonable for
general exposure to broad
categories of fruits and
vegetables.
Below ground
vegetable
concentration
Uses empirical
Root Concentration
Factor which is function
of Kow, VGbg, an
empirical correction
factor, and soil water
concentrations
Separating below ground
with above ground vegeta-
was critical and supported
by the literature; approach
based on laboratory experi-
ments with barley roots.
The VG^ empirically des-
cribes the difference in
barley roots and bulky
underground vegetables;
although assignment of
0.01 is rationally based,
arguments presented could
estimate it instead at
0.10 or 0.001; algorithm
also a function of Kow,
which is uncertain by
2 orders of magnitude;
most likely change in Kow
decreases concentration by
up to an order of magnitude
Comparison with the literature
indicates estimates of below
ground vegetables may be low
by an order of magnitude; how-
ever, the trend that below
ground vegetables have higher
transfers from soil to plant
as compared to above ground
vegetables was correctly
captured.
Air-to-leaf
Transfer
for Above
Ground
Vegetation
Uses air-to-leaf factor
developed in laboratory
conditions for 10 com-
pounds transferred to
azalea leaves; empiri-
cally corrects for plants
like fruit/veg. that have
much less transfer to
inner parts as compared
to leaves
Sound data on which to
develop the empirical
transfer factor, which
is also presented as a
function of Kow and H
(Bacci, et al., 1990)
air to plant transfer
algorithm is critical as
researchers believe this
process to dominate in
real world settings for
dioxin-like compounds
Empirical correction factor
is necessary, but values
selected could also be low
or high, as above; like
algorithm above, transfers
critically a function of Kow;
most likely alternate Kow
would increase concentra-
tion estimates by up to
an order of magnitude
Like root uptake, this algorithm
should be considered for future
revisions. The theoretical basis
for vapor phase transfers to
plants is sound, and the transfer
factor is based on sound data. A
refined correction factor might
consider photodegradation of
residues on plant surfaces.
surfaces.
(continued on next page)
10-68
7/31/92
-------
Table 10.11. (cont'd)
DRAFT-DO NOT QUOTE OR CITE
Assumption/
Method
Approach
Rationale
Uncertainty
Comments
Paniculate Model for soil contami-
Phase nation solves for air-
Deposition borne particulate phase
for Above concentration and applies
Ground deposition rate; ISC model
Vegetation estimates deposition rate
for stack emission source
source category; plant con-
centrations based on
deposition rates, plant
mixing volumes, canopy
cover and washoff
Algorithm developed for
radionuclide impact to
agriculture (Baes, et al.,
1984) but applied to other
contaminants sorbed to air-
borne particulates; parti-
culate deposition as mecha-
nism of plant contamination
speculated to be of concern
for 2,3,7,8-TCDD in early
literature. Model results
show particulate deposition
critical for stack emissions,
less critical for soil
contamination.
Fugitive dust emission
algorithms (wind erosion, ash
handling) developed from data
but not field validated; ISC
modeling theory well founded;
particulate depositions
underestimated, but higher
depositions, at most, will
double fruit/vegetable
concentrations while
not impacting grass and fodder
The basic approach given con-
taminant deposition rates is
defensible; model results imply
particulate deposition is insig-
nificant for soil contamination
but significant for stack
emissions of particulates.
Overall: All ingestion parameters assumed are evaluated as reasonable for general exposure to broad categories of fruits and vegetables.
However, great variability is expected if using these procedures on a specific site where home gardening practices can be more precisely
ascertained. A contaminant concentration ratio was defined as the concentration of contaminant in vegetation divided by the concentration in
the soil. A comparison of the modeled ratios with those found in the literature showed that, whereas the modeled ratios tended to be lower for
all vegetation (above and below ground fruit and vegetation, grass and cattle fodder) by about an order of magnitude, important trends noted in
the literature were duplicated by the model. These are higher below ground vegetable ratios as compared to above ground vegetable ratios,
and higher ratios for perennials (grasses, e.g.) as compared to vegetables. A key assumption in the vegetation algorithm, that dioxin-like
compounds to not translocate from root to shoot, was verified by two experiments. Vapor-phase contributions to vegetation dominated the
contaminated soil source categories - a minor contribution was made by particulate deposition. This assessment likely underestimated
vegetative concentrations attributable to particulate phase concentrations, although increasing those contributions by an order of magnitude
only as much as doubled vegetable concentrations, while still not impacting grass of fodder concentrations. A critical empirical parameter was
the above and below ground correction factors, VGag and VGbg, both set at 0.01. A different assumption for its derivation might set it at 0.10,
which would increase estimated concentrations and perhaps make concentration ratios more in line with literature values. Other experimentally
derived empirical factors describing the transfer of compounds from soil to below ground vegetables and air to above ground vegetation were a
function of contaminant Kow. An alternate value of log Kow for 2,3,7,8-TCDD would more likely be higher than lower, given a literature range
of 6.15 to 8.5, and a selected value of 6.64. Increasing log Kow tends to decrease below ground vegetation, by as much as an order of
magnitude, while increasing above ground vegetation by as much as an order of magnitude.
A final distinction was required which divided unprotected fruit/vegetables to those
which grow underground and those which grow above ground. Different algorithms were
used to transfer soil residues to plants depending on whether they were above or below
ground. Using the same data once again, it was estimated that no fruits were grown
underground (unprotected or protected), and that 37% of unprotected vegetables were
grown underground.
The result of these two distinctions was to estimate total consumptions rates of
unprotected fruit as no below ground and 88 g/day above ground consumption; for
10-69
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
unprotected vegetables, total consumption included 76 g/day above ground and 28 g/day
below ground.
The overall average fraction of total vegetable and fruit consumption which is
homegrown is estimated as 0.25 and 0.20, respectively (EPA, 1989). EPA (1989)
recommends 90th percentile assumptions for these parameters of 0.40 (vegetables) and
0.30 (fruit), which were assumed in the high end scenarios of this assessment. EPA
(1989) notes a wide range of fraction homegrown for individual vegetables, 0.04-0.75,
and fruits, 0.09-0.33.
All these assumptions discussed: total consumption rates, protected or
unprotected, above or below ground, and fraction home grown, are probably reasonable
for general assessment purposes as long as exposures are to the broad categories of fruits
or vegetables, and not for individual fruits or vegetables. For a site specific assessment,
there will likely be wide variability on the types of produce grown at home, what
percentage of that is unprotected, and so on. Finally, and as is also true for beef and milk
exposures, this assessment only considers the impact of home-grown fruits and
vegetables. In rural settings, it is plausible that a large percentage of an individual's total
fruit and vegetable intake comes from nearby and impacted sources, more than the 20-
40% assumed in this assessment. If that is the case, than contact fractions should be set
at 1.0, and exposures would increase 2-5 times from what they are estimated as in this
assessment.
10.2.11.2. Comparing Predicted Versus Measured Plant Concentrations
There have been several studies which have measured plant concentrations for
plants grown in soils with known concentrations of 2,3,7,8-TCDD. One quantity that can
be estimated from these studies is a plant:soil contaminant concentration ratio. This
quantity is analogous to the contaminant concentration ratio defined for evaluating the
erosion algorithm (see Section 10.2.3 above), and equals the concentration in the plant
divided by the concentration in soil in which the plant is growing. Contaminant
concentration ratios predicted to have occurred can be compared against those that have
been measured in the various studies. It is important to note that this is not a validation
exercise. None of the experimental or field conditions for the literature studies were
duplicated in this exercise. Comparing contaminant concentration ratios only gives a sense
10-70 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
of how literature and modeled transfers from soil to plant compare. The literature articles
measuring soil and resulting plant concentrations of 2,3,7,8-TCDD are summarized in
Table 10.1 2. This table also includes contaminant concentration ratios.
What is common about all these studies is that the soil in which the plants were
grown likely represents the source of 2,3,7,8-TCDD. It seems logical, therefore, that
results from these studies can be compared with model estimations where soil in which
the plants grow is the ultimate source. This is only true for the "on-site" source category.
For that category, the soil is contaminated to an unspecified depth with dioxin-like
compounds. That soil provides the source for underground plant to root transfers and
above ground volatilization and dust emissions for air-to-leaf transfers and particulate
depositions. The same is not true for the other source categories, where the source of
vapor phase and particulate phase contaminants is either distant incinerator stacks or
areas of contaminated soil. Because the stack emission, off-site, and ash landfill source
to be distinctly different than the studies summarized in Table 10.12, only the contaminant
concentration ratios from the on-site source category will be examined.
The following contaminant concentration ratios were estimated for the two
scenarios demonstrating the on-site source category in Chapter 9, Scenarios 1 and 2:
below ground vegetables - 1.45x10"3, above ground vegetables/fruit - 3.6x10"5, grass -
2.4x10"2, and fodder 1.2x10~2. Some observations from contaminant concentration
ratios found in the literature and these estimated by the model are:
1) The contaminant concentration ratios for above and below ground
fruits/vegetables, 10~5 and 10"3 respectively, are about an order of magnitude lower than
found by Wipf, et al. (1982) for experiments conducted outdoors and indoors, on fruits
and vegetables, using Seveso contaminated soils. Wipf showed ratios for above ground
fruit grown outdoors to be in the range of 0.0009 to 0.0042, which is higher than the
model's 10~5 above ground ratio. Wipf found no residues at 1 ppb detection limit.in the
cobs and kernels of corn, although a small amount in the sheaths of corn, leading to a ratio
of 0.0 for the edible part of corn and 0.0008 for the sheath. He showed (washed and
unwashed) contaminant ratios for carrots in the range of 0.01 to 0.17. These ratios are
higher than the 10~3 ratio estimated by the model. Wipf's underground ratio range, 0.01-
0.17, was two orders of magnitude higher than they found for above ground fruits and
vegetables, 0.0009-0.0042. This two order of magnitude difference is similar to the
10-71 7/31/92
-------
Table 10.12. Summary of plant concentration vs. soil concentration data for 2,3,7,8-TCDD.
CO
Plant/Soil
Concentrations
Contaminant
Ratio
Reference and Comments
(9-42 ppt)/
10 ppb
.0009-.0042 Wipf, et al., 1982; analysis of apples, pears, plums, figs, peaches, and apricots grown in
Seveso, Italy year following contamination; apples, pears, and peaches showed >95% of
whole fruit concentrations listed here was in the peels; analysis of vegetative samples in less
contaminated areas showed non-detections at 1 ppt detection limit; reference was unclear as
to whether reported concentrations in fruit was based on fresh or dry weight.
(8-9 ppt)/
10 ppb
.0008
Wipf, et al., 1982; concentrations listed were those found in sheaths of corn grown year
following Seveso contamination; none found in cobs and kernels at 1 ppt detection limit.
TI
6
O
(54-167 ppt)/
d-5 ppb)
.01-.17
(0.8-9.2 ppb)/
(2.7-8.3 ppb)
.24-1.73
Wipf, et al., 1982; results are for greenhouse carrots grown in Seveso contaminated soil; the
54 ppt concentration listed was for carrot peels and inner portions; the 167 ppt listed includes
the 54 ppt plus additional residues found in wash water and can be described as "unwashed"
concentration; 96% of 167 ppt unwashed concentration includes that found in wash water
(67%) and peels (29%).
Coccusi, et al., 1979; results are for carrots, potatoes, narcissus, and onions grown on
contaminated soil the spring following the Seveso contamination; aerial plant part ratios were
0.25-0.40 - underground part ratios were 0.23-1.73; residues in contaminated plants were
found to dissipate when contaminated plants transplanted to unpolluted soils; results
contradict those from Seveso in Reference 1 noted above; results were expressed in fresh
plant weight and fresh soil basis.
(continued on next page)
D
c
O
H
m
O
•J3
O
H
m
CD
NJ
-------
Table 10.12. (cont'd)
Plant/Soil
Concentrations
Contaminant
Ratio
Reference and Comments
O
I ^
GO
(1-63 ppt)/
(12-3300 ppt)
0.003-0.35 Sacchi, et al., 1986; data was for: "aerial parts" (sic) of bean and maize plants, tritiated
TCDD amended soil with concentrations ranging as noted, taken at different intervals including
7, 34 and 57 days (one test), 17, 34, and 57 days (another test), 8 and 77 days, and 8 and
49 days, and in tests where soil was and was not amended with peat. Results showed
increasing plant concentrations with increasing soil concentrations, but the ratio of plant to soil
concentrations was inversely related to increasing soil concentrations (lowest ratios at highest
soil concentrations). Soils without peat had higher ratios than soils with peat. Plant
concentrations were on a fresh weight basis.
(0.2-4.2 ppt)/
(2-6000 ppt)
.00008-.3
ND (DL =
60 ppb
ppb)/
<0.017
(10-760 ppt)/
411 ppt
.02-1.85
Hulster, A. 1991; potatoes (shoot and tuber results) and lettuce grown outside on soil and in
indoor potted plants PCDD and PCDF concentrations in soil ranging from 2.4 to 6071 ng/kg
(ppt). Plant:soil ratio trend followed similar trend as of Sacchi above - highest ratios were at
the lowest soil concentrations. Plant concentrations given in dry weight basis.
Isensee and Jones, 1971; results are for mature oat and soybean tops, and oat grain and the
bean of soybean, in soil treated with [14C]TCDD to achieve a concentration of 60 ppb - no
residues of TCDD were found; ratios of 0.14 and 0.28 were found for 2,4,-dichlorophenol
(DCP) in oat and soybean tops, and 0.20 for 2,7-dichlorodibenzo-p-dioxin (DCDD) in oat tops;
trace amounts of DCP and DCDD were found in the bean of soybean.
Young, 1983; data was perennial grasses and broadleaf plants grown on soils which had
received 9 years of 2,4,5-T herbicide applications - the herbicides were contaminated with
dioxin - 8 years prior to the plant studies and at the time of the study, had known 2,3,7,8-
TCDD soil concentrations at approximately 400 ppt to a total depth of 15 cm over the test
area. Plant tops (culm, leaves, and stems) had concentrations 10-75 ppt, or ratios of .02-. 18.
Plant crown (portion just above the soil surface) had a concentration of 270 ppt, or a ratio of
0.66. Plant roots had concentrations of 710-760, or a ratio greater than 1.7. It was not
known whether concentrations are on a fresh or dry weight basis.
O
30
O
O
z
O
H
D
C
O
H
m
O
3
O
H
m
CO
CO
ro
-------
DRAFT-DO NOT QUOTE OR CITE
difference in modeled contaminant concentration ratios. This trend of estimating higher
ratios for below ground vegetation as compared to above ground vegetation was noted in
several of the studies summarized in Table 10.12.
Wipf's contaminant concentration ratios, however, are much different than those of
Coccusi, et al. (1979) under similar experimental conditions. Coccusi's ratios ranged from
0.24 to 1.73 for several vegetables. Hulster (1991) tested different soil concentrations of
TE (toxic equivalents of 2,3,7,8-TCDD) ranging from 2 to 6000 ppt in potted and outdoor
experiments including potatoes and lettuce. Contaminant concentration ratios were lowest
at the highest concentrations in the 10~5 to 10~4 range, consistent with this assessment,
but were in the 10~2 to 10~1 for the lowest concentrations, again much higher than the
ratios in this assessment.
2) The trend in Hulster's data of having higher ratios at lower soil concentrations
was also found in Sacchi (1986). This trend will not be duplicated with the modeling
framework of this assessment, unless the user changes model parameters with changes in
soil concentration. It is unclear why such a trend would occur in experimental conditions.
One possible explanation for below ground vegetation is that binding sites in the
vegetation become limiting when the supply of available contaminant to bind increases.
For above ground vegetation, binding site limitations with increases in air concentration
may also be the case, or volatilization from soil may be unhindered at low concentrations
but inhibited at high concentrations.
3) Intuitively, perennial plants grown in soil contaminated with the dioxin-like
compounds ought to have higher concentrations compared to annual agronomic crops
grown in similar conditions. These compounds resist degradation in soils, and although
they are likely to be less persistent in plants than soils, it seems likely that there would be
a positive accumulation of residues in vegetation not removed from the soil over a number
of years compared to agronomic crops that are removed each year. Young (1983)
sampled perennials in soils previously and extensively treated with 2,4,5-T. At the time of
plant sampling, the soil had measured concentrations of 400 ppt of 2,3,7,8-TCDD to a
depth of 1 5 cm (this depth is about the depth to which grass and weed roots extend).
Plant tops (culm, leaves, and stems) had ratios of .02-. 18, plant crown (portion just above
the soil surface) had a ratio of 0.66, and plant roots had a ratio greater than 1.7. This
assessment estimated the highest contaminant concentration ratios for grass and fodder at
10-74 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
10~2, which compares favorably to the plant top ratios reported by Young. The difference
between grass/fodder and fruits/vegetables in this model is expressed in the empirical
parameter, VGag, which is set at 1.00 for grass and 0.50 for fodder, and 0.01 for
fruits/vegetables. Fruit/vegetable concentrations would be similar to grass/fodder if the
empirical parameter VGag was set to the same value for both types of vegetation. The
basis for parameter assignment was not on a difference in annual or perennial residue
accumulations. Rather, the above-ground VG parameter assignments was based on the
differences between the leaves of Bacci's experiments (1990, 1992) used to derive a
vapor phase air-to-leaf transfer factor and the above ground fruits/vegetables, grass, and
fodder. Similarly, the below ground VGbg assignments were based on differences between
roots of young barley plants of Brigg's (1982) experiments and below ground vegetables.
4) The early literature is not consistent in its reporting of fresh weight vs. dry
weight, and in some articles, such information is not provided. Also, the early literature
was unclear as to the mechanism of plant contamination. Outer portions of below ground
vegetation contained concentrations that seem unambiguously due to sorption. However,
translocation with transpiring water has been hypothesized as the cause for contamination
of above ground portions in two studies (Coccusi, et al. 1979; Sacchi, et al., 1986), while
contamination by soil particles has been hypothesized as the cause for contamination of
above ground vegetation in two other studies (Young, 1983; Wipf, et al., 1982). Current
thought on highly hydrophobic organic contaminants such as the dioxin-like compounds is
that sorption to outer portions of below ground vegetation is principally the cause of
transfer from soil to plant. Little within plant translocation is expected to occur. Above
ground portions of vegetation are thought to be principally impacted by vapor phase
transfers. These issues are further discussed in the next section.
10.2.11.3. Uncertainty Evaluation of the Vegetation Concentration Algorithm
A key assumption for the algorithm in this assessment is that there is an
insignificant translocation of residues taken in by roots to above ground portions of plants.
Recent literature on 2,3,7,8-TCDD attributes below ground vegetation concentrations to
sorption onto outer plant surfaces, and above-ground concentrations to vapor-phase
transfers 2,3,7,8-TCDD into shoot, leaves, and other above-ground plant material.
The specific issue of uptake and translocation via transpiration was investigated
10-75 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
using soybean and corn plants grown hydroponically in carefully constructed growth
chambers (McCrady, et al., 1990). Roots and the hydroponic growth solution were
separated from the shoots and leaves of these plants using two separate chambers, one
inverted over the other. Separate air-flow systems for each chamber included traps for
volatile organics. Mass balance on the tritiated TCDD experiments was able to recover
98% in the soybean experiment and 86% for the corn experiment. Most of the recovered
material was found in the roots; 75% for soybeans and 67% for corn, with the second
highest recovery was on the inside surface of the root chamber, around 15% for both
experiments. Recovered TCDD was also found, in order of decreasing percentage, in the
growth solution, root chamber air, shoot chamber air, and shoots. The recovery from the
shoots was negligible at 0.004% and 0.001 % of the total TCDD for the soybean and corn,
respectively. McCrady, et al. (1990) concluded that transpiration stream transport of
2,3,7,8-TCDD to plant shoots is an insignificant mechanism of plant contamination, and
that volatilization of TCDD is an important transport mechanism that can result in
significant quantities of airborne TCDD being absorbed by plant shoots.
Briggs, et al. (1982) provide another way to evaluate the translocation of
contaminants from roots to above ground vegetation. Experiments with barley roots in
growth solution led to the development of an empirical parameter describing the efficiency
of transport of organic chemicals to plant shoots from root uptake. This parameter is
called the Transpiration Stream Concentration Factor (TSCF) and is defined as
(concentration in transpiration stream)/(concentration in external solution). The empirical
formula presented for this factor is:
TSCF = 0.784 e
~[ lo& Kow - 1-78 ]2 / 2.44 (10-5)
Given a log Kow for 2,3,7,8-TCDD of 6.64, TSCF is solved for as roughly 5 * 10"5.
Assuming that the concentration of external solution term for the experimental conditions
of Briggs' experiments is equivalent to the concentration in soil water in a field situation,
then the TSCF for 2,3,7,8-TCDD implies that the transpiration stream water of a plant is
over 5 orders of magnitude lower than the soil water concentration. Like McCrady's
experiments, this also shows the insignificance of translocation of residues from roots to
shoots.
10-76 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
The experiments of Bacci, et al. (1990, 1992) in carefully controlled laboratory
conditions demonstrated the transfer of 10 air-borne vapor-phase contaminants, including
TCDD, into azalea leaves. His work formed the basis of the air-to-leaf transfer algorithm
of this and several related efforts. Although some of the early research speculated that
particulate depositions were the cause for above ground concentrations, the current
thought is that vapor phase transfers dominate the impact to above ground vegetation.
Model results on the proportion of above ground plant concentrations that are due
to air-to-leaf transfer and particulate deposition were examined. Results from this
examination for 2,3,7,8-TCDD are summarized in Table 10.13. Results were opposite for
soil based contamination source categories, the on-site, off-site, and ash landfill source
categories, versus the stack emission source category. For the soil-based categories, it
was found that the air-to-leaf transfer of vapor-phase contaminants generally accounted
for over 90% of total above ground plant concentrations. However, for the stack emission
source category, particulates originating from the stack and depositing onto vegetation at
the site of exposure provided the largest contributor to plant concentrations, between
83.8% and 99.0% for the three types of vegetation.
An examination of the vapor-phase and particulate-phase concentrations of 2,3,7,8-
TCDD in air provide one explanation for this discrepancy. For the stack emission source
category, vapor phase concentrations are one approximately order of magnitude lower
than particulate phase concentrations, as predicted by the ISC model. For the other three
soil contamination source categories, predictions of vapor phase concentrations are
approximately one order of magnitude higher than particulate phase concentrations.
Another part of the explanation is the rate of particle deposition. The ISC internally
calculates a deposition rate of approximately 0.06 m/sec, which is based on particle size
distributions and includes wet deposition. For the soil contamination source categories, a
gravitational deposition velocity of 0.01 m/sec was used. In summary, the stack emission
source category predicts higher contributions by particulates because of a greater reservoir
of air-borne particulate phase contaminants and because of a greater deposition velocity,
as compared to the soil contamination source categories.
10-77 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.13. Contribution of above ground vegetation concentrations from air-to-leaf
transfers and paniculate depositions.1
Description2
Air-to-leaf
Transfers
Particulate
Deposition
Scenarios 1 & 2: On-site
Soil Source Category
vegetables/fruit
pasture grass
fodder
89.2
99.4
98.8
10.8
0.6
1.2
Scenarios 3 & 6: Off-site and
Ash Landfill Source Categories
vegetables/fruit
pasture grass
fodder
Scenarios 4 & 5: Stack Emission
Source Category
vegetables/fruit
pasture grass
fodder
93.8
99.6
99.5
1.0
16.2
12.3
6.2
0.4
0.5
99.0
83.8
87.7
1 Results are in percent of total contribution and are specific to 2,3,7,8-TCDD.
Scenarios 1 and 2 demonstrated the "on-site" source category, where soil at the exposed individuals home was
contaminated at a level of 1 ng/kg (ppt) 2,2,7,8-TCDD; Scenario 3 demonstrated the "off-site" source category, where soil
at a contaminated site 150 meters away was initialized at 1 >/g/kg (ppb) 2,3,7,8-TCDD; Scenario 6 demonstrated the "ash
landfill" source category, where soil at a nearby landfill where ash was disposed of was initialized at 0.7 >/g/kg (ppb)
2,3,7,8-TCDD; Scenarios 4 and 5 demonstrated the "incinerator stack emission" source category - in Scenario 4, the
exposed site was 2000 meters from the incinerator, and in Scenario 5, the exposed site was 500 meters from the
incinerator.
It is very likely that the soil contamination source categories, and to some extent
the stack emission source category, are underestimating the particulate phase
contributions, for at least four reasons:
• All algorithms estimating air-borne contaminant concentrations for soil
contamination only estimate concentrations of PM-10, or inhalable size particulates,
those 10 jjm size diameter and less, while the ISC model considers all size
particulates emitted from stacks. Larger size air-borne particulates, while not
10-78
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
inhalable, would deposit onto vegetation.
• Dry deposition only is considered for the soil contamination categories, whereas
the ISC model includes dry plus wet deposition;
• For the source categories involving soil contamination distant from the site of
exposure, the off-site and the ash landfill source category, only these off-site
locations provide the source of air-borne particulates. Meanwhile, algorithms are in
place estimating exposure site contamination, albeit to thin surface levels.
Certainly, the reservoir of air-borne particulates depositing onto vegetation would
also include contributions from where the vegetation is located.
• The modeling does not consider the splash effect of rainfall, which would
deposit soil onto the lower parts of plants. This would make the most impact for
grass and for vegetables near the ground surface such as lettuce.
The precise impact of these factors might be investigated more fully in a later
assessment. However, it may turn out that the increase in particulate contributions to
vegetation impacts is still low compared to vapor phase transfers for the soil
contamination source categories. Tests were run increasing the amount of particulate
phase contaminants depositing onto vegetation by an order of magnitude to the on-site
and the ash landfill source category. Minimal impact was noted for the on-site source
category. The vapor phase/particulate phase contributions to above ground fruits and
vegetables, originally 98%/2% (from Table 10.13 above), changed to 84%/16% with an
order of magnitude increase in particulate phase contributions. Vegetable concentrations
only increased 16%. The impact was less for grass and fodder, with less than a 1%
increase in concentrations. The impact was greater for the ash landfill source category.
The fruit and vegetable vapor phase/particulate phase breakdown, originally 89%/11 %,
swung in favor of particulate phase, 46%/54%. Vegetable and fruit concentrations nearly
doubled. The impact to grass and fodder was still marginal, with only a 5% increase.
This analysis does raise questions about the impact of particulate depositions onto
vegetation, particularly fruits and vegetables. However, a ten-fold increase in particulate
depositions would only appear, at most, to double estimated fruit and vegetable
concentrations, while not impacting grass and fodder calculations.
Particulate deposition was the significant contributor to plant concentrations when
incinerators were the source. Section 10.3 discusses use of the ISC model to model
10-79 ' 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
particulate deposition rates, and use of the model otherwise. While that section notes
several areas of variability and uncertainty, the trend for higher particulate phase air-borne
contaminants at exposure sites as compared to vapor phase probably holds up under
different emission technologies, emission rates, stack heights, topography, climate, and so
on. The ISC model is based on accepted atmospheric transport modeling approaches, and
is widely accepted and used throughout EPA. The emission rates of dioxin-like compounds
used in this assessment are based on data, and limited validation of the model for
downwind impacts shows model results to be consistent with observations. The
conclusion that particulate deposition is of principal concern for vegetation concentration
estimation is probably a reasonable one.
The result that vapor phase transfers are more critical than particulate phase
depositions for above ground vegetation is tentative. The uncertainty in that result is not
only a function of uncertainties concerning particulate phase contributions to plant
contributions, but also the algorithms estimating vapor phase concentrations. Section
10.2.9. above discusses uncertainty issues associated with vapor phase concentrations.
Critical for estimating plant concentrations from air-borne concentrations include the
concentration, the air-to-leaf transfer factor, Bvpa, and the empirical correction factor,
The laboratory experiments used to derive the air-to-plant transfer factor, Bvpa, are
evaluated as sound and appropriate as the starting point for current purposes. However,
some correction is necessary when applying that factor for estimating fruits and vegetable
concentrations due to this mechanism, since this vegetation is distinctly different than the
azalea leaves of the experiments of Bacci, et al. (1990). Leaves are hypothesized as
analogous to outer portions of bulky fruits and vegetables. Also, the literature discussed
above did find that the large majority of fruit and vegetable 2,3,7,8-TCDD residues in outer
portions of the fruit. The empirical parameter, VGag, was introduced to correct for the
difference in azalea leaves and the vegetation pertinent to this assessment. Assignments
to this parameter were made based mostly on a surface volume to whole fruit volume
ratio. This ratio was shown to be in the range of 0.03 to 0.09. An additional correction
was made based on literature showing additional residue loss after peeling or washing.
The final VGag for fruits and vegetables was 0.01 . It was assumed to be equal to 1 .0 for
grass, which is thought to be directly analogous to leaves, and 0.5 for cattle fodder, which
10-80 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
is a combination of leafy hay and corn grain (which includes protected grain and leafy
vegetation).
The comparison of literature contaminant concentration ratios with modeled ratios
above did show that the methodologies of this assessment may, in fact, be
underestimating vegetative concentrations. This could imply that the VG empirical
correction parameters, both for above and below ground, may be too low (the below
ground VG, VGbg, was also 0.01). The surface area volume was estimated assuming a
very thin outer shell of 0.03 cm. Consideration of some within plant translocation would
lead to an assumption of a thicker shell, and a higher surface area volume to whole fruit
volume ratio. Below and above ground vegetable/fruit concentrations are linearly related
to the VG parameters - increasing them to 0.10, possibly a more plausible value, would
increase concentrations by a factor of ten.
A refined surface area volume to whole fruit volume ratio may be one consideration
for further development of an experimentally based empirical parameter such as VG.
Another consideration is the degradation of dioxin-like residues sorbed to vegetation
surfaces. Bacci's data was developed for short time intervals, in the order of hours and
days, and under laboratory lighting conditions. Data is now being developed which
demonstrates photodegradation of dioxin-like compounds in plants in field conditions
(McCrady and Maggard, 1992). The use of VG term of 0.01 for vegetables may already
account for this empirically, but the VG of 0.50 for fodder and 1.00 for grass may result in
predicted concentrations from air-to-leaf transfer being too high.
Finally, one has to consider the uncertainty associated with the empirical transfer
factors, including the Root Concentration Factor, RCF, describing the transfer from soil
water to below ground vegetables, and the air-to-leaf transfer parameter, Bvpa, describing
the vapor phase transfer to above ground vegetation. Both these parameters are critically
a function of the octanol water partition coefficient, Kow, a parameter itself which is
uncertain. The Bvpa is additionally a function of Henry's Constant.
The range of log Kow given in Chapter 2 (Section 2.3.4) for 2,3,7,8-TCDD is 6.15
to 8.5. The value chosen for this assessment is 6.64. Equation (5-21) gives the empirical
relationship estimating the RCF as a function of log Kow. The RCF for log Kow equalling
6.15, 6.64, and 8.5 are 1642, 3915 (which is the value chosen for this assessment), and
105925. Below ground vegetable concentrations are linearly and directly related to RCF;
10-81 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
this might imply that assuming a higher log Kow would lead to greater below ground
vegetable concentrations. However, the equation for estimating such concentrations,
Equation (5-18), shows them also to be an inverse function of the adsorption partition
coefficient, Kd, which is also estimated as an empirical function of log Kow in this
assessment. Increasing Kow also increases Kd, or increases sorption to soil, which tends
to reduce concentrations. Plant concentrations are a function of RCF/Kd. Log Kows of
6.15, 6.64, and 8.5 translate to the following RCF/Kd ratios: 0.19, 0.14, and 0.005.
Therefore, an assumed log Kow of 8.5 results in over an order of magnitude reduction in
below ground vegetation from what was assumed in this assessment, whereas a lower log
Kow of 6.15 results in roughly a 36% increase in estimated concentrations.
A similar dual effect is noted when assuming a different Kow to estimate air-to-leaf
vapor phase transfers. Equation (5-23) shows the transfer factor, B , to be a function of
log Kow. Increasing log Kow to 8.5 over the assumed 6.64 results in a Bvpa over 700
times higher. However, increasing log Kow again would imply an increase in Kd, and
increase of sorption to soil, and a subsequent reduction in volatilization flux. Air
concentrations are decreased by roughly an order of magnitude with a log Kow of 8.5 for
2,3,7,8-TCDD. With roughly two orders of magnitude increase in the transfer factor, but
an order of magnitude decrease in air concentrations, the net effect is to increase the
vapor transfers to plants by an order of magnitude with an increase in log Kow to 8.5.
Henry's Constants are less variable for the dioxins and furan compounds, as seen in
Table 2.2. The value assumed for 2,3,7,8-TCDD, 1.65 x 10~5 atm-m3/mol, is the highest
of listed values, with the lower values being on the order of 1 x 10~6 atm-m3/mol.
Reducing H to this value increases the B parameter by a factor of 1 6. Again, changing
H influences air concentrations - they would decrease by a factor of 4 with the noted
reduction. Therefore, reducing H by an order of magnitude has the net effect of increasing
the vapor phase transfers by a factor of 4.
10.2.12. Beef and Milk Ingestion
Concentrations in beef and milk are a function of cattle ingestion of contaminated
soil, pasture grass, and cattle fodder. Therefore, previous sections are relevant to
estimating beef and milk concentrations. Section 10.2.3. on soil erosion describes
uncertainties with estimating soil concentrations at an exposure site distant from a site of
10-82 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
soil contamination. Section 10.2.11. above describes uncertainties associated with
estimating grass and fodder concentrations.
The first section below describes uncertainties in the ingestion rates of beef and
milk. The next section compares model estimations of beef and milk concentrations with
those that have been found worldwide. An evaluation of alternate modeling approaches is
then made. Finally, uncertainties in the parameters of the algorithm estimating beef and
milk concentrations are described. Table 10.14 summarizes uncertainties associated with
the beef and milk exposure pathways.
10.2.12.1. Beef and Milk Ingestion Rates
Data on rates of milk and beef consumption were taken from surveys summarized
in EPA (1989). Whereas the survey data may lead to adequate estimates for per capita
consumption of these products, EPA (1989) cautions that farm families who home
slaughter or who home produce dairy products may have higher consumption rates. Data
is unavailable for these situations. Another consideration for application to real world rural
situations is that farming and non-farming families may be obtaining cattle food products
from local farms which may also be impacted by dioxin-like compounds. This possibility
was not addressed in this assessment.
The fractions of meat or milk intake coming from the farmer's home supplies was
determined in a survey of 900 rural farm households (USDA, 1966). The 0.44 (44%) of
meat and 0.40 of dairy contact fractions from this survey were, appropriately, proportions
of total dietary intake that is home-produced and consumed by farming families.
Therefore, more certainty is expected for these contact fractions as compared to ingestion
rates.
The trend analysis for the example scenarios in Chapter 9 indicated that the
greatest exposures occur for beef, milk, and fish. Therefore, the rate of consumption of
impacted beef and milk is critical. The range of beef fat consumption noted in surveys
summarized in EPA (1989) is 14.9 to 26.0 g/day, but a single high consumption rate of
30.6 g/day was noted. If this high rate is more typical of home-producing farm
families,than the value of 22 g/day selected for this assessment may be 28% low (28% is
(30.6-22)/(30.6)*100%). The single high rate of 35 g/day of milk fat is significantly
higher than the 8.9-10.7 g/day range noted in EPA (1989) and the 10 g/day ingestion rate
10-83 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.14. Uncertainties associated with beef and milk ingestion exposure algorithms.
Assumption/
Method
Approach
Rationale
Uncertainty
Comments
Ingestion rates
22 g/day beef fat
10 g/day milk fat
Literature showed 14.9-
26.0 g/day beef fat, and
18.8-43 g/day milk fat;
ranges developed from 3
surveys
Shape of distribution of
consumption not well
defined; study and survey
showing 43 g/day milk fat
less well documented than
other 2 surveys.
Beef and milk home producers
may tend to ingest more than
average families.
Contact rates
0.44 for beef fat
0.40 for milk fat
Data from USDA survey
including percent of
annual consumption of
beef and milk homegrown.
Likely to be substantial
differences between families;
some may not home slaughter.
Again, home producers may
obtain more than 44 or 40% of
of beef and milk from their
own supplies.
Beef and milk
fat concentra-
tion estimates
Developed by Fries and
Paustenbach (1990); bio-
concentration factor
multiplied by propor-
tionally weighted concen-
trations of soil, grass,
and fodder.
Their principle premise
was that 2,3,7,8-TCDD
bioconcentrates equally
in beef and milk fat;
their literature survey
developed key parameters
used here as well.
Uncertainties associated with
soil, pasture grass, and fodder
carry over into beef and milk
fat concentrations; other
uncertainties with parameters
as noted below.
Section 10.2.12.3 shows how
current approach is the same
as earlier approaches which used
whole beef and milk biotransfer
factors and similar models for
particle deposition impact to
soil and vegetation.
Key parameters
and assumptions
Bioconcentration factor,
F for 2,3,7,8-TCDD of
5.0; Soil bioavailability
Bs of 0.65; and assump-
tions on proportions of
diet in soil, grass, and
fodder
F developed by Fries and
Paustenbach from data,
which also showed lower
F for higher chlorinated
dioxin-like compounds;
also developed Bs and
tested different diet
assumptions.
Of three noted, assumptions on Fattening of beef cattle prior
proportions of diet most variable to slaughter is not considered
and having the most impact;
hypothetical worst and best
scenarios for cattle ingestion
tested and shown to result in
roughly an order of magnitude
range due to these assumptions
alone.
and could result in 50% or more
reductions in fat concentrations.
Overall: Model predicted beef and milk fat concentrations of 2,3,7,8-TCDD resulting from a 1 ppt soil concentration, chosen to be similar to
low, perhaps background, concentrations found by researchers, compared favorably with beef and milk concentrations taken by researchers
similarly looking for background concentrations in milk or beef and/or sampling from farms with no known nearby source of possible 2,3,7,8-
TCDD contamination. A similar favorable comparison was noted for beef and milk concentrations resulting from incinerator depositions.
Comparison with earlier modeling approaches showed that the current approach is the same as earlier approaches, although mathematically
formulated differently. Earlier approaches also estimated cattle dose of 2,3,7,8-TCDD from contaminated air (directly) and contaminated
ground water - these earlier estimations showed these contributions to be minimal, and they were not considered in this assessment. Finally,
earlier assessments considered the practice of fattening beef cattle prior to slaughter by feeding them residue-free grains. These efforts
estimated over a 50% reduction in beef concentration due to residue degradation or elimination and/or dilution with increases in body fat. This
assessment does not consider this practice. Including a depuration algorithm or other cattle production practices which influence
concentrations in fat should be considered in future refinements of the models in this assessment.
10-84
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
for milk fat may be 71 % too low.
10.2.12.2. Comparison of modeled beef and milk concentra tions with
concentrations found
The example scenario in Chapter 9 demonstrating the on-site source category
(where the soil at the place of residence/farming/exposure is the source of contamination)
had soil concentrations initialized at 1 ng/kg (ppt) 2,3,7,8-TCDD. This concentration was
chosen because it was similar to concentrations of 2,3,7,8-TCDD found in studies where
researchers had measured what they characterized as "rural" or "background" soils. Beef
and milk fat concentrations of 2,3,7,8-TCDD estimated with this soil concentration were
0.4 and 0.1 ppt 2,3,7,8-TCDD, respectively. Assuming fat contents for beef and milk of
0.22 and 0.035, respectively, whole beef and milk concentrations are estimated as 0.09
and 0.001 ppt. Beef and milk fat concentrations for an exposure site located 500 meters
from a hypothetical incinerator, another of the example scenarios in Chapter 9, were 1.3
and 0.8 ppt. Corresponding whole beef and milk concentrations were 0.3 and 0.03 ppt.
The other two source categories were sites of higher soil concentration located near sites
of exposure. One, termed the off-site source category, had a 4 hectare site contaminated
with 2,3,7,8-TCDD at 1 /vg/kg (ppb) located 150 meters from an exposure site. This
concentration was selected based on similar 2,3,7,8-TCDD concentrations found in sites
of elevated contamination, such as Superfund sites. Soil concentrations at the site of
exposure were estimated to be 0.61 ppb, or 610 ppt. Concentrations in beef and milk fat
were 170 and 46 ppt, respectively, which corresponds to whole product concentrations of
37 and 1.6 ppt. The second source category with elevated soil concentrations near a site
of exposure is the ash landfill category. Initial 2,3,7,8-TCDD concentrations on a 27
hectare landfill site were estimated based on expected ash concentrations. Landfill
2,3,7,8-TCDD concentrations were 700 ppt, and nearby exposure site soil concentrations
were 563 ppt. Resulting beef and milk fat concentrations were 161 and 45 ppt,
respectively, and whole product concentrations were 35 and 1.6 ppt.
Several studies have measured concentrations of dioxin-like compounds in farm
products (dairy, beef, etc.) where it is not known whether the products are impacted by a
nearby known source - market basket surveys may be characterized this way - or where
samples were obtained on farms chosen specifically because no identifiable source was
10-85 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
nearby. It may be appropriate to compare results from these types of studies to results
from the first source category noted above, where soil concentration levels were low and
maybe typical of background or ambient levels. In such a survey taken in Germany (Beck,
et al. 1989), the fat content of meat and eggs averaged 0.23 ppt, and for dairy products
averaged 0.14 ppt 2,3,7,8-TCDD. Another random sample of food products from
Germany found average cow's milk fat concentration to average 0.4 ppt for ten samples,
ranging up to 1.9 ppt (Furst, 1990). Concentrations of 2,3,7,8-TCDD were not found in 3
beef fat samples, with a detection limit of 0.5 ppt (Furst, 1990). Samples of whole milk
from seven farms in England chosen because they were well removed from known sources
of dioxin showed a mean concentration of 0.009 ppt (detection limit = 0.004 ppt; Startin,
et al., 1990). No 2,3,7,8-TCDD residues were detected in three whole milk samples in
Switzerland (detection limit = 0.013 ppt) where no identifiable source of dioxins was
nearby (Rappe, et al., 1987).
The example scenario results from the incinerator source showed beef and milk
concentrations higher than those for the first example source category, but still near 1.0
ppt. The literature also shows comparable trends for sampling which occurred near
incinerators. The study sampling remote farms in England also sampled two farms near
incinerators and two farms near industrial centers. Whereas samples from remote farms
averaged 0.009 ppt for whole milk, two concentrations near the incinerators were 0.034
and 0.036 ppt 2,3,7,8-TCDD, and the samples near the industrial centers were 0.043 and
0.081 ppt (Startin, et al., 1990). The study noted above sampling milk from locations
remote from 2,3,7,8-TCDD sources in Switzerland, and not detecting residues, also
sampled three locations that were within 1000 meters of incinerators. Whole milk
concentrations near the incinerators were 0.021, 0.038, and 0.049 ppt.
Sampling of beef and milk near areas of elevated soil concentrations, or where
cattle were raised on soils with known high concentrations of 2,3,7,8-TCDD, were not
found in the literature. There are some studies on other animals indicating high tissue
concentrations in areas of high soil contamination of 2,3,7,8-TCDD. Lower, et al. (1989)
studied animal tissues for wild animals in the abandoned town of Times Beach, Missouri,
and compared their results for similar wild animals tissue concentrations found in Eglin Air
Force Base in Florida; Seveso, Italy; and Volgermeerpolder in Holland. With 2,3,7,8-TCDD
soil levels in these areas in the hundreds to thousands of ppt, tissue levels for earthworm,
10-86 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
mouse, prairie vole, rabbit, snake, and liver samples from some of these animals, were in
the tens to hundreds to thousands of ppt.
There is an episode of beef and dairy cows (as well as lambs and pigs) being raised
on lots where the soil was heavily contaminated with polybrominated biphenyls (PBB;
details of the extent and cause of contamination are given in Fries and Jacobs, 1986;
results cited are from Fries, 1985). Soil concentrations to which dairy and beef cows
were exposed were 830 and 350 //g/kg (ppb), respectively. Body fat of the dairy cows
had PBB concentrations of 305, 222, and 79 ppb (dairy heifers, primaparous dairy, and
multiparous dairy, respectively). Body fat for the beef cows exposed to 350 ppb soil
levels were 95 and 137 ppt (cows and calves, respectively). Milk fat concentrations from
the primaparous dairy and multiparous dairy cows exposed to 830 ppb soil levels were 48
and 18 ppt.
Fries estimated a quantity which is also useful for purposes of comparison - this
quantity is the ratio of concentration in animal fat to concentration in soil. His justification
for deriving this ratio is that soil was speculated as the principal source of body burdens of
PBB in the data listed above. For the source categories where contaminated soil is the
source of dioxin-like compounds, the on-site, off-site, and ash landfill source categories, a
similar assumption is warranted. Ratios he derived for body fat of dairy heifers ranged
from 0.10 to 0.37, while it was 0.02 and 0.06 for milk fat. For body fat of beef cows,
these ratios were 0.27 and 0.39. Fries also measured a ratio of 1.86 for sows and gilts.
He attributes much higher sow ratios due to their tendencies to ingest more soil.
Analogous ratios can be derived for the contaminated soil source categories, and for beef
and milk fat. For the onsite source category with low soil concentrations, beef fat to soil
and milk fat to soil ratios were 0.4 and 0.1, respectively. For the off-site and ash landfill
source categories, ratios were the same at 0.28 for beef fat and 0.08 for milk fat. These
compare favorably with PBB ratios derived by Fries (1985).
The incinerator source category does not have comparable results. The beef and
milk fat to soil ratios were 1.00 and 0.61. The reason this occurs is that vegetation is
more impacted by emission depositions as compared to soil transfers. The soil
concentration resulting from the emission depositions is 1.27 ppt, which compares to the
1.00 ppt assumed for the on-site source category. However, the grass and fodder
concentrations resulting from incinerator depositions are 0.20 and 0.14 ppt, respectively,
10-87 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
whereas in the on-site source category, grass and fodder concentrations are 0.02 and
0.01 ppt. With an order of magnitude higher vegetation concentration resulting from
incinerator emission depositions, beef and dairy impacts are greater than estimated when
soil contamination is the source. It is not appropriate to compare the 1.00 and 0.61 fat to
soil ratios with PBB ratios developed by Fries, for this reason.
10.2.12.3. Alternate Modeling Approaches for Estimating Beef and Milk
Concentrations
Webster and Connett (1990) compared five models which estimated the 2,3,7,8-
TCDD content of cow's milk from 2,3,7,8-TCDD air contamination. The five models were
described in Michaels (1989), Connett and Webster (1987), Stevens and Gerbec (1988),
Travis & Hattermer-Frey (1987), and McKone and Ryan (1989). Ironically, a sixth model
by Fries and Paustenbach (1990), noted by Webster and Connett as available but received
too late for inclusion in their article, formed the basis for the approach taken in this
assessment.
All five models compared by Webster and Connett have the same basic framework.
Particulate-bound 2,3,7,8-TCDD deposits onto the ground and vegetation (cattle fodder
and pasture grass). Algorithms to estimate resulting vegetation and soil concentrations in
these models are the same ones used in this approach, although parameter assignments
are different. A daily dosage of 2,3,7,8-TCDD to the cattle is calculated and converted to
a concentration in whole milk using a "biotransfer factor". This same structure was used
to estimate concentrations in beef, using a beef biotransfer factor different than the milk
biotransfer factor. Mathematically, this is expressed as:
Cm,b = Fm,b Dose (10-6)
where:
Cm b = concentration in whole milk/beef, mg/kg
Fm b = milk/beef biotransfer factor, day/kg
-------
DRAFT-DO NOT QUOTE OR CITE
fm b = fat content of milk/beef, unitless
Q = daily mass intake of cattle in experiment, kg
Dose = total daily dose of 2,3,7,8-TCDD, mg/day
= I (8j * Cj * Qj)
3: = relative bioavailability on intake vehicle j (soil, air, vegetation, etc)
Cj = concentration of 2,3,7,8-TCDD in vehicle j, mg/kg (or equivalent
units)
Q: = mass of vehicle j intake, kg (or equivalent units)
Further details on the models can be found in their primary references and in Webster and
Connett's comparison. Some highlights, including comparisons of the five approaches to
the approach taken in this assessment, are:
1) Two of the approaches, that of Stevens and Gerbec (1988), and McKone and
Ryan (1989), consider inhalation of contaminated air by cattle to contribute to their daily
dose of 2,3,7,8-TCDD. One of the approaches, that of Travis and Hattemer-Frey (1987),
considers ingestion of contaminated water by cattle. A later assessment by Travis and
Hattemer-Frey (1991) has all the components of their earlier assessment, and adds cattle
inhalation exposures. This assessment does not consider cattle inhalation of contaminated
air nor ingestion of contaminated water in estimating beef and milk concentrations.
However, these intakes were shown to be insignificant when estimated by these
researchers. Stevens and Gerbec estimate inhalation contributions to be less than 0.05%
(0.0005 in fractional terms) of total daily dose, or an essentially insignificant amount.
Travis and Hattemer-Frey (1991) estimate inhalation to contribute between 0.3 and 1.0%
to milk and beef concentrations, respectively. McKone and Ryan (1989) did not provide
sufficient information to easily determine the relative contribution of inhalation on
estimation of cattle beef and milk concentrations by their estimations. Travis and
Hattemer-Frey (1987, 1991) estimate water contributions to be less than 0.01% (0.0001)
of total daily cattle dose of 2,3,7,8-TCDD.
2) None of the approaches considered vapor phase transfers from air to plant,
although Webster and Connett recommended its inclusion in their article. The later
assessment by Travis and Hattemer-Frey (1991) on 2,3,7,8-TCDD did include vapor phase
transfers into vegetation consumed by cattle. According to results of the example
10-89 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
scenarios in this assessment, these transfers appear to be particularly critical when the
source of contamination is soil. This is discussed in Section 10.2.11.3, where vapor
phase transfers were shown to contribute over 90% of 2,3,7,8-TCDD contamination of
above ground vegetation. Where the source was incinerator emissions, however,
particulate depositions dominated plant concentrations, with contributions between 88 and
99% of total above-ground plant concentrations. Three of the assessments evaluated by
Webster and Connett were specific to incinerator emissions and estimated plant
concentrations based on particulate deposition (Stevens and Gerbec (1988), Connett and
Webster (1987), and Michaels (1989)), so an omission of vapor phase transfers in these
assessments may not have been critical. However, one of them assumed particle
depositions did not impact vegetation (Michaels (1989)), but impacted soil, and calculated
plant concentrations based on a root uptake algorithm. Webster and Connett discuss why
this particular approach, which neglected particle deposition onto vegetation, greatly
underestimated vegetative concentrations.
3) One of the assessments, Stevens and Gerbec (1988), assumed a specific mode
of beef cattle production - in particular, one which included placing the cattle on a grain-
only diet for fattening prior to slaughter. They assumed the grain was residue free, and
therefore, residues in cattle would depurate during the last 1 30 days of their lives.
Assuming a half-life of 2,3,7,8-TCDD in cattle of 11 5 days, they showed a 54% reduction
in beef concentrations due to this practice. Fries and Paustenbach (1990) also identified
cattle production practices as critical, and evaluated the impact of a 120-day grain only
diet prior to slaughter in their modeling. They note that cattle can gain as much as 60-
70% in body weight, so dilution can also result in lower beef concentrations at slaughter.
Procedures are not described in this assessment to estimate the reduction of
concentrations in beef and milk fat due to depuration or dilution periods, although the
procedure is simple. Assuming first order kinetics sufficiently describes reduction in
concentrations during a period prior to slaughter, the fractional reduction during such a
period is given as, 1 - exp(-kdt), where kd is the depuration rate constant, in days"1, and t
is the depuration period, in days. The rate constant can be estimated from the depuration
half-life, HL, as 0.693/HL. The 115 day half-life assumed by Stevens and Gerbec (1988)
corresponds to a rate constant of 0.006 day"1, and assuming a 130 day depuration period,
the fractional reduction is easily calculated as 0.54 (i.e., 1 - exp(-kdt)). The amount
10-90 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
remaining after 130 days is estimated as the initial amount multiplied by 0.46 (i.e., exp(-
M»-
4) Two of the assessments did not assume any cattle ingestion of contaminated
soil, and two of the assessments estimated the contribution to milk concentrations due to
ingestion of contaminated soil was minor at 1 and 2%. Only one of the assessments,
Travis and Hattermer-Frey (1987), estimated any significant impact due to soil ingestion,
attributing 19% of the concentration due to ingestion of contaminated soil. Their later
assessment (Travis and Hattemer-Frey (1991)) estimated soil to contribute 29 and 20% of
beef and milk concentration estimations, respectively. They estimated this high a
contribution by contaminated soil even though they assumed that contaminated soil
comprised 1% of the total dry matter intake by cattle. Fries and Paustenbach (1990)
recognized the importance of cattle soil ingestion, evaluating scenarios where cattle soil
ingestion ranged from 1 to 8% of total cattle dry matter intake. This example scenarios in
Chapter 9 assumed that beef cattle ingestion of contaminated soil was 8% of their total
dry matter intake, and 2% of a dairy cattle's intake was contaminated soil. The
percentage of beef and milk concentrations of 2,3,7,8-TCDD attributed to soil, fodder, and
pasture grass, when soil contamination is the source and when incinerator emissions are
the source, are given in Table 10.15. As seen, cattle soil ingestion is a significant
contributor to beef and milk concentration estimations using procedures and assumptions
in this assessment, explaining 11 to 71 % of the beef and milk concentrations. Two key
explanations for the differences in results in this assessment versus the ones in the
literature include: the other assessments assumed less soil ingestion, 0.5% in Stevens and
Gerbec (1988) and 1-3% in Travis and Hattemer-Frey (1987) and McKone and Ryan
(1989), and differences in soil and plant concentration estimations (due to different mixing
depth assumptions for soil concentrations, particle deposition rates for plant
concentrations, and so on).
The various approaches used different parameter values for the biotransfer factor,
intake quantities, initial air concentrations, and others. The critical focus of the Webster
and Connett (1990) comparison, is the milk fat bioconcentration factor, BCFmf. As shown
in Equation (10-6), the biotransfer factor, Fm/ is estimated using experimental data which
yields a milk fat bioconcentration factor, BCFmf. Experiments most relied upon by these
modelers are those described in Jensen, et al. (1981), and Jensen and Hummel (1982). A
10-91 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.15. Estimated relative importance of soil, pasture grass, and fodder ingestion to
the estimation of beef and milk concentrations when soil and incinerator emissions are the
source of 2,3,7,8-TCDD contamination.
Percent of beef and milk fat concentrations due to:
Description Soil Pasture Grass Fodder
1. Soil Contamination
Beef
Milk
II. Incinerator Emissions
Beef
Milk
71 29
51 7
26 73
11 11
<1
42
1
78
key difference in the early modeling approaches is the interpretation of these two and
other studies and the resulting assignment of BCFmf, with values ranging from 5 to 25.
Webster and Connett (1990) discuss issues of experimental interpretation.
Parameter assignments and assumptions (cattle soil ingestion vs. no ingestion, etc.)
obviously all impact estimations and can be a critical source of variation and uncertainty in
estimates of beef and milk concentrations. The uncertainty associated with the modeling
framework described above was explored by McKone and Ryan (1989) using Monte Carlo
techniques. They found that the 90% confidence range for human exposure to 2,3,7,8-
TCDD, where the source was air contamination and the human exposure route was
through milk, spanned two to three orders of magnitude.
The approach taken by all five researchers centers on the milk biotransfer factor,
abbreviated Fm in Webster and Connett (1990) and in units of day/kg. Beef
bioaccumulation was modeled in the same way using a beef biotransfer factor, Fb. Travis
and Arms (1988) developed this concept to the fullest, taking several data sets from the
10-92 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
literature on a variety of contaminants and animals, to derive empirical formulas for Fb and
Fm, which they termed Bb and Bm, as a function of contaminant octanol water partition
coefficient, Kow:
log Bb = log Kow - 7.6 (10-7a)
log Bm = log Kow - 8.1 (10-7b)
Given a log Kow of 6.64 for 2,3,7,8-TCDD (assumed in this assessment), Bb is solved for
as 0.110 and Bm is solved for as 0.034. Travis and Hattemer-Frey (1991) use 0.80 and
0.03 for Bb and Bm.
Simple transformations can show how the earlier approaches and the approach of
Fries and Paustenbach (1990), the one used in this assessment, are the same. First, the
concentration of dioxin-like compounds in the fat of beef and milk is given in this
assessment by (also as Equation (5-24) in Chapter 5):
Cfat = (FDFSBSACS) + (FDFgACg) + (F DFf AC/) (10-8)
where:
Cfat = concentration in beef fat or milk fat, mg/kg
F = bioconcentration ratio of contaminant as determined from cattle
vegetative intake (pasture grass or fodder), unitless
DF_ = fraction of cattle diet that is soil, unitless
o
Bs = bioavailability of contaminant on the soil vehicle relative to the
vegetative vehicle, unitless
ACS = average contaminant soil concentration, mg/kg
DF = fraction of cattle diet that is pasture grass, unitless
AC = average concentration of contaminant on pasture grass, mg/kg
y
DFf = fraction of cattle diet that is fodder, unitless
ACf = average concentration of contaminant in fodder, mg/kg.
Transformation steps are: 1) factor out the F from Equation (10-8) , 2) multiply the top
10-93 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
and bottom of Equation (10-8) by total dry matter intake by cattle, Q; 3) the multiplication
of Q by the diet fraction terms in the numerator, DFs, DFg, and Dff, gives the values for
soil dry matter intake, Qs, grass - Qg, and fodder - Of, 4) with F factored out, and Q*DFs
replaced by Qs, etc., the parenthetical now reads, (Qs*Bs*ACs + Qg*ACg + Qf*ACf) -
this is the "Dose" term defined above in Equation (10-6), 5) finally, multiply the right hand
side of Equation (10-8) by fat content, say fm for milk, which would transform the right
and hence left hand side of that equation to whole product concentration. Transformed
Equation (10-8) is analogous to Equation (10-6):
C = -^2 [ (Qs Bs ACS ) + ( Qg ACg ) + ( QfACf ) ] (10-9)
Fries and Paustenbach (1990) discuss one critical theoretical assumption not
explored in the earlier literature: that 2,3,7,8-TCDD bioaccumulates equally in beef fat and
milk fat - that the BCFmf and BCFbf are equal. The differences in observed concentrations
in beef and milk are attributed to differences in the diets of cattle raised for beef versus
those raised for milk. The key difference Fries and Paustenbach cite is the tendency for
beef cattle to graze while lactating cattle are more often barn fed. Grazing cattle intake
more contaminated soil than barn fed cattle. Fries and Paustenbach justify their selection
of 5.0 for F, the bioconcentration factor, for 2,3,7,8-TCDD, which is also used in this
assessment. They also derive F for higher chlorinated dioxin-like compounds from
experimental data, noting that the F value is less with higher chlorination. Webster and
Connett (1990) made the analogous observation, saying that 2,3,7,8-TCDD equivalents
transferred from air to milk less efficiently than 2,3,7,8-TCDD.
Some conclusions from this analysis of earlier efforts for estimating
bioconcentration in beef and milk are:
• Although the framework of the earlier approaches looks different than the
framework used in this assessment, they are actually the same with a simple
mathematical transformation;
• The possible dosage to cattle of 2,3,7,8-TCDD via contaminated air or water
was considered in earlier assessments, but was not found to be a significant
pathway, and was not considered in this assessment;
10-94 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Earlier assessments did not consider soil ingestion to the extent considered
in this assessment; the results of this assessment are that 11-71 % of the
beef and milk concentrations were attributed to contaminated soil ingestion
by cattle;
Earlier assessments did not consider vapor phase transfers to vegetation
consumed by cattle; this assessment implies that this transfer is particularly
critical for soil contamination, but less critical for incinerator emissions;
Even though the structure of the analysis has been consistent from the
earlier to the current approaches, different assumptions on parameter values
greatly impacts modeling results. The critical bioconcentration factor, earlier
termed BCFm (for milk) and termed F in this assessment, was assumed to be
5 to 25 for 2,3,7,8-TCDD in different assessments. This assessment uses
an F value of 5 for 2,3,7,8-TCDD. Using Monte Carlo techniques on this
model structure for estimating human exposure to milk resulting from air
contamination of 2,3,7,8-TCDD, McKone and Ryan (1989) showed a 90%
confidence interval spanning 2 to 3 orders of magnitude.
10.2.12.4. Uncertainty of Model Parameters Estima ting Beef and Milk
Concentrations
Earlier sections have discussed the uncertainty of soil and vegetation concentration
estimations. Needless to say, any differing assumption or parameter value which impacts
soil and/or vegetation concentrations will also impact beef and milk concentrations.
Exercises undertaken to evaluate the sensitivity of model estimations to soil and
vegetation concentrations were not undertaken once again to evaluate the modeled impact
to beef and milk. However, beef and milk concentrations are linearly related to soil and
vegetation concentration - see Equation (10-10) above. Using the information in Table
10.15 above, and referring to earlier sections for other sensitivity information, one can
note key relationships and estimate impacts. For example, it is easily shown that beef and
milk concentrations are linearly related to soil concentrations when soil is the source of
contamination. Specifically, vapor-phase air concentrations are linearly related to soil
concentrations, and vegetation concentrations are principally a function of vapor phase
10-95 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
transfers for the soil contamination source categories. Therefore, doubling soil
concentrations will double the porti.on of the impact to beef and milk concentrations due to
soil ingestion, and will also result in a doubling of vegetation concentrations; the net
impact would appear to be roughly a doubling of beef and milk concentrations with a
doubling in soil concentration. A similar analysis shows that particle deposition rates and
concentrations are the principal quantities for beef and milk concentrations for the stack
emission source category. Because of the greater impact of vegetation to beef and milk
concentrations when incinerator emissions are the source (see Table 10.13), assumptions
and quantities involved in plant concentration estimation for depositing particles become
critical.
This section will focus on the three remaining areas of variability and uncertainty.
One is the parameter describing the relative bioavailability of soil in relation to plants, Bs.
The second is the bioconcentration factor, termed F. The third is the dietary assumptions
of beef and dairy cattle - the proportion of diet in soil, pasture grass, and cattle fodder, as
well as the assumption that cattle fodder and pasture grass are grown on the soil that is
contaminated and is therefore impacted. The basis for assignments of all parameters in
these areas was provided in Fries and Paustenbach (1990). Their justifications will be
noted, but further details are provided in that and supporting references.
The literature survey done by Fries and Paustenbach (1990) showed that soil was
likely to be a less efficient vehicle for internal absorption than cattle feed, but not by
much. They hypothesized that absorption of 2,3,7,8-TCDD with cattle feed as the vehicle
was approximately 50%, and that absorption with soil as the vehicle was 30-40%. This
provided the basis for the assignment of 0.65 for Bs. It is unlikely that Bs would vary
greatly. Tests were run for Bs equal to 0.5 and 0.8. The response to estimated beef and
milk fat concentrations to these changes, when soil was the source of contamination, was
minimal. Beef concentrations increased and decreased 16% with an increase to 0.8 and a
decrease to 0.5. Milk concentrations increased and decreased 10% with the same
changes. There was an even smaller impact for the incinerator source with these changes
in Bs. Beef concentrations increased and decreased by 6%, and milk concentrations were
impacted by only 2%.
The bioconcentration factor, F, applicable to both beef and milk fat, was justified by
Fries and Paustenbach (1990) to be in the range of 4 to 6 for 2,3,7,8-TCDD. They note
10-96 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
that earlier estimates in the literature based on experiments placing F in the range of 12 to
above 20 assumed that 100% of the delivered dose was absorbed. Evidence suggests
that percent absorption is more appropriately 50%. They also note that some of an earlier
estimates by Jensen, et al. (1981) were based on a 28-day feeding trial and then
projecting to steady state. Sixty-day studies of cattle ingestion of PCBs (Fries, 1982)
estimated F at about 5, and other studies running 160 days and involving cattle ingestion
of hexa-, hepta-, and octa-CDDs, estimated F at 2 and below (Parker, et al., 1980).
The analysis by Fries and Paustenbach (1990) suggests that there is not much
variation in F for 2,3,7,9-TCDD. There may be an argument that an F for milk should be
lower than an F for beef since lactation is a form of residue elimination. However, the
data is not available to support this position, and Fries and Paustenbach (1990) argue that
the principle variation noted in milk and beef fat concentration is due to diet. In any case,
milk and beef fat concentrations are linearly related to F, as seen in Equation (5-24) above;
doubling F to 10 would double concentrations of 2,3,7,8-TCDD, and so on.
The impact of different dietary assumptions can be evaluated with variations in the
parameters describing the proportion of diet in soil, grass, and fodder, and also the
parameters describing the proportion of contaminated land to which the cattle are
exposed. The latter set of parameters were used to determine the average concentrations
on soil, grass, and fodder. For the example scenarios, it was assumed that all the soil,
grass, and fodder were impacted - all soil ingested had concentrations initially specified or
solved for, etc.
Rather than define several alternate assumptions, a set of extreme assumptions
were defined and tested. Currently, for estimating beef concentrations, the proportion of
dietary intakes are 8% soil, 90% pasture grass, and 2% fodder, all of which are impacted
by 2,3,7,8-TCDD. A worst-case scenario might have cattle ingesting 15% soil, 83%
pasture grass, and 2% fodder (Fries and Paustenbach (1990) suggest a bounding estimate
for cattle soil ingestion of 18%). A reasonable best-case scenario might have beef cattle
ingest 2% soil, 65% pasture grass, and 33% residue-free grain or other fodder. These
scenarios were tested for source categories involving contaminated soil and for the stack
emission source category. For the soil source categories, the worst case assumptions
increased beef concentrations by 60% and the best case assumptions decreased beef
concentrations by 61 %. For the emission source category, the worst case assumption
10-97 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
increased beef concentrations by 17% and the best case reduced concentrations by 40%.
The current assumptions for lactating cattle were 2% soil, 8% pasture grass, and
90% fodder, all of which were impacted. A worst-case scenario might have dairy cattle
pastured more with 4% soil intake, 25% pasture grass intake, and 71 % fodder, all
impacted. A best-case scenario might have less soil ingestion at 1%, some pasturing with
9% pasture grass, and 90% residue-free fodder. For the soil source categories, the worst
case scenario increased milk concentrations by 61%, and a decrease of 66% was noted
for the best case scenario. For the stack emission scenario, the worst case increased
concentrations by 16% and the best case decreased concentrations by 83%.
This analysis shows that there can be nearly an order of magnitude difference in
predicted fat concentrations between the highest and lowest estimates due to dietary
assumptions, with the assumptions for the example scenarios being near the middle of this
possible range. Assuming unimpacted fodder can evaluate the impact of a lower lifetime
intake of contaminant. However, the practice of fattening beef cattle with hypothesized
residue-free grain prior to slaughter was not evaluated in this exercise. Fries and
Paustenbach (1990) speculated that lower fat concentrations would result from a 60-70%
weight gain and dilution effects during this period, while Stevens and Gerbec (1988)
assumed elimination would reduce body fat concentrations by roughly 50%. Either way,
this assessment may have overestimated beef fat concentrations by a factor of two if the
practice of fattening prior to slaughter occurs for a specific site.
10.3. UNCERTAINTY AND VARIABILITY IN ANALYSIS OF DIOXIN-LIKE COMPOUNDS
EMITTED FROM MUNICIPAL SOLID WASTE INCINERATION
Chapter 6 described the basis for evaluating the environmental fate and transport of
dioxin-like compounds that may be discharged to the atmosphere from the stack of a
municipal solid waste incinerator. The procedures were applied to a hypothetical MSW
incinerator that was arbitrarily located in Tampa, Florida. Uncertainty could be analyzed
using Monte Carlo Simulation techniques to generate probability distributions of input
parameters used in the modeling of the emissions from the incinerator. If it is known how
the data is distributed, e.g., normal distribution, log-normal distribution, etc., the random
values of these parameters could be generated. Key to this analysis is a relatively large
set of measured values. With regard to the modeling of emissions of the dioxin-like
10-98 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
compounds, not enough discrete measurements of congener-specific emissions under
varying conditions and operations have been ascertained. The limiting factor has always
been the analytical costs involved. Therefore, this section is not a true uncertainty
analysis, rather it addresses basic variability in the key data inputs and outputs. Because
the algorithms supporting the air dispersion and deposition analysis are linear, variability
can be estimated if only one dependent variable to the analysis is changed, and the other
variables are held constant. Variability can be addressed in terms of the spread of
modeled input values and their effect on the spread of the modeled output values of the
estimated environmental impacts. A second issue is the estimated extent or magnitude of
change.
The principle components of the analysis of estimating spacial and temporal ground-
level concentrations of the stack emissions of dioxin-like compounds from the hypothetical
incinerator are the following:
(1) The probability of emission of dioxin-like compounds from the stack of the
incinerator.
(2) The quantity or magnitude of release of dioxin-like compounds from the stack
of the incinerator.
(3) The representativeness of the hypothetical incinerator in comparison to actual
facilities.
(4) The estimation of dilution, dispersion, wet and dry surface deposition of the
dioxin-like compounds using the Industrial Source Complex Model.
(5) Comparison of ambient measurements with the predicted ambient air
concentration of dioxin-like compounds.
Each of these elements will be discussed separately.
10.3.1. The Probability of Emission Of Dioxin-Like Compounds
The emission of PCDD and PCDFs from the stacks of municipal solid waste
combustors has been well characterized through stack gas sampling and laboratory
analysis. It is certain that all incineration systems that only incinerate municipal solid
waste will emit chlorinated congeners of dibenzo-para-dioxins and dibenzofurans
(Rappe,1991; Cleverly, 1991; U.S.EPA,1987,1989; Siebert, 1991; OTA,1989). There are
no instances whereby dioxin has not been analytically determined to be present in the
10-99 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
stack gas emissions of an operating system. This is true regardless of incinerator design,
the type of incinerator technology, the geographical location of the facility, whether
processed or unprocessed refuse is incinerated, whether the incinerator is or is not
equipped with an air pollution control device, or by the type of air pollution control system
that is employed. Therefore the probability that any incinerator burning municipal solid
waste will emit to the atmosphere PCDDs and PCDFs is absolute. In other words the
probability of occurrence in the stack emissions can be expressed as Pr = 1, or 1 chance
in 1.
The emission of congeners of dioxin-like polychlorinated biphenyls is highly
uncertain. There exists only limited data of PCB emissions from MSW incinerators, none
of which have been speciated on a congener-specific level. The emission of dioxin-like
congeners of PCBs was inferred from the limited data of homologue groups of PCBs by
applying an assumption that the specific congener has a probability of occurrence equal to
the ratio of the individual congener to all congeners existing within the homologue group.
There is no way of addressing the reliability of this estimation procedure without congener-
specific PCB emissions data. However, if it is the case, as the author has assumed in
Chapter 6, that PCB emissions are a consequence of PCB contamination present in the
MSW that is burned in the incinerator, and not a product of chemical synthesis, then the
emission database can be used in infer a probability or likelihood that PCBs will be emitted
from any MSW incinerator. A probability can in implied from recent test reports of the
detection of total PCBs in the raw municipal solid waste taken as a sample prior to
combustion in an incinerator (U.S. EPA; Environment Canada, 1991). A total of 13
separate samples were taken corresponding to 13 discrete days of the month. No PCBs
were measured (level of detection: 0.1 ng/g) in the MSW in 5 of the 13 sampling days.
From this limited data a probability of 5 chances out of 13 of detecting measurable PCBs
on any day of random sampling can be implied for this particular incinerator, or Pr of 0.39.
Again this is only a relative probability based on a small sample that may be specific only
to the conditions of the MSW charged to that particular incinerator. There simply is no
additional data that could be found in the literature to increase confidence in the
estimation. What may be important is the fact that the emission of PCBs may not be a
continuous process during MSW combustion. This observation can be sustained by
comparing the probability of release of PCB from the stack of the incinerator with the
10-100 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
MSW prior to combustion. The stack tests showed the exact frequency of detection over
the same sampling interval, e.g., PCBs were detected in the flue gas 5 out of 13 test
samples, or a Pr of 0.39. The stack release of PCBs will be highly dependent on
contamination in the raw refuse and the failure of the incineration system to sustain 100
% thermal destruction of the molecules.
Tiernan (1992; personal communication from T. Tiernan, Professor of Chemistry
and Director of the Toxic Contaminant Research Program, Wright State University, Dayton
Ohio, July, 1992) has reviewed the "reasonableness" of the PCB homologue emissions
data as derived in Table 6.7 of Chapter 6. The assumption made to derive that table was
that PCB stack emissions are a function of the incomplete thermal destruction of total
PCBs present as a contaminant in the MSW. Working with this premise, Tiernan assumed
an equal mixture of Aroclors 1016, 1242, 1248, 1254, and 1260 in the MSW as
comprising the total PCBs present. By summing the congeners present in this equal
distribution of PCB Aroclors for dichloro- through octachloro-biphenyl homologue groups, a
normalized distribution can be constructed. This distribution can be compared to the
predicted distribution derived in Table 6.7. This comparison is given in Figure 10.1. As
can be seen, the relative distribution of congeners derived by Tiernan closely approximates
the relative distribution of emission factors indicated in Table 6.7. The exception seems to
be PentaCB. This may lend further support to the theory that PCB contamination in the
MSW is responsible for PCB air emission from the stack of the MSW incinerator.
10.3.2. The Quantity of Release of Dioxin-Like Compounds From the Stack of the
Incinerator
Emission factors of dioxin-like compounds were estimated from test reports of
technologies that are most similar in design and operation as the hypothetical MSW
incinerator. The use of multiple test reports from multiple facilities of the same basic
technology should increase the confidence that the predicted emissions of the
contaminants will be in good agreement and fairly representative of emissions from actual
facilities of this size, e.g., 2727 metric tons of MSW combusted per operating day.
Simplified statistical techniques were used to derive a representative emission factor for
each dioxin-like congener. It was assumed that the emissions data for 10 individual
incinerators is normally distributed. It was assumed that only state-of-the art incineration
10-101 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
c
o
(A
O
TJ
01
*-
a
E
**
M
UJ
Tiernan Distribution
*— Chapter 6 Distribution
Figure 10.1. Comparison of relative distributions of MSW stack emissions of PCB
congeners as derived by Tiernan (see text. Section 10.3.1} and as given in
Table 6.7, Chapter 6.
systems would be included in the universe of emissions for purposes of this analysis.
Itwas assumed that only data derived under ideal test conditions would be used. No data
was included if the test report showed malfunctioning of the incinerator during stack
testing. Thus only mass burn heat recovery systems of modern features design to delimit
the formation of organic compounds were selected. Only tests reporting the identification
of individual congeners of dioxin-like compounds were used, when available, to minimize
the necessity for extrapolating unknown values. These selection criteria may present a bias
toward underestimating the magnitude of potential emissions in the hypothetical case.
When assuming a normal distribution of the data, the arithmetic mean becomes a
measure of central tendency, that is, a measure of the center of the distribution of the
finite data set. Statistical measures of the dispersion of the data about the point of central
tendency are the standard deviation (s), and the mean absolute percentage deviation
(MAPD). In a normal distribution of data, (s) is equal to the positive square root of the
variance of the measurements. The Empirical Rule states that given a distribution of
measurements that is approximately bell-shaped, the mean plus or minus one standard
deviation will contain approximately 68% of the measurements. The MAPD is defined as
the absolute deviation of each measurement divided by the mean (times 100%). This gives
the percentage deviation (or difference) of every emission measurement from the mean.
10-102
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Subsequent addition of these percentages divided by the number of measurements, in this
case 10 facilities, yields the mean absolute percentage deviation (MAPD).
To investigate the variability of the magnitude of emissions, statistical analysis will
focus on three congeners: 2,3,7,8-TCDD (TCDD), 2,3,7,8-TCDF (TCDF), and OctaCDD
(OCDD). Table 10-1 6 is a summary of the statistical analysis of dispersion of the emission
measurements for these three specific congeners. The mean mass emission factor for
TCDD, OCDD, and TCDF derived from the testing of 10 facilities chosen for this analysis is
4.33//g/metric ton (Mt) of MSW burned, 150/yg/Mt, and 49.1 //g/Mt, respectively. The
range of the measurements is from 0.28 to 15.6 //g/Mt for TCDD; 1 2.8 to 92.3 /yg/Mt for
OCDD, and 4.06 to 129 /yg/Mt for TCDF. The MAPD is 68.36 % for TCDD, 97.86 % for
OCDD, and 83.83 % for TCDF. The standard deviation for the set of measurements is
5.03/yg/Mt for TCDD, and 195//g/Mt for OCDD, and 47.9 for TCDF.
It appears from this data set that the emission of TCDD, OCDD and TCDF is highly
variable. This is the case even when efforts were made to segregate emissions data by
design type (mass burn energy recovery), and whether or not combustion engineering
principles employed represent state-of-the-art. Although there is an absolute certainty that
any incineration system burning MSW will emit dioxin from the stack to the atmosphere, it
may be that the magnitude of concentration, hence the magnitude of emission, is a
random event. This may be in keeping with the theories supporting a de novo synthesis on
the surface of particulate matter in regions of the incinerator where the flue gases have
cooled to between 200° and 350° C. The rate limiting factors are: relative range of
temperature; amount of chlorine in the MSW; the amount of HCI produced and required for
acting as a chlorine donor;the presence and amount of phenolic precursor compounds, and
the amount of inorganic ions present to act as a catalyst in the chemical synthesis of
dioxin. Of these factors, only temperature can be controlled to reduce variability. The
other rate limiting factors are random variables, because the physical/chemical composition
of MSW burned in the incinerator is constantly changing with time. MSW is heterogenous
in composition. If it is true that the magnitude of concentration of the stack emission of
dioxin is random and highly variable, then segregating emissions on the basis of "good"
incinerators versus "bad" incinerators (for the purpose of deriving a representative
emission factor) solely on the basis of the magnitude of concentration of dioxin may be a
weak argument, and that a more inclusive data set may be more representative of the
10-103 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10.16. Basic Statistical Analyses of Emissions of Dioxin Used in the Hypothetical
MSW Incinerator Scenario.
Facility
1
2
3
4
5
6
7
8
9
10
Mean3
St.Devb
MAPDC
Facility
1
2
3
4
5
6
7
8
9
10
Mean9
St.Devb
MAPDC
TCDD
U/g/Mt)
2.10
7.85
9.40
0.39
1.94
15.6
1.97
0.28
2.91
0.82
4.33
5.03
TCDF
U/g/Mt)
NA
105
129
11.4
24.4
78.0
NA
18.0
23.0
4.06
49.1
47.9
Ab.
Dev1
U/g/Mt)
2.23
3.52
5.08
3.93
2.38
1.13
2.36
4.05
1.41
3.51
68.36%
Ab.
Dev1
U/g/Mt)
NA
55.9
79.9
37.7
24.7
28.9
NA
31.1
26.1
45.0
Deviat2
(%)
51.50
81.29
117.32
90.76
54.97
26.10
54.50
93.53
32.56
81.06
Deviat2
(%)
NA
113.85
162.73
76.78
50.31
58.86
NA
63.34
53.16
91.65
83.83%
OCDD Ab
Dev.1
U/g/Mt) (//g/Mt)
12.8 137
15.4 135
206 56
60.80 89.2
37.6 112
374 224
604 454
14.3 136
92.3 57.7
83.0 67.0
150
195
Key:
a arithmetic mean
b standard deviation
Deviat.
(%)
91.33
90.00
37.33
59.47
74.67
149.33
302.67
90.67
38.47
44.67
97.86
c mean absolute percentage
deviation
1 absolute deviation
mean
from the
2 percentage deviation from the
mean
10-104
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
wide range of potential variation in incinerator emissions. If a more inclusive set of
emissions from existing databases on mass burn, heat recovery incinerators (equipped with
electrostatic precipitators) are included in the universe of emission factors, the range of
potential average emissions of 2,3,7,8-TCDD would span roughly four orders of magnitude
in relative concentration, e.g., 0.2/jg/Mt to 2000//g/Mt (U.S.EPA, 1987). No such
variation was noted in the data set in this analysis. Emissions from incinerators equipped
with dry-scrubbers combined with fabric filters are from 1 % to 10 % of emissions from
incinerators equipped with electrostatic precipitators (U.S. EPA, 1989; U.S. EPA,1987).
This provides additional information to support a hypothesis that the formation and
subsequent emission of chlorinated dioxins and dibenzofurans from the incineration of
MSW is absolutely certain, but the magnitude of concentration of emission may be random
and highly variable over some discrete time interval. The hypothesis warrants further
testing through additional combustion engineering research. It may be that emission
factors that show small variability, high correlations with plant operations (such as
temperature of the flame in the combustion zone, and % carbon monoxide monitored in
stack emissions), were derived as a result of committing a type I error, e.g, the researcher
wishes to select critical measurements so that the variance is small. Critical data may be
excluded by some subjective criteria that, in itself, becomes difficult to measure the overall
effect on the distribution of data. These criteria may include one or more of the following:
(1) eliminating all facilities without heat recovery; (2) eliminating facilities that experience
anomalies in operation during test periods; (3) eliminating facilities on the basis of some
arbitrary capacity cutoff; (4) eliminating facilities on the subjective basis that the emission
appears to be too great in magnitude.
10.3.3. The Representativeness of the Hypothetical Incinerator In Comparison to Actual
Facilities
A third element to investigating uncertainty in the analysis of the emission of
dioxin-like compounds is the "representativeness" of the hypothetical incinerator. This can
be addressed by examining the distribution of design type of existing and projected
facilities; by examining the distribution of plant capacity, e.g., metric tons incinerated over
24 hours; by comparisons to the distribution of stack height relative to plant capacity.
In 1989, the year when figures were last available, there were approximately 200
10-105 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
municipal solid waste incineration facilities operating in the U.S. (U.S.EPA,1989). Figures
10-2 and 10-3 are an approximate distribution (broken down by percent mass of total
MSW combusted) of existing incineration technologies operating in the U.S, and
technologies projected to be operational in the year 1995. From these distributions it is
apparent that the mass burn technology currently dominates, and will continue to
dominate, the type of technology incinerating most of the MSW in the U.S. Therefore the
selection of the mass burn technology for the hypothetical incinerator is representative of
actual and projected technologies.
A second issue relevant to the question of how representative the hypothetical
incinerator is of existing technology is the combustion capacity. The hypothetical
incinerator has a combustion capacity of 2727 Mt per day. Figure 10-4 is a histogram of
the distribution of mass burn MSW technologies by combustion capacity. The capacity of
2727 metric tons per day is within the upper 75th percentile in the distribution of mass
burn MSW incinerators that exist and are planned out to the year 1995. This means that 3
out of every 4 of the mass burn incinerators are smaller in capacity than the capacity
chosen for the hypothetical case.
A third area involving representativeness is the height of the stack assumed for the
hypothetical case. The stack height that is assumed is 46 meters. Figure 10-5 relates
facility capacity to stack height of 11 7 existing MSW incinerators operational in the U.S.
Although there is a poor correlation between capacity and stack height, e.g., r2 = .37, the
distribution gives some reference to the selection made. The scatter plot of 117 existing
MSW incinerators in the U.S. shows that for the capacity of 2727 Mt/day, the stack
height of 46 meters is fairly representative of actual facilities, and is a reasonable
assumption.
The final criteria addressing "representativeness" of the hypothetical MSW
incinerator is the type of pollution control devices assumed for control of dioxin emissions.
The hypothetical facility has two pollution control scenarios: electrostatic precipitators
(ESPs), and dry scrubber combined with a fabric filter (DSFF). The distribution of air
pollution control technologies of the existing MSW incinerators is as follows: 43.67% are
controlled by ESPs; 20.25% are controlled by DSFF; 5.7% are controlled by dry scrubber
combined with ESPs; and 13.29% have no form of pollution control (EPA, 1992a).
Therefore, about 64% of all existing MSW incinerators are controlled by either ESPs or
10-106 7/31/92
-------
DBAFT-DO NOT QUOTE OR CITE
12%
6%
24%
58%
Mass Burn
FCF
Modular
Fluidized-bed
Figure 10-2. Percent
distribution of existing incinerators.
3.8%
6.4%
24.7%
,1%
Mass burn
B Fluidized-bed
£3 Modular
Fl9u,e10-3.
o, ~
-------
DRAFT-DO NOT QUOTE OR CITE
80.0%
60.0% -
«
>
~ 40.0% -
3
E
3
0 20.0% -
0.0%
Cumulative %
Adapted from U.S. EPA, 1989
182-455 455-909 909-1818 1818-2727
Capacity Metric tons/day
Figure 10-4. Capacity distribution of existing and planned mass burn incinerators.
120
100
60
0
y-23.785+ 2.3559e-2x RA2-OJ374
Q
hypothetical incinerator
stack H
1000 2000 3000
capcity(Vd)
4000
Figure 10.5. Scatter plot of stack height relative to capacity for 117 MSW incinerators
operation in the United States (EPA, 1987).
10-108
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
DSFF. Based on this distribution, the control scenarios assumed for the hypothetical case
seem reasonable.
10.3.4. Air Dispersion Analysis of Incinerator Emissions
Air dispersion analysis was performed using the Industrial Source Complex Model.
The model is intended to give approximate estimations of atmospheric dispersion and wet
and dry deposition flux, and does not give absolute values. The model is used in the
context of predicting future states based on known characteristics of geographical
location, local meteorological conditions, temporal rates of emissions, and the physical
description of the facility.
The ISC is a Gaussian plume model. Downwind concentrations of the dioxin-like
chemicals are calculated as a function of stack height, the mass emission rate, the wind
speed, and general atmospheric conditions. The Gaussian model assumes that the
emission concentrations predicted by the model will fit a normal distribution. The principal
assumptions in the Gaussian model are (Kapahi, 1991):
• The air concentration of the chemical at a fixed distance from the source is
directly proportional to the emission rate from the source;
• The air concentration of a given chemical is inversely proportional to the wind
speed corresponding to the effective height of release of the chemical into the air;
• The predicted ground-level concentration of the chemical approaches zero at
large distances from the initial point of release.
• The model is steady-state.
• The model assumes constant wind speed, wind direction, and atmospheric
stability over time and space for a given time period.
In general the stochastic features of the model have been shown to predict annual
average ambient air concentrations of a chemical emission from an industrial source to
within a factor of one-order of magnitude of measured values, and in some cases, within a
factor of 3 to 4 -fold of field measurements (Cohrssen ,1989). This modeling error spans
both sides of the predicted concentration, that is the actual concentration may be plus or
minus this amount of the predicted value. The most sensitive aspects to variability in
modeled predictions of ambient air impacts, if emissions are held constant, are stack
height (height of the release), and terrain (flat verses complex topography). To investigate
10-109 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
modeling variability, the EPA placed a prototype hypothetical hazardous waste incinerator
in flat terrain and elevated terrain in geographical areas around the U.S. (U.S. EPA,
1991 a). Then the stack height was varied at these particular locations. Numerous runs of
the ISC model were made at twelve specific sites to compare and contrast the influence of
stack height and terrain on predicted ambient air concentrations of various mass emission
rates of specific inorganic pollutants. A series of tables were developed from this
sensitivity analysis from which the numerical estimation of the variability as a function of
stack height and terrain can be inferred. When the hypothetical hazardous waste
incinerator was modeled in flat terrain, e.g., topography within a distance of 5 km is not
above the height of the stack, and the stack height was varied from 4 meters to 120
meters, the variability in the predicted ambient air concentration spanned two orders of
magnitude (100). The lower stack height resulted in a predicted ambient air concentration
that was 100 times greater than the concentration predicted using the tallest stack height.
When the hypothetical hazardous waste incinerator was located in complex terrain over
the same range of physical stack heights, the variability in estimated groundlevel
concentration of the subject pollutant spanned two orders of magnitude (100-fold). In the
latter case the stack height was computed as the terrain-adjusted stack height by
subtracting from the physical stack height the influence of terrain on plume rise. From the
limited sensitivity analysis of hazardous waste incinerators it can be assumed that the
predictions of spacial ground-level ambient air concentrations of dioxin-like chemicals could
differ from values in Tables 6-16 and 6-17 by two-orders of magnitude in consideration of
changes in stack height or changes in terrain. For example. Table 6-1 6 shows that the
maximum annual average ambient air concentration of 2,3,7,8-TCDD predicted near the
hypothetical MSW incinerator is 1.65 femtograms/cubic meter (fg/m3) for the stack height
of 46 meters, and assuming flat terrain. If only the stack height is varied from 20 meters
to 120 meters, and all other modelling parameters are held constant, then the predicted
ambient air concentration would be approximately 10 times greater and 10 times less than
the estimated concentration, respectively. The uncertainty is broader when considering
the influence of topography on predictability of the ground-level concentrations from the
model. If only terrain elevation is varied at a distance of 5 km from the hypothetical MSW
incinerator from zero elevation to 46 meters, e.g., the height of the stack, then the
predicted ambient air concentration of 2,3,7,8-TCDD would be approximately ten times
10-110 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
greater. The tables derived in the hazardous waste incineration analysis have a limitation of
elevation of terrain to the height of the stack.
The most uncertain aspect to the modeling is the estimation of dry and wet
deposition flux of dioxin-like compounds on the vicinity of the hypothetical MSW
incinerator. Contributing most to this uncertainty seems to be the settling velocities and
scavenging coefficients estimated for specific particle size diameters (Cohrssen, 1989;
Doran, 1985). Seinfeld (1986) found that particles over 20 microns in diameter settle
primarily by gravity, whereas smaller particles deposit primarily by atmospheric turbulence
and molecular diffusion. Considerable, but non-quantifiable, uncertainty exists with
respect to deposition velocities of particles 0.1 to 1.0 microns in diameter (Seinfeld, 1986).
The uncertainty is difficult to define. The wide variation of predicted deposition velocities
as a function of particle size, atmospheric turbulence and terrain adds to this uncertainty
(Sehmel, 1980). However, Gaussian plume dispersion models have been field validated for
their ability to spatially predict dry deposition flux over some specified distance (Doran,
1985). In a series of field experiments conducted by Pacific Northwest Laboratory (Doran,
1985) zinc sulfide was used as a depositing tracer gas, and sulfur hexafluoride was used
as a non-depositing tracer gas to compare and contrast modeling results with field
measurements of dry deposition and atmospheric diffusion of the gases. The tracer was
released from a height of 2 meters, and all releases were made under relatively stable
atmospheric conditions. Five sampling stations were located downwind of the release from
100 to 3200 meters. The results of these experiments showed good agreement with the
predicted verses the measured deposition of the tracer ZnS. The overall correlation
coefficient between predicted and measured deposition concentration was found to be
0.82 (Doran, 1985), but the models marginally over-predicted deposition flux near the
source of release, and under-predicted deposition flux at 3200 meters.
Travis and Yambert (1991) have evaluated the uncertainty in modeling the dry
deposition flux of particulates using four standard Gaussian plume dispersion models.
Since deposition flux is dependent on deposition velocity for a given particle mass and
diameter, comparisons were made between model-generated deposition velocities and
measured values found in the open literature for particles ranging from 0.01 to 30 microns
in diameter. It was found that measured deposition velocities for a given particle size in
the scientific literature exhibit variability spanning roughly two orders of magnitude. The
10-111 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
analysis of the mean predicted deposition velocities to mean measured values showed that
most measured data exceeded the predicted data for all four models. Moreover, the
models underestimated the mean deposition velocities for particles in the range of
diameters from 0.05 to 1.0 microns.
Similar uncertainty probably exists with regard to scavenging of various diameter
particles by various intensity of rainfall. Seinfeld (1986) has calculated scavenging
coefficients in terms of the removal efficiency of particles of a given size by rain droplets
having a given momentum. Seinfeld (1986) found that the scavenging coefficient of a
given particle diameter corresponding to a given rainfall intensity can be calculated based
on physical laws, but there is a complete absence of research data to verify these
calculations. Hence it is not possible to address the accuracy nor uncertainty of the wet
deposition flux estimated in Tables 6-18 and 6-19.
10.3.5 Comparison of predicted ambient concentrations to ambient measurements
In the analysis of the environmental impact associated with the emissions from the
hypothetical MSW incinerator, predictions have been made with respect to the spacial and
temporal distribution of ambient air concentrations of specific dioxin-like compounds
(Tables 6-16, 6-17). Comparing these predictions to ambient air measurements taken
downwind of an operating MSW incinerator provides a qualitative check of the relative
correctness of the estimated concentrations. There are certain caveats to such a
comparison:
• Measured concentrations are in different geographical settings from the
hypothetical case, and therefore reflect different meteorological conditions and
topography.
• Measured concentrations most likely reflect stack emissions of dioxin-like
compounds from the actual incinerator that significantly differ from the hypothetical case.
No stack measurements were provided by the investigator in the literature cited.
• Measured values are a "snapshot" in time. Therefore comparisons are between
annual average predicted concentrations and less than annual average monitored
concentrations.
• Measured values may reflect differences in analytical sensitivities; ambient
monitoring instrumentation; and statistical reliability of placement of the sampling network
10-112 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
with respect to measuring the true ambient air impact.
Given these limitations in the database, average ambient air concentrations of the
dioxin-like congeners measured in the vicinity of four operational MSW incinerators were
summarized and compared with the predicted annual average ambient air impact from the
hypothetical incinerator. Table 10-17 is a summary of this data. The range of
measurements specific to each congener is reported in the Table. The qualitative "reality"
check of the adequacy of the modeling of air emissions from the hypothetical case
involves a comparison of the range of measured data to see if the predicted values fall
within the range. As can be determined from the analysis, most of the predicted values do
fall within the range of measured values.
10-113 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table 10-17. Comparison between measured ambient air concentrations with predicted
ambient air impacts of the hypothetical MSW incinerator (values as pg/m3)
Congeners 1
2,3,7,8-TCDD 0.004
1,2,3,7,8-PeCDD 0.011
1,2,3,4,7,8-HxCDD 0.025
1,2,3,6,7,8-HxCDD 0.023
1,2,3,7,8,9-HxCDD 0.011
1,2,3,4,6,7,8-HpCDD 0.040
OctaCDD 0.330
2,3,7,8-TCDF 0.006
1,2,3,7,8-PeCDF 0.010
2,3,4,7,8-PeCDF 0.010
1,2,3,4,7,8-HxCDF 0.030
1, 2,3,6,7, 8-HxCDF 0.030
1,2,3,7,8,9-HxCDF 0.020
2,3,4,6,7,8-HxCDF 0.030
1,2,3,4,6,7,8-HpCDF 0.060
1,2,3,4,7,8,9-HpCDF 0.040
OctaCDF 0.010
Reference
2 3
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
.238
.010
.005
.013
.015
.150
.590
.031
.019
.029
.050
.022
.005
.010
.135
.010
.059
0.020
0.220
0.019
0.710
0.360
NA
7.400
0.380
0.420
0.430
0.270
0.240
0.020
0.210
NA
NA
0.780
#
0
0
0
0
0
2
2
1
0
0
0
0
0
0
2
0
4
.015
.015
.243
.390
.063
.000
.950
.470
.247
.687
.110
.453
.043
.763
.080
NA
.748
Range
0.004
0.011
0.005
0.013
0.011
0.041
0.330
0.006
0.010
0.010
0.030
0.030
0.005
0.010
0.060
0.010
0.010
to 0
to 0
to 0
to 0
to 0
to 2
to 7
to 1
to 0
to 0
to 0
to 0
to 0
to 0
to 2
to 0
to 0
.238
.220
.190
.710
.063
.000
.400
.470
.420
.687
.270
.453
.043
.763
.080
.040
.780
Modeled
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0.
0
.002
.015
.004
.010
.004
.008
.062
.020
.020
.020
.012
.010
.010
.008
.010
0003
.030
Within
Range
Yes
Yes*
Yes*
Yes*
No
No
No
No
Yes
Yes
Yes*
Yes*
Yes
Yes*
No
No
Yes
References: 1. Conn. DEP,1988; 2. Conn. DEP,1989; 3-Rappe, 1987; 4. Smith, 1989.
Notes: a. All values are average values, b. Non detects were treated as measured values.
NA = Not Available.
* Predicted value judged to be close to the lower bound of the range.
10-114
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
REFERENCES FOR CHAPTER 10
Ambrose, R.B., T.A. Wood, J.P. Connolly, and R.W. Schanz. 1988. WASP4, A
Hydrodynamic and Water Quality Model -- Model Theory, User's Manual, and
Programmer's Guide. Athens Environmental Research Laboratory, Office of
Research and Development, US EPA. EPA/600/3-87/039.
Bacci, E., D. Calamari, C. Gaggi, M. Vighi. 1990. Bioconcentration of organic chemical
vapors in plant leaves: experimental measurements and correlation. Environmental
Science and Technology 24: 885-889.
Bacci, E., M.J. Cerejeira, C. Gaggi, G. Chemello, D. Calamari, and M. Vighi. 1992.
Chlorinated dioxins: volatilization from soils and bioconcentration in plant leaves. In
Press in Bulletin of Environmental Contamination and Toxicology.
Baes, C.F., R.D. Sharp, A.L. Sjoreen, and R.W. Shor. 1984. A review and analysis
of parameters for assessing transport of environmentally released radionuclides
through agriculture. Prepared for the U.S. Department of Energy under contract no.
DE-AC05-840R21400.
Beck, H., K. Eckart, W. Mather, and R. Wittkowski. 1989. PCDD and PCDF body burden
from food intake in the Federal Republic of Germany. Chemosphere 18: 417-424.
Bopp, R.F., M.L. Gross, H. Tong, M.J. Simpson, S.J. Monson, B.L. Deck, and F.C. Moser.
1991. A major incident of dioxin contamination: sediments of New Jersey
estuaries. Environmental Science and Technology 25: 951-956.
Brady, N.C. 1984. The Nature and Properties of Soils. Ninth Edition. Macmillan
Publishing Company. New York, NY 10022.
Briggs, G.G., R.H. Bromilow, and A.A. Evans. 1982. Relationships between lipophilicity
and root uptake and translocation of non-ionised chemicals by barley. Pesticide
Science 13:495-504.
Cleverly, D.H.; Morrison, R.M.; Riddle, B.L.;Kellam, R.G. (1991). Regulatory analysis of
pollutant emissions, including polychlorinated dibenzo-p-dioxins (CDDs) and
dibenzofurans (CDFs), from municipal waste combustors. In: Hattermer-Frey, H.A.;
Travis,C., eds., Health effects of municipal waste incineration. Boca Raton,FL, CRC
Press, pp.47-65.
Coccusi, S., F. DiGerolamo, A. Verderio, A. Cavallaro, G. Colli, A. Gorni, G. Invernizzi, and
L. Luciani. 1979. Absorption and translocation of tetra-chlorodibenzo-p-dioxin by
plants from polluted soil. Experientia 35(4): 482-484.
Cohrssen, J., Covello, V. 1989. Risk analysis: A guide to principles and methods for
analyzing health and environmental risks. U.S. Council on Environmental Quality,
Office of the President. National Technical Information Service #PB 89-137772.
10-115 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Conn DEP. 1988. Ambient background monitoring for selected PCDDs/PCDFs in the
vicinity of the Mid-Connecticut Resource Recovery Facility. Connecticut
Department of Environmental Protection, Air Compliance Unit, Hartford, CT. Project
report by Combustion Engineering,Inc,Windsor,CT., Document # 1790-002/6350-
001.Final Report. September,1988.
Conn DEP. 1989. Ambient background monitoring for selected PCDDs/PCDFs in the
vicinity of the Bristol Resource Recovery Facility.Connecticut Department of
Environmental Protection,Air Compliance Unit, Hartford, CT. Project report by
Ogden Martin Systems,Fairfield,NJ. Document # 5160-003-400.Final Report.
February 1989.
Connett, P. and T. Webster. 1987. An estimation of the relative human exposure to
2,3,7,8-TCDD emissions via inhalation and ingestion of cow's milk. Chemosphere
15: 2079-2084.
Connolly, J.P. and C.J. Pedersen. 1987. A thermodynaic-based evaluation of organic
chemical accumulation in aquatic organisms. Environmental Science and
Technology 22:99-103.
Cook, P.M., M.K. Walker, D. W. Kuehl, and R.E. Peterson. 1991. p. 143-168 in Gallo,
M.A., R.J. Scheuplein, and K.A. Van Der Heijden, eds., Banbury Report 35:
Biological Basis for Risk Assessment of Dioxins and Related Compounds. Cold
Spring Harbor Laboratory Press, Plainview, New York 11803.
Doran, J.C., Horst, T.W. (1985). An evaluation of gaussian plume-depletion models with
dual-tracer field measurements. Atmos. Environ. 19:939-951.
Foster, G.R., and T.E. Hakonson. 1987. Erosion Losses of Fallout Plutonium. In Howard,
W.A., and R. Fuller, eds. Dynamics of Transuranics and Other Radionuclides in
Natural Environments. NTIS-87014456. Springfield, VA.
Fries, G.F. 1985. Bioavailability of soil-borne polybrominated biphenyls ingested by farm
animals. Journal of Toxicology and Environmental Health 16: 565-579.
Fries, G.F., and L.W. Jacobs. 1986. Evaluation of Residual Polybrominated Biphenyl
Contamination Present on Michigan Farms in 1978. Research Report 477 from the
Michigan State University Agricultural Experiment Station, East Lansing, Michigan.
December, 1986.
Fries, G.F. and D.J. Paustenbach. 1990. Evaluation of potential transmission of 2,3,7,8-
tetrachlorodibenzo-p-dioxin-contaminated incinerator emissions to humans via
foods. Journal of Toxicology and Environmental Health 29: 1-43.
Furst, P., C. Furst, and W. Groebel. 1990. Levels of PCDs and PCDFs in food-stuffs from
the Federal Republic of Germany. Chemosphere 20: 787-792.
10-116 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Gillette, D.A. 1981. Production of dust that may be carried greate distances. In: Desert
Dust: Origin, Characterists. and Effect on Man, edited by Troy Pewe. Geological
Society of America Special Paper 186, pp. 11-26.
Gotz, R., and E. Schumacher. 1990. Polychloriete dibenzo-p-dioxine (PCDDs) und
plychlorierte dibnezofurane (PCDFs) in sedimenten und fischen Aus Dem Hamburger
Hafen. Chemosphere 20: 51-73.
Hawley, J.K. 1985. Assessment of health risk from exposure to contaminated soil. Risk
Analysis 5(4): 289-302.
Huckins, J.N. T.R. Schwartz, J.D. Petty, L.M. Smith. 1988. Determination, fate, and
potential significance of PCBs in fish and sediment samples with emphasis on
selected AHH-lnducing congeners. Chemosphere 17: 1995-2016.
Hulster, A. 1991. Transfer of PCDD/PCDF from Contaminated Soils into Two Vegetable
Crops. Unpublished data presented in a poster at Dioxin 91.
Hwang, ST., J.W. Falco, and C.H. Nauman. 1986. Development of Advisory Levels for
Polychlorinated Biphenyls (PCBs) Cleanup. Exposure Assessment Group, Office of
Research and Development, U.S. Environmental Protection Agency. EPA/600/6-
86/002.
Isensee, A.R.; Jones, G.E. 1971. Absorption and translocation of root and foliage applied
2,4,-dichlorophenol, 2,7-dichlorodibenzo-p-dioxin and 2,3,7,8-tetrachlorodibenzo-p-
dioxin. J. Agric. Food Chem. 19(6): 1210-1214.
Jackson, D.R., M.H. Roulier, H.M. Grotta, S.W. Rust, and J.S. Warner. 1986. Solubility
of 2,3,7,8-TCDD in Contaminated Soils, pp. 185-200 in Rappe, C., G. Choudhary,
and L.H. Keith, eds., Chlorinated Dioxins and Dibenzofurans in Perspective. Lewis
Publishers, Inc.
Javitz, H. 1980. Seafood Consumption Data Analysis. SRI International. Final report
prepared for EPA Office of Water Regulations and Standards. EPA Contract 68-01-
3887.
Jensen, D.J. and R.A. Hummel. 1982. Secretion of TCDD in milk and cream following
the feeding of TCDD to lactating cows. Bulletin of Environmental Contamination
and Toxicology 29: 440-446.
Jensen, D.J., R.A. Hummel, N.H. Mahle, C.W. Kocker, and H.S. Higgins. 1981. A
Residue study on beef cattle consuming 2,3,7,8-tetrachlorodibenzo-p-dioxin.
Journal of Agriculture and Food Chemistry 29: 265-268.
Kapahi, R. 1991. Modeling the dispersion of toxic air pollutants emitted from municipal
• solid waste incinerators.In: Hattermer-Frey, H.A.; Travis,C., eds.. Health effects of
municipal waste incineration. Boca Raton,FL., CRC Press, pp.67-82.
10-117 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Karickhoff, S.W., D.S. Brown, and T.A. Scott. 1979. Sorption of hydrophobia pollutants
on natural sediments. Water Research 13: 241-248.
Kenaga, E.E. and C.A.I. Goring. 1980. Relationship between water solubility, soil-
sorption, octanol-water partitioning, and bioconcentration of chemicals in biota. In
Eaton, J.SC., Parrish, P.R., and Hendrics, A.C. (eds). Aquatic Toxicology. ASTM
STP 707. Philadelphia, PA: American Society for Testing and Materials.
Kjeller, L.O., S.E. Kulp, S. Bergek, M. Bostrom, P.A. Bergquist, C. Rappe, B. Jonsoson,
D.De Wit, B. Jansson, and M. Olsson. 1990. Levels and possible sources of
PCDD/PCDF in sediment and pike samples from Swedish lakes and rivers (Part
One). Chemosphere 20: 1489-1496.
Kuehl, D.W., B.C. Butterworth, A. McBride, S. Kroner, and D. Bahnick. 1989.
Contamination of fish by 2,3,7,8-tetrachloridibenzo-p-dioxin: a survey of fish from
major watersheds in the United States. Chemosphere 18: 1997-2017.
Linsley, R.K., M.A. Kohler, and J.L.H. Paulus. 1982. Hydrology for Engineers. McGraw-
Hill, Inc., New York, NY.
Lower, W.R., A.F. Yanders, C.E. Orazio, R.K. Puri, J. Hancock, and S. Kapila. 1989. A
survey of 2,3,7,8-tetrachlorodibenzo-p-dioxin residues in selected animal species
from times Beach, Missouri. Chemosphere 18: 1079-1088.
Lyman, W.J., W.F. Reehl, and D.H. Rosenblatt. 1982. Handbook of Chemical Property
Estimation Methods. Mcgraw-Hill, Inc., New York, NY.
McCrady, J.K, and S.P. Maggard. 1992. The uptake and photodegradation of 2,3,7,8-
TCDD sorbed to grass foliage. Draft, to be submitted for publication.
McCrady, J.K., C. McFarlane, and L.K. Gander. 1990. The transport and fate of 2,3,7,8-
TCDD in soybean and corn. Chemosphere 21:359-376.
McKone, T.E. and P.B. Ryan. 1989. Human exposures to chemicals through food chains:
an uncertainty analysis. Environmental Science and Technology 23: 11 54-11 63.
Meyer, C., D. O'Keefe, D. Hilker, L. Rafferty, L. Wilson, S. Conner, K. Aldous, K.
Markussen, and K. Slade. 1989. A survey of twenty community water systems in
New York State for PCDDs and PCDFs. Chemosphere 19:21-26.
Michaels, R. 1989. Health Risk Assessment for the Planned St. Lawrence County, New
York, Resource Recovery Facility. RAMTRAC, Long Island City, New York.
MRI, 1990. Special Management Standards for Municipal Waste Combustion (MWC) Ash.
Submitted by: Midwest Research Institute, 425 Volker Boulevard, Kansas City, MO
64110-2299, to US EPA, Municipal sSolid Waste Program. EPA Contract No. 68-
01-7287. June 29, 1990.
10-118 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Nestrick, T.J., L.L. Lamparski, N.N. Frawley, R.A. Hummel, C.W. Kocher, N.H. Mahle,
J.W. McCoy, D.L. Miller, T.L. Peters, J.L. Pillepich, W.E. Smith, and S.W. Tobey.
1986. Perspectives of a large scale environmental survey for chlorinated dioxins:
overview and soil data. Chemosphere 15: 1453-1460.
Niimi, A.J., and B.C. Oliver. 1989. Distribution of polychlorinated biphenyl congeners and
other halocarbons in whole fish and muscle from Lake Ontario salmonids.
Environmental Science and Technology 23: 83-88.
Norwood, C.B, M. Hackett, R.J. Powell, B.C. Butterworth, K.J. Williamson, and S. M.
Naumann. 1989. Polychlorinated dibenzo-p-dioxins and dibenzofurans in selected
estuarine sediments. Chemosphere 18:553-560.
Oliver, E.G., and A.J. Niimi. 1988. Trophodynamic analysis of polychlorinated biphenyl
congeners and other chlorinated hydrocarbons in the Lake Ontario ecosystem.
Environmental Science and Technology 22: 388-397.
OTA. (1989). Facing America's trash. What next for municipal solid waste? Office of
Technology Assessment,U.S.Congress, Washington, DC.
Pao, E.M., K.H. Fleming, P.M. Guenther, et al. 1982. Foods commonly eaten by
individuals: amount per day and per eating occassions. U.S. Department of
Agriculture, home Economics Report No. 44.
Parker, C.E., W.A. Jones, H.B. Matthews, E.E. McConnell, and J. Mass. 1980. The
chronic toxicity of technical and analytical pentachlorophenol in cattle. II.
Chemical analysis of tissues. Toxic. Appl. Pharmacol. 55: 359-369.
Parkerton, T.F. 1991. Development of a Generic Bioenergetics-Based Model for Predicting
the Bioaccumulation of Persistent Sediment-Associated Contaminants - Final
Report, New Jersey Department of Environmental Protection, Trenton, NJ.
Poiger, H. and C. Schlatter. 1980. Influence of solvents and adsorbents on dermal and
intestinal adsorption of TCDD. Food Cosmet. Toxicol 18: 477-481.
Rappe, C. 1991. Introduction, profiles, and patterns. In: Procedings of the 11th
international symposium on chlorinated dioxins and related compounds. Research
Triangle Park, NC, September 23-27, 1991, pg 17.
Rappe, C., P.A. Bergqvist, and L.A. Kjeller. 1989. Levels, trends and patterns of PCDDs
and PCDFs in Schandinavian environmental samples. Chemosphere 18: 651-658.
Rappe, C., and L.O. Kjeller. 1987. PCDDs and PCDFs in environmental samples, air,
particulates, sediments, and soil. Chemosphere 16: 1775-1780.
10-119 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Rappe, C., M. Nygren, G. Lindstrom, H.R. Buser, 0. Blaser, and C. Wuthrich. 1987.
Polychlorinated Dibenzofurands and Dibenzo-p-dioxins and other chlorinated
contaminants in cow milk from various locations in Switzerland. Environmental
Science and Technology. 21: 964-970.
Reed, L.W., G.T. Hunt, B.E. Maisel, M. Hoyt, D. Keefe, and P. Hackney. 1990. Baseline
assessment of PCDDs/PCDFs in the vicinity of the Elk River, Minnesota generating
station. Chemosphere 21: 159-171.
Sacchi, G.A., P. Vigano, G. Fortunati, S.M. Coccusi. 1986. Accumulation of 2,3,7,8-
TCDD from soil and nutrient solution by bean and maize plants. Experientia 42:
586-588.
SAIC. 1986. Science Applications International Corp. Conversion factors for use in
estimating environmental transport of contaminated soil from unregulated disposal
sites. Prepared for the U.S. Environmental Protection Agency, Office of Solid
Waste, Washington, D.C., under EPA contract 68-01-7624.
Sehmel, G.A. 1980. Paritcle and gas dry deposition. A review. Atmos. Environ., vol.
14: 983-1012.
Seinfeld, J.H. 1986. Atmospheric chemistry and physics of air pollution. New York,NY.,
John Wiley and Sons.
Siebert, P.C.; Alston-Gulden, D.; Jones, K.H. (1991). An analysis of worldwide resource
recovery emissions data and the implications for risk assessment.In: Hattermer-
Frey, H.A.; Travis,C., eds.. Health effects of municipal waste incineration. Boca
Raton,FL., CRC Press, pp.2-46.
Smith, L.M., T.R. Schwartz, and K. Feltz. 1990. Determination and occurrence of AHH-
Active polychlorinated biphenyls, 2,3,7,8-tetrachloro-p-dioxin and 2,3,7,8-
tetrachlodibenzofuran in Lake Michigan sediment and biota. The question of their
relative toxicological significance. Chemosphere 21: 1063-1085.
Startin, J.R., M. Rose, C. Wright, I. Parker, and J. Gilbert. 1990. Surveillance of British
foods for PCDDs and PCDFs. Chemosphere 20: 793-798.
Stevens, J.B. and E.N. Gerbec. Dioxin in the agricultural food chain. Risk analysis 8(3):
329-335.
Swackhammer, D.L., B.D. McVeety, and R.A. Hites. 1988. Deposition and evaporation of
polychlorinated congeners to and from Siskiwit Lake, Isle Royale, Lake Superior.
Environmental Science and Technology 22: 664-672.
Thomann, R.V. 1989. Bioaccumulation model of organic chemical distribution in aquatic
food chains. Environmental Science and Technology 23: 699-707.
10-120 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Thomann, R.V., J.P. Connolly, and T.F. Parkerton. 1992. An equilibrium model of
organic chemical accumulation in aquatic foodwebs with sediment interaction. In
Press, to appear in Environmental Toxicology and Chemistry.
Travis, C.C. and A.D. Arms. 1988. Bioconcentration of Organics in Beef, Milk, and
Vegetation. Environmental Science and Technolology, Vol 22(3): 271-274.
Travis, C.C. and H.A. Hattemer-Frey. 1987. Human exposure to 2,3,7,8-TCDD.
Chemosphere 16: 2331-2342.
Travis, C.C. and H.A. Hattemer-Frey. 1991. Human exposure to dioxin. The Science of
the Total Environment 104: 97-127.
Travis, C., Yambert, M. (1991). Accounting for dry depostion in incinerator risk
assessments. ln:C.C. Travis, editor. Municipal waste incineration risk assessment.
Plenum Press, New York, NY., pp 53-63.
Turner, D.B. 1970. Workbook of Atmospheric Dispersion Estimates. PHS publication no.
999-AP-26 (NTIS PB 191482), Research Triangle Park, EPA, North Carolina.
USDA. 1983. Food Consumption: Households in the United States, Seasons and Year
1977-1978. U.S. Department of Agriculture.
USDA. 1966. Household food consumption survey 1965-1966. Report 12. Food
Consumption of Households in the U.S., Seasons and Years 1965-1966. United
States Department of Agriculture, Washington, D.C. U.S. Government Printing
Office.
U.S. EPA. 1985. Rapid Assessment of Exposure to Paniculate Emissions from Surface
Contamination Sites. Office of Health and Environmental Assessment, Office of
Research and Development, US EPA. EPA/600/8-85/002.
U.S. EPA. 1987. National Dioxin Study Report to Congress. Office of Solid Waste and
Emergency Response, U.S. Environmental Protection Agency, EPA 530-SW087-
025. August, 1987.
U.S. EPA. 1987a. Emission data base for municipal waste combustors. Office of Solid
Waste and Emergency Response, Office of Air and Radiation, Office of Research
and Development, Washington, DC., EPA/530-SW-87-021b, June, 1987.
U.S. EPA. 1987b. Characterization of the municipal waste combustion industry.Office of
Solid Waste and Emergency Response, Washington,DC., EPA-530-SW-87-021h.
June, 1987.
U.S. EPA. 1988. Compilation of Air Pollution Emission Factors. Volume 1. Stationary
Point and Area Sources. Fourth Edition. Supplement B. U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. Sepetember, 1988.
10-121 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
U.S. EPA. 1989. Exposure Factors Handbook. Exposure Assessment Group. Office of
Health and Environmental Assessment, Office of Research and Development. U.S.
Environmental Protection Agency. EPA/600/8-89/043. July, 1989.
U.S. EPA. 1989a. Background paper, municipal waste combustors air emission
standards. Office of Air Quality Planning and Standards, Research Triangle Park,
NC. April 26, 1989.
U.S. EPA. 1990a. Lake Ontario TCDD Bioaccumulation Study Final Report. Cooperative
study including US EPA, New York State Department of Environmental
Conservation, New York State Department of Health, and Occidental Chemical
Corporation. May, 1990.
U.S. EPA. 1990b. Integrated Risk Assessment for Dioxins and Furans from Chlorine
Bleaching in Pulp and Paper Mills. Office of Pollution Prevention and Toxics
(formerly Office of Toxic Substances), U.S. EPA. EPA 560/5-90-011.
U.S. EPA. 1990c. Methodology for Assessing Health Risks Associated with Indirect
Exposure to Combustor Emissions. Office of Health and Environmental
Assessment, Office of Research and Development, US Environmental Protection
Agency. EPA/600/6-90/003.
U.S. EPA. 1991. A Methodology for Estimating Population Exposures from the
Consumption of Chemically Contaminated Fish. Prepared by Tetra Tech, Inc.,
Fairfax VA. for Office of Policy, Planning, and Evaluation and Office of Research
and Development, US EPA. EPA/600/9-91/017.
U.S. EPA. 1991 a. Burning of hazardous waste in boilers and industiral furnaces, Final
Rule. Federal Register, volume 56, number 35. U.S. Government Printing Office,
Wash. DC., February 21, 1991, pp 7134- 7240.
U.S. EPA. 1992. Dermal Exposure Assessment: Principals and Applications. Exposure
Assessment Group, Office of Health and Environmental Assessment, Office of
Research and Development, US EPA. EPA/600/8-91/011A [DRAFT].
U.S. EPA. 1992a. Economic impact and preliminary regulatory impact analysis for
proposed MACT-based emission standards and guidelines for municipal waste
combustors. Office of Quality Planning and Standards, Research Triangle Park, NC.
EPA-450/3-91-029. January. 1992
U.S.EPA; Environment Canada. 1991. The environmental characterization of RDF
combustion technology. Mid-Connecticut facility, Hartford,CT. Test program and
results, volume II., June 1991.
Webster, T. and P. Connett. 1990. The use of bioconcentration factors in estimating the
2,3,7,8-TCDD content of cow's milk. Chemosphere 20: 779-786.
10-122 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Wipf, H.K., E. Homberger, N. Neuner, U.B. Ranalder, W. Vetter, and J.P. Vuilleumier.
1982. TCDD levels in soil and plant samples from the seveso area. In: Chlorinated
Dioxins and Related Compounds: Impact on the Environment. Eds. Hutzinger, 0.,
et al., Pergamon, NY.
Wischmeier, W.H., and D.D. Smith. 1965. Predicting Rainfall-Erosion Losses from
Cropland East of the Rocky Mounts. Agriculture Handbook 282. U.S. Department
of Agriculture, Agricultural Research Service.
Young, A.L. 1983. Long term studies on the persistence and Movement of TCDD in a
natural ecosystem. In: Human and Environmental Risks of Chloronated Dioxins and
Related Compounds, Eds. Tucker, R.E., A.L. Young, A.P. Gray, Plenum Press, New
York, NY.
10-123 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
11. CONCLUSIONS AND RECOMMENDATIONS
This chapter presents the conclusions regarding how people are exposed to the
dioxin-like compounds and recommendations for further research to resolve uncertainties in
these estimates.
11.1. CONCLUSIONS
The following are principal findings and conclusions of this assessment:
1. The dioxin-like compounds are commonly found in soils, sediments and biota
throughout the world. Concentrations in nonindustrial rural areas are generally lower than
in urban or industrial areas. The higher chlorinated hepta- and octa- dioxin and furan
congeners were generally more frequently found and at higher concentrations than the
tetra- through hexa- congeners in environmental and exposure media. Selected literature
references containing occurrence information on dioxin-like compounds were reviewed in
an attempt to estimate background concentrations of these compounds in exposure media.
The estimated background exposure (assuming all pathways are equally additive) to all
dioxin-like compounds was estimated by multiplying the average media levels by typical
contact rates. This analysis suggests a background exposure level in the range of 20-90
pg of TEq/day using data for world-wide sources. This estimate is highly qualified
considering factors such as: sparseness of world-wide data, judgements as to whether
data used represents "background" exposures, the assumption that all exposures
considered to derive this estimate occur simultaneously contrasting the fact that several
exposures, particularly ingestion of meat products other than beef, were not considered,
and so on. Data on tissue levels suggest that body burden levels among industrialized
nations are reasonably similar. Chapter 3, Section 3.8 contains more details on this
exercise.
2. Typical exposure levels were also estimated by applying pharmacokinetic models
to body burden data. Using this approach, exposure levels to 2,3,7,8-TCDD are estimated
to be about 20 to 40 pg/day. This is consistent with the analysis for exposure to total
11-1 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
dioxin-like compounds as described above. More details of this analysis are presented in
Chapter 8.
3. Procedures to estimate individual exposures to four categories of dioxin-like
contamination sources were described and demonstrated in this assessment. These
categories include: 1) on-site soils - the soil contamination and the exposure occur at the
same site, 2) off-site soil - the contaminated soil is located distant from the site of
exposure, 3) incinerator stack emissions - individuals near incinerator stacks are directly
exposed via inhalation of impacted air and indirectly exposed as a result of deposition of
contaminated particulates onto soil and vegetation, and 4) ash landfill - similar to the off-
site soil category except that source material is incinerator ash rather than soil. Exposure
assessments associated with these four source categories were demonstrated in Chapter
9. The example assessments involved development of scenarios which were carefully
crafted to be plausible and meaningful. However, such assessments are based on
numerous site-specific assumptions and therefore may not be generalizable to other sites.
For example, demonstration of the incinerator stack emission source category was based
on realistic emission rates, but only for a specific technology, air transport modeling was
accomplished using the Industrial Source Complex (ISC) model using climate data from a
specific location, and exposed individuals resided specific distances from the incinerator.
All source categories were demonstrated on three example compounds: a dioxin (2,3,7,8-
TCDD), a furan (2,3,4,7,8-PCDF), and a PCB (2,3,3',4,4',5,5'-HPCB). Finally, the example
scenarios were crafted to be consistent with current exposure assessment guidelines
which recommend assessment of "central" and "high-end" exposures. The following
observations are offered based on the demonstration of these source categories in Chapter
9 - again, the assertion is not being made that these observations are fully generalizable
conclusions:
• The highest human exposures estimated in these examples were associated
with ingestion of farm products: ingestion of beef and dairy products. Fish
ingestion was of principal concern when not overshadowed by ingestion of these
farm products. The apparent reasons for the prevalence of food chain exposures
include the tendency of dioxin-like compounds to bioaccumulate in food products of
11-2 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
high fat content, coupled with the ingestion rates of these food products. Lesser
but still noteworthy exposures occur through contact with soils including soil
ingestion and soil dermal contact. Other exposure pathways considered, including
inhalation (of dioxin-like compounds in a vapor phase and associated with dust
particles), fruit and vegetable ingestion, and water ingestion, were found to be of
lesser consequence in comparison to these other exposure pathways.
• Estimates of central exposures were generally less than an order of magnitude
lower than high-end exposure estimates. This trend was the result of exposure
parameter assignments and assumptions; the differences between a central and a
high end pattern of exposure was expressed primarily in terms of exposure
parameters - exposure durations, intake rates, etc. For example, exposure
estimates for the soil ingestion pathway for the central scenario demonstrated for
the on-site source category was one-fourth the exposure for soil ingestion for the
high end on-site source category demonstration scenario. The childhood pattern of
soil ingestion was modeled assuming 5 years of ingestion between the ages of 2
and 6 for both the central and high end scenarios. Children were assumed to ingest
0.8 g/day soil in the high end scenarios and 0.2 g/day in the central scenarios,
explaining the factor of four difference. One key difference between central and
high end example scenarios was that central exposures were assumed to occur
within a "residential" setting and high end within a "farm" setting. Those in the
farm setting were assumed to obtain a portion of their beef and milk intake from
home-produced and impacted cattle, whereas individuals in the residential setting
were assumed not to be exposed to impacted beef and milk. In that sense,
individuals in farm settings were evaluated as generally more exposed because of
these two exposure pathways. This strategy for defining central and high end
exposures is just one among many acceptable alternatives and should not be
interpreted as a general recommendation for future assessments of exposure to
dioxin-like (or other) compounds.
• The on-site source category was demonstrated assuming soil levels of 1 ng/kg
(ppt) which was characterized as typical of soil levels found by researchers
11-3 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
specifically targeting "background" or "rural" areas. The stack emission source
category was demonstrated using emission rates specific to a fabric filter combined
with semi-dry alkaline scrubbers (abbreviated DSFF in Chapter 6) emission control
technology. This was characterized in Chapter 6 as a high level of emission control
in use for a relatively small number of incinerators currently operating, but
expected to be more prevalent for new incinerators. Exposure media
concentrations and resulting exposures assuming these soil concentrations and
incinerator emission source strengths were found to be comparable. The off-site
soil source category was demonstrated assuming 1 x/g/kg (ppb) concentrations in a
contaminated area located 150 meters from a farm (the example scenario was
described as a high end scenario). This soil concentration was characterized as
typical of soils of known industrial contamination of the dioxin-like compounds.
The ash landfill soil concentrations were developed given estimated ash
concentrations of the example compounds based on the same emission control
technology. Resulting landfill concentrations were also in the low ppb range, and
exposed individuals resided 150 meters from the landfill. Exposures estimated for
individuals living near these areas of relatively high soil concentrations were 2 to 3
orders of magnitude higher as compared to the on-site and stack emission exposure
estimates. Like other scenario observations, these results are not offered as
generalizations. Key qualifiers include: they are specific to the source strengths
assumed and are specific to that source alone (i.e., total exposures might include
impacts from more than one source), proximity to the source strengths, realism of
estimated exposure media concentrations, and so on.
Further observations from the demonstration of source categories are given in Chapter 9,
Section 9.6.
4. In contrast to a few years ago, consensus is emerging as to causes and extent
of vegetative contamination by the dioxin-like compounds. It now appears that vegetation
is impacted as a result of three processes: sorption to outer portions of below ground
parts, "air-to-leaf" transfers of vapor-phase contaminants to outer above ground plant
surfaces, and deposition and sorption of air-borne particulate-phase contaminants onto
11-4 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
outer portions of above ground vegetation. Translocation from outer to inner portions
appears to be minimal, and translocation from root to shoot appears to be essentially
nonexistent. This document presents unique approaches for estimating vegetative impacts
from these routes. Also, estimates of fruit and vegetable ingestion exposures considered
below vs. above ground and protected vs. unprotected fruits. The models estimating
vegetative concentrations were examined in Chapter 10. Some observations from that
examination are now listed. It is noted that these observations are specific to the models
and the model parameters used for demonstration purposes. These observations are not
field verified and might change with a different set of parameters:
• When the source of contamination is soil, 90% and more of the estimated
above ground vegetative concentrations are explained by volatilization followed by
air-to-leaf transfers. In contrast, when incinerator emissions are the source, above
ground vegetative concentrations are dominated by particulate depositions.
• When the source of contamination is soil, below ground vegetation
concentrations exceed those of above ground vegetation by 1-2 orders of
magnitude. When the source of contamination is incinerator emissions, the reverse
is true: above ground vegetation concentrations exceed below ground
concentrations by roughly 2 orders of magnitude.
The analysis of the vegetative concentration models is given in Chapter 10, Section
10.2.11.
5. With some modifications, procedures in this assessment to estimate beef and
milk concentrations are similar to procedures developed during the 1980s. One principal
difference was that most earlier assessments did not include air-to-leaf transfers in their
estimates of pasture grass and cattle fodder concentrations. The other principal difference
was that earlier assessments did not consider soil ingestion by cattle to the extent
considered in this assessment. Analysis showed that critical components of the beef and
milk models include the assumptions of soil ingestion by cattle, and estimates of pasture
grass and cattle fodder concentrations of the dioxin-like compounds. A dichotomy in
11-5 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
model results between the soil source categories and the incinerator emission categories,
similar to the dichotomies noted above, was also observed in predictions of beef and milk
concentrations in the demonstration of source categories. For the soil source categories,
ingestion of soil explained 51-71% of estimated beef and milk concentrations, the
remainder due to ingestion of impacted pasture grass and cattle fodder. For the incinerator
source category, soil ingestion explained only 11-26% of estimated concentrations, the
remainder due to pasture grass and cattle fodder ingestion. The analysis of the beef and
milk concentration algorithms are given in Chapter 10, Section 10.2.12.
6. There are two basic approaches for estimating concentrations in fish. One is
based on water concentrations (soluble phase or total concentrations) and one is based on
bottom sediment concentrations. Estimates of fish concentrations have classically been
made based on water concentrations and a "bioconcentration factor" (BCF), and this
approach has been used in assessment of exposures to 2,3,7,8-TCDD. Bioconcentration
refers to the net accumulation of a chemical from exposure via water only, and BCFs are
most often obtained in laboratory conditions. Another measure of the potential for
chemicals to accumulate in fish tissue is termed the "bioaccumulation factor" (BAF).
Bioaccumulation refers to the net accumulation of a chemical from exposure via food and
sediments as well as water. Fish exposure to strongly hydrophobic contaminants such as
the dioxin-like compounds is not via passive movement through the gills but rather through
the food chain. This argument supports the use of a BAF over a BCF if estimating fish
tissue concentrations based on water column concentrations. This assessment presents
the BAF procedure (Chapter 10, Section 10.2.8.3). The focus of this assessment,
however, is on an approach to estimating fish tissue concentrations based on bottom
sediment concentrations. This procedure uses an organic carbon-based, lipid-based "biota
sediment accumulation factor" (BSAF), which is defined as the ratio of contaminant
concentrations in the lipid of fish to the organic-carbon based concentrations in bottom
sediments. In theory, the BAF and BSAF are analogous indicators of bioaccumulation, and
should therefore lead to similar estimations of fish tissue concentrations if used for that
purpose. One key advantage of the BSAF approach is that bottom sediment
concentrations of the dioxin-like compounds are generally at levels amenable to analytical
measurement, and a relatively large amount of data exists on concentrations of these
11-6 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
compounds in bottom sediments. In contrast, water concentrations of hydrophobic
contaminants are very low so as to preclude analytical measurements; development of
BAFs require computer model simulations to estimate the necessary water column
concentrations. A limited comparative modeling test was conducted on the BAF and BSAF
approaches. General conclusions cannot be made from this exercise due to the
uncertainties in the input parameters, the uncertainties in the modeling of sediment and
water concentrations, and so on. However, the exercise showed that the BSAF and BAF
approaches arrived at comparable concentrations; within a factor of ten with the BAF
approach generally predicting higher than the BSAF approach. Details of this sensitivity
exercise can be found in Chapter 10, Section 10.2.8.3.
11.2. RECOMMENDATIONS
Further research is recommended in the following areas. These are not listed in any
specific priority order.
1. Throughout this document the lack of congener-specific data is cited as a major
source of uncertainty. For example, congener-specific data is lacking for basic chemical
properties such as octanol-water partition coefficients, degradation rates, and vapor
pressures. Also this data is lacking for estimation of incinerator emission factors,
metabolic rate constants, and bioavailability and biotransfer factors. Thus, gathering more
data on the congener-specific properties of these compounds is a high priority for further
research.
2. The use of pharmacokinetics in body burden analysis has shown great potential
for estimating exposure levels. In order to reduce the uncertainty in these procedures,
increased collection of biological samples and improvements in PK model structure and
input parameters are recommended. In addition, further research should be conducted on
the application of these procedures to estimating target organ dose, absorbed dose,
lactational/placental transfers, and effects on offspring.
11-7 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
3. There are key uncertainties in estimating fish tissue concentrations using the
BSAF approach, as there are using the BAF approach. Because of their affinity for organic
carbon, the fate and transport of dioxin-like compounds in water bodies is likely to be more
a function of sediment-related processes rather than water-related processes. Key
sediment processes in water bodies include: sorption/desorption, importance and
prevalence of dissolved organic materials in the water column,
deposition/suspension/resuspension, downstream sediment transport, and so on. Although
some preliminary information on sediment modeling in surface water bodies is presented in
Appendix D, much more development is needed, especially for evaluating pipe discharges.
Further development is needed for the BSAF approach for estimating fish tissue
concentrations. A key issue that has been identified is whether BSAFs that have been
developed for one species and water body are generalizable to another species and another
water body. This question will be difficult to answer because of the several uncertainties
associated with BSAF development that were discussed in Chapter 5: fish migratory
patterns, variability in fish lipid content and other differences within and between species,
study design with regard to fish and sediment sampling, ecosystem differences, and so on.
However, after careful examination of existing data sets and considering key differences
between species (invertebrates vs. vertebrates, fresh water vs. salt water, bottom feeders
vs. water column feeders, etc.), it may be possible to develop a workable system for BSAF
assignment based on key considerations.
4. Considering that beef and dairy exposures are identified as critical exposures in
this assessment, more information is needed on several of the components of the model to
estimate beef and milk concentrations. Such information includes: cattle soil ingestion
rates, pasture grass concentrations and mechanisms of transfer from the air/soil to pasture
grass, inventories of cattle production practices (such as fattening prior to slaughter) and
the impact of these practices to cattle food product concentrations, and models and data
to further develop the bioconcentration factor (termed F in this assessment and translates
concentrations on cattle dry matter intake to concentrations in beef and milk fat).
5. This assessment did not evaluate all possible exposure pathways. Other
potential pathways include ingestion of other farm products such as eggs, chicken, and
11-8 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
pork. The occurrence of these compounds in ground water is expected to be minimal
based on strong sorption to soils. However, they have been found in ground water below
and near sites of industrial contamination. Co-occurrence with other organic compounds,
co-occurrence with solvents, and transport associated with oils have been cited as causes
of enhanced mobility in these settings. Exposures through mother's milk or placental
transfers (as noted above) need to be evaluated and compared with other exposure
pathways. Exposure resulting from bleached paper products, such as occurrences of
2,3,7,8-TCDD leached from milk containers, has also been cited as a route of exposure.
Further evaluation of these pathways is recommended.
6. Recent assessments of exposure to dioxin-like compounds have concluded that
an inventory of currently identified sources of contamination falls well short of explaining
environmental concentrations that have been measured. Researchers have claimed that
key sources of dioxin-like compounds in the environment have yet to be identified. This
issue wasn't explored in this assessment, but is certainly worthy of careful consideration.
Complete source identification and comparative contributions by different sources should
be explored in future assessments.
7. Chapter 10 demonstrates the breadth of variability and uncertainty in the
models and model parameters described in this assessment; some of the more important
areas of uncertainty have been noted in previous recommendations. Reducing uncertainty
in all areas will increase the reliability of estimates using procedures in this assessment and
similar ones. Areas not discussed above include: emission rates of dioxin-like compounds
from different stack emission technologies, vapor/particulate partitioning in air from stack
emitted and soil volatilized residues, overland transport of residues from one site to nearby
land areas and to surface water, and the impact of photodegradation on air-borne residues
and concentrations in biota and surface soils.
11-9 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
APPENDIX A. ENVIRONMENTAL CHEMISTRY
The tables in this appendix are discussed in Chapter 2. References listed at the end
of each table are included in the reference list at the end of Chapter 2. Following are the
tables included in this Appendix:
A-1 Physical and chemical properties for the dioxin-like compounds
A-2 Physical and chemical properties for the dioxin, furan, and
PCB congeners
A-1 7/31/92
-------
TABLE A-l. P-CHEM PROPERTIES for the DIOXIN-LIKE CONGENERS
Chemical
CAS No.
Melting
Point
c°*
Melting
Point
Ref.
Water
Solubility
mg/T
ws
Temp.
WS
Ref.
Vapor
Pressure
mm Hg*
VP
Temp.
VP
Ref.
Henry's
Constant
atm-m'/mol*
Henry's
Constant
Ref.
Log
KJ1
Log
K..
Ref.
Photo
Quantum
Yield"
Photo
Quantum
Yield
Ref.
Tetrachlorodibenzo-p-dioxins (MW= 321 .98)
2.3.7.8-TCDD
1746-01-6
Homologue Group Average
305-306
9
1.90E-05
(3.50E-04)
25
25
1.2
20
7.40E-10
(1.36E-08)
25
25
2,3
20
(1.65E-05)
(4.06E-05)
2,3
20
6.64
(6.52)
2,4
20
2.2E-03b
22
PentachIorodibenzo-p-dioxins(MW=356.42)
1,2,3,7,8-PeCDD
40321-76-4
Homologue Group Average
240-241
9
(1.20E-04)
20
20
(4.35E-10)
(5.60E-10)
25
25
9
20
(2.19E-06)
20
6.64
(6.57)
10
20
Hexachlorodibenzo-p-dioxins(MW=390.87)
1.2.3.4.7.8-HxCDD
39227-28-6
l,2.3.6.7.8-HxCDD
57653-85-7
l,2.3.7.8.9-HxCDD
19408-74-3
Homologue Group Average
273-275
285-286
243-244
9
9
9
4.40E-06
(4.40E-06)
20
20
6
20
(3.82E-11)
(3.60E-11)
(4.88E-11)
(4.35E-11)
25
25
25
25
9
9
9
20
(4.47E-06)
(4.47E-06)
19
20
(7.79)
(7.25)
8
20
1.1E-04'
23
Heptachlorodibenzo-p-dioxins (MW = 425 .3 1)
1,2,3,4,6.7,8-HpCDD
35822-46-9
Homologue Group Average
264-265
9
2.40E-06
(2.40E-06)
20
20
6
20
(5.62E-12)
(5.62E-12)
25
25
9
20
(1.31E-06)
(1.31E-06)
19
20
(8.20)
(8.20)
8
20
1.53E-05C
23
Octachlorodibenzo-p-dioxins (MW= 460.76)
1,2,3,4,6,7.8.9-OCDD
3268-87-9
325-326
6
4.00E-07
20
6
8.25E-13
25
9
(6.74E-06)
5
(8.60)
8
2.26E-05C
23
A-2
-------
TABLE A-l (continued)
Chemical
CAS No.
Melting
Point
c°*
Melting
Point
Ref.
Water
Solubility
mg/r
ws
Temp.
WS
Ref.
Vapor
Pressure
mm Hg*
VP
Temp.
VP
Ref.
Henry's
Constant
atm-m'/mol*
Henry's
Constant
Ref.
Log
K«'
Log
K«
Ref.
Photo
Quantum
Yield'
Photo
Quantum
Yield
Ref.
Tetrachlorodibenzoiurans (MW = 305 .98)
2,3,7,8-TCDF
51207-31-9
Homologue Group Average
227-228
21
4.19E-04
(4.19E-04)
22.7
22.7
11
20
(1.50E-08)
(2.49E-08)
25
25
21
20
(1.44E-05)
(1.44E-05)
19
20
6.53
(6.21)
10
20
Pentachlorodibenzofurans (MW=340.42)
1,2,3,7.8-PeCDF
57117-41-6
2,3,4,7,8-PeCDF
57117-31-4
Homologue Group Average
225-227
196-
196.5
21
21
2.36E-04
(2.36E-04)
22.7
22.7
11
20
(1.72E-09)
(2.63E-09)
(2.57E-09)
25
25
25
21
21
20
(4.99E-06)
(4.99E-06)
19
20
6.79
6.92
(6.44)
10
10
20
Hexachlorodibenzofurans(MW=374.87)
1,2,3,4,7,8-HxCDF
70648-26-9
1.2.3.6.7,8-HxCDF
57117-44-9
1,2,3,7,8,9-HxCDF
72918-21-9
2.3,4.6,7,8-HxCDF
60851-34-5
Homologue Group Average
225.5-
226.5
232-234
246-249
239-240
21
21
21
21
8.25E-06
1.77E-05
(1.30E-05)
22.7
22.7
22.7
11
11
20
(2.40E-10)
(2.18E-10)
(1.95E-10)
(2.83E-10)
25
25
25
25
21
21
21
20
(1.43E-05)
(6.08E-06)
(1.02E-05)
19
19
20
6.96E-04b
25
Heptachlorodibenzofarans (MW =409 .3 1)
1.2.3,4.6.7.8-HpCDF
67562-39-4
236-237
21
1.35E-06
22.7
11
(3.53E-11)
25
21
(1.41E-05)
19
7.92
10
A-3
-------
TABLE A-l (continued)
Chemkal
CAS No.
1.2.3.4,7,8.9-HpCDF
55673-89-7
Homologue Group Average
Melting
Point
Co.
221-223
Melting
Point
Ref.
21
Water
Solubility
rng/1*
(1.35E-06)
ws
Temp.
22.7
WS
Ref.
20
Vapor
Pressure
mm Hg*
(4.65E-11)
(4.65E-11)
VP
Temp.
25
25
VP
Ref.
21
20
Henry's
Constant
atm-m'/mol*
(1.41E-05)
Henry's
Constant
Ref.
20
Log
K_«
(7.92)
Log
K«
Ref.
20
Photo
Quantum
Yield-
Photo
Quantum
Yield
Ref.
Octachlorodibenzofurans (MW=444.76)
1,2,3,4.6,7,8.9-OCDF
39001-02-0
258-260
21
(1.16E-06)
25
11
3.75E-12
25
21
(1.89E-06)
19
(8.78)
8
Tetrachloro-PCB (MW=291.99)
3,3',4,4'-TeCB
32598-13-3
3,4,4',5-TeCB
70362-50-4
453
(410)
17
17
5.69E-04
(2.92E-03)
25
25
12
17
(1.34E-05)
(1.77E-05)
25
25
18
18
9.40E-05
(1.48E-04)
13
16
6.21
(6.36)
15
15
Pentachloro-PCB (MW=326.44)
2,3,3',4,4'-PeCB
32598-14-4
2.3,4,4',5-PeCB
74472-37-0
2,3',4,4',5-PeCB
31508-00-6
3,3',4.4',5-PeCB
57465-28-8
(398)
(392)
378
(398)
17
17
17
17
(2.06E-03)
(2.59E-03)
(2.06E-03)
(1.03E-03)
25
25
25
25
17
17
17
17
(5.86E-06)
(9.00E-06)
(8.50E-06)
(2.90E-06)
25
25
25
25
18
18
18
18
(6.00E-05)
(1.14E-04)
(1.16E-04)
(5.40E-05)
16
16
16
16
(6.65)
(6.65)
(6.74)
(6.89)
15
15
15
15
Hexachloro-PCB (MW=360.88)
2,3,3'.4,4',5-HxCB
38380-08-4
2,3,3',4,4'.5'-HxCB
69782-90-7
(414)
(414)
17
17
(3.61E-04)
(3.61E-04)
25
25
17
17
(1.39E-06)
(1.23E-06)
25
25
18
18
(2.20E-05)
(6.60E-05)
16
16
7.13
7.20
14
14
A-4
-------
TABLE A-1 (continued)
Chemical
CAS No.
2,3',4,4',5,5'-HxCB
52663-72-6
3,3',4,4',5,5'-HxCB
32774-16-6
Melting
Point
C01
(408)
(485)
Melting
Point
Ref.
17
17
Water
Solubility
mg/P
(3.61E-04)
(3.61E-05)
WS
Temp.
25
25
ws
Ref.
17
17
Vapor
Pressure
mmHg*
(1.95E-06)
(1.52E-06)
VP
Temp.
25
25
VP
Ref.
18
20
Henry's
Constant
atm-urVmoP
(1.23E-04)
(5.90E-05)
Henry's
Constant
Ref.
16
16
Log
K~'
7.26
7.47
Log
*~
Ref.
14
14
Heptachloro-PCB (MW=396.33)
2,3,3',4,4',5,5'-HpCB
39635-31-9
(431)
17
(6.28E-05)
25
17
(3.04E-07)
25
18
3.00E-3
19
(7-71)
15
Photo
Quantum
Yield-
Photo
Yield
Ref.
Footnote References
* Values are presented as they appeared in the referenced article. Values in ( ) are calculated.
b Tested in solution of water:acetonitrile (1:1 v/v) at 313 nm.
c Tested in solution of water:acetonitrile (2:3 v/v) at 313 nm.
1. Marple et al. (1986a)
2. USEPA (1990)
3. Podoll et al. (1986)
4. Marple et al. (1986b)
5. Shiu et al. (1988)
6. Friesen et al. (1985)
8. Burkhard and Kuehl (1986)
9. Rordorf(1987)
10. Sijm et al. (1989)
11. Friesen et al. (1990)
12. Dickhut et al. (1986)
13. Dunnivant and Elzerman (1988)
14. Risby et al. (1990)
15. Hawker and Connell (1988)
16. Sabljic and Gusten (1989)
17. Abramowitz and Yalkowsky (1990)
18. Foreman and Bidleman (1985)
19. Calculated by the VP/WS ratio technique
20. Average of all literature values (measured and calculated) within a homologue group
21. Rordorf(1989)
22. Dulin et al. (1986)
23. Choudhry and Webster (1987)
A-5
-------
TABLE A-2. P-CHEM PROPERTIES for the DIOXIN, FURAN, and PCB CONGENERS
Chemical
CAS No.
Melting
Point
|"«O«
Melting
Point
Ref.
Water
Solubility
mg/1'
ws
Temp.
WS
Ref.
Vapor
Pressure
mm Hg*
VP
Temp.
VP
Ref.
Henry's
Constant
atm-m'/nioP
Henry's
Constant
Ref.
Log
K./
Log
K»
Ref.
Photo
Quantum
Yield'
Photo
Quantum
Yield
Ref.
Tetrachlorodibenzo-p-dioxins (MW= 321.98)
1,2,3,4-TCDD
30746-58-8
1,2,3,6-TCDD
71669-25-5
1,2,3,7-TCDD
67028-18-6
1.2,3.8-TCDD
53555-02-5
1,2,3.9-TCDD
71669-26-6
1,2,4,6-TCDD
71669-27-7
1.2,4,7-TCDD
71669-28-8
1,2,4,8-TCDD
71669-29-9
1.2,4,9-TCDD
71665-99-1
1,2,6,7-TCDD
40581-90-6
1.2.6,8-TCDD
67323-56-2
1,2,6,9-TCDD
40581-91-7
172-175
6
6.30E-04
4.30E-04
25
20
5
6
4.80E-08
(7.50E-09)
25
25
9
9
(3.72E-05)
(7.39E-06)
5
19
6.60
6.86
(6.91)
6.48
6.39
6.10
6.25
6.25
6.10
6.43
5
10
8
10
10
10
10
10
10
10
5.42E-040
24
A-6
-------
TABLE A-2 (continued)
Chemical
CAS No.
1,2.7,8-TCDD
34816-53-0
1.2,7,9-TCDD
71669-23-3
1,2,8,9-TCDD
62470-54-6
1,3,6,8-TCDD
33423-92-6
1,3,6,9-TCDD
71669-24-4
1,3,7,8-TCDD
50585-46-1
1,3,7,9-TCDD
62470-53-5
1,4,6,9-TCDD
40581-93-9
1,4,7,8-TCDD
40581-94-0
2,3,7,8-TCDD
1746-01-6
Homologue Group Average
Melting
Point
fO*
219-220
193.5-
195
305-306
Melting
Point
Ref.
6
9
9
Water
Solubility
mg/I*
3.20E-04
1.90E-05
(3.50E-04)
ws
Temp.
20
22
25
WS
Ref.
6
1.2
20
Vapor
Pressure
mm Hg'
(5.25E-09)
(6.30E-09)
7.40E-10
(1.36E-08)
VP
Temp.
25
25
25
25
VP
Ref.
9
9
2,3
20
Henry's
Constant
atm-ui'/uiol*
6.81E-05
(1.65E-05)
(4.06E-05)
Henry's
Constant
Ref.
7
2,3
20
Log
K,w-
6.38
6.86
7.20
6.25
6.30
7.06
6.38
6.39
6.64
(6.52)
Log
*-
Ref.
10
10
8
10
10
8
10
10
2,4
20
Photo
Quantum
Yield'
2.17E-03'
2.2E-03"
Photo
Quantum
Yield
Ref.
24
22
Pentachlorodibenzo-p-dioxins (MW=356 ,42)
1,2,3,4,6-PeCDD
67028-19-7
6.30
10
A-7
-------
TABLE A-2 (continued)
Chemical
CAS No.
1,2,3,4.7-PeCDD
39227-61-7
1.2,3.6,7-PeCDD
71925-15-0
1,2,3.6,8-PeCDD
71925-16-1
1,2-3,6,9-PeCDD
82291-34-7
1,2,3,7,8-PeCDD
40321-76-4
1.2,3,7.9-PeCDD
71925-17-2
1,2,3,8,9-PeCDD
71925-18-3
1,2,4,6,7-PeCDD
82291-35-8
1,2,4,6,8-PeCDD
71998-76-0
1,2.4,6,9-PeCDD
82291-36-9
1,2,4,7,8-PeCDD
58802-08-7
1,2,4,7,9-PeCDD
82291-37-0
Melting
Point
Co.
195-196
240-241
206
Melting
Point
Ref.
9
9
9
Water
Solubility
mg/1"
1.20E-04
ws
Temp.
20
WS
Ref.
6
Vapor
Pressure
mm Hg*
(6.60E-10)
(4.35E-10)
(5.85E-10)
VP
Temp.
25
25
25
VP
Ref.
9
9
9
Henry's
Constant
atm-m'/mol*
(2.58E-06)
Henry's
Constant
Ref.
19
Log
«„„•
(7.44)
6.74
6.53
6.24
6.64
6.40
6.60
6.20
Log
K~
Ref.
8
10
10
10
10
10
10
10
Photo
Quantum
Yield'
9.8E-05'
Photo
Quantum
Yield
Ref.
23
A-8
-------
TABLE A-2 (continued)
Chemical
CAS No,
1,2,4,8,9-PeCDD
82291-38-1
Homologue Group Average
Melting
Point
f*0*
Melting
Point
Ref.
Water
Solubility
mg/1*
(1.20E-04)
WS
Temp,
20
WS
Ref.
20
Vapor
Pressure
mm Hg*
(5.60E-10)
VP
Temp.
25
VP
Ref.
20
Henry's
Constant
aUn-m'/moP
(2.19E-06)
Henry's
Constant
Ref.
20
Log
K..-
(6.57)
Log
K~
Ref.
20
Photo
Quantum
Yield'
Photo
Quantum
Yield
Ref.
Hexachlorodibenzo-p-dioxins (MW = 390. 87)
1,2,3,4,6,7-HxCDD
58200-66-1
1,2,3,4,6,8-HxCDD
58200-67-2
1,2,3,4,6,9-HxCDD
58200-68-3
1,2,3,4,7,8-HxCDD
39227-28-6
1,2,3,6,7,8-HxCDD
57653-85-7
1,2.3,6.7.9-HxCDD
64461-98-9
1,2,3,6,8.9-HxCDD
58200-69-4
1.2,3.7,8,9-HxCDD
19408-74-3
1,2,4,6,7,9-HxCDD
39227-62-8
1,2,4,6,8,9-HxCDD
58802-09-8
273-275
285-286
243-244
238-240
9
9
9
9
4.40E-06
20
6
(3.82E-11)
(3.60E-11)
(4.88E-11)
(5.10E-11)
25
25
25
25
9
9
9
9
(4.47E-06)
19
6.85
(7.79)
7.59
7.59
6.85
6.85
10
8
10
10
10
10
1.1E-04C
23
A-9
-------
TABLE A-2 (continued)
Chemkal
CAS No.
Homoiogue Group Average
Melting
Point
c«.
Melting
Point
Ref.
Water
Solubility
mg/1*
(4.40E-06)
WS
Temp.
20
WS
Ref.
20
Vapor
Pressure
mm Hg*
(4.35E-11)
VP
Temp.
25
VP
Ref.
20
Henry's
Constant
atm-m'/mol"
(4.47E-06)
Henry's
Constant
Ref.
20
Log
K^'
(7.25)
Log
K~
Ref.
20
Photo
Quantum
Yield'
Photo
Quantum
Yield
Ref.
Heptachlorodibenzo-p-
-------
TABLE A-2 (continued)
Chemical
CAS No.
1,2,4,7-TCDF
83719-40-8
1,2,4,8-TCDF
64126-87-0
1,2,4,9-TCDF
83704-24-9
1,2,6,7-TCDF
83704-25-0
1,2,6,8-TCDF
83710-07-0
1,2,6,9-TCDF
70648-18-9
1,2,7,8-TCDF
58802-20-3
1,2,7,9-TCDF
83704-26-1
1,2,8,9-TCDF
70648-22-5
1,3,4,6-TCDF
83704-27-2
1,3,4,7-TCDF
70648-16-7
1,3,4,8-TCDF
92341-04-3
Melting
Point
c»-
191-193
199-200
210-211
184-185
Melting
Point
Ref.
21
21
21
21
Water
Solubility
mg/1'
ws
Temp.
WS
Ref.
Vapor
Pressure
mm Hg*
(2.25E-08)
(2.02E-08)
(1.80E-08)
(2.48E-08)
VP
Temp.
25
25
25
25
VP
Ref.
21
21
21
21
Henry's
Constant
atm-m'/mol"
Henry's
Constant
Ref.
Log
K..-
6.31
6.25
6.23
6.25
6.31
6.23
6.13
Log
*„
Ref.
10
10
10
10
10
10
10
Photo
Quantum
Yield-
Photo
Quantum
Yield
Ref.
A-ll
-------
TABLE A-2 (continued)
Chemical
CAS No.
1,3,4.9-TCDF
83704-28-3
1,3,6,7-TCDF
57117-36-9
1.3,6.8-TCDF
71998-72-6
1.3,6,9-TCDF
83690-98-6
1,3,7,8-TCDF
571 17-35-8
1.3,7,9-TCDF
64560-17-4
1,4,6,7-TCDF
66794-59-0
1,4.6,8-TCDF
82911-58-8
1,4,6,9-TCDF
70648-19-0
1,4,7,8-TCDF
83704-29-4
1,6,7,8-TCDF
83704-33-0
2.3.4,6-TCDF
83704-30-7
Melting
Point
/^O«
176.5-
177
177-178
206.5-
207.5
180-181
153-154
Melting
Point
Ref.
21
21
21
21
21
Water
Solubility
mg/1*
WS
Temp,
WS
Ref.
Vapor
Pressure
mm Hg*
(2.78E-08)
(2.70E-08)
(1.88E-08)
(2.62E-08)
(3.98E-08)
VP
Temp.
25
25
25
25
25
VP
Ref.
21
21
21
21
21
Henry's
Constant
atm-mVmoI*
Henry's
Constant
Ref.
Log
K,,,-
5.89
6.37
6.34
6.34
6.15
5.60
6.17
6.11
Log
K^
Ref.
10
10
10
10
10
10
10
10
Photo
Quantum
Yield-
Photo
Quantum
Yield
Ref.
A-12
-------
TABLE A-2 (continued)
Chemical
CAS No.
2.3.4,7-TCDF
83704-31-8
2,3.4,8-TCDF
83704-32-9
2.3.6,7-TCDF
57117-39-2
2,3.6,8-TCDF
57117-37-0
2.3.7,8-TCDF
51207-31-9
2,4,6,7-TCDF
57117-38-1
2,4,6,8-TCDF
58802-19-0
3,4,6,7-TCDF
57117-40-5
Homologue Group Average
Melting
Point
Co.
172.5-
173
176-177
195-196
202-203
227-228
164-
165.5
198-200
Melting
Point
Ref.
21
21
21
21
21
21
21
Water
Solubility
mg/1*
4.19E-04
(4.19E-04)
WS
Temp.
22.7
22.7
WS
Ref.
11
20
Vapor
Pressure
mm Hg*
(2.92E-08)
(2.78E-08)
(2.10E-08)
(1.95E-08)
(1.50E-08)
(3.30E-08)
(2.02E-08)
(2.49E-08)
VP
Temp.
25
25
25
25
25
25
25
25
VP
Ref.
21
21
21
21
21
21
21
20
Henry's
Constant
atin-m'/mol"
(1.44E-05)
(1.44E-05)
Henry's
Constant
Ref.
19
20
Log
K.,'
6.06
6.31
6.73
6.53
6.25
6.17
(6.21)
Log
1C
Ref.
10
10
10
10
10
10
20
Photo
Quantum
Yield*
Photo
Quantum
Yield
Ref.
PentachIorodibenzofurans(MW=340.42)
1,2,3,4,6-PeCDF
83704-47-6
1,2,3.4,7-PeCDF
83704-48-7
1.2,3,4,8-PeCDF
67517-48-0
194-195
177-178
21
21
(2.70E-09)
(3.60E-09)
25
25
21
21
6.53
6.79
10
10
A-13
-------
TABLE A-2 (continued)
Chemkal
CAS No.
1,2,3.4,9-PeCDF
83704-49-8
1,2,3,6,7-PeCDF
57117-42-7
1, 2,3.6. 8-PeCDF
83704-51-2
1,2,3,6,9-PeCDF
83704-52-3
1,2.3,7,8-PeCDF
57117-41-6
1,2,3,7,9-PeCDF
83704-53-4
1.2,3,8,9-PeCDF
83704-54-5
1,2,4,6,7-PeCDF
83704-50-1
1,2,4,6,8-PeCDF
69698-57-3
1,2,4,6,9-PeCDF
70648-24-7
1.2,4,7,8-PeCDF
58802-15-6
1,2,4,7,9-PeCDF
71998-74-8
Melting
Point
c«.
205-207
225-227
178.5-
179.5
204-205
236-238
196-197
Melting
Point
Ref.
21
21
21
21
21
21
Water
Solubility
mg/T
WS
Temp.
WS
Ref.
Vapor
Pressure
mm Hg*
(2.25E-09)
(1.72E-09)
(3.52E-09)
(2.32E-09)
(1.50E-09)
(2.63E-09)
VP
Temp.
25
25
25
25
25
25
VP
Ref.
21
21
21
21
21
21
Henry's
Constant
atin-m'/ujoP
Henry's
Constant
Ref.
Log
K»'
6.26
6.33
6.79
6.27
6.34
6.59
6.26
6.19
Log
K~
Ref.
10
10
10
10
10
10
10
10
Photo
Quantum
Yield'
1.30E-02b
Photo
Quantum
Yield
Ref.
25
A-14
-------
TABLE A-2 (continued)
Chemical
CAS No.
1,2,4,8,9-PeCDF
70648-23-6
1.2,6,7,8-PeCDF
69433-00-7
1,2,6,7,9-PeCDF
70872-82-1
1,3,4,6,7-PeCDF
83704-36-3
1,3,4,6,8-PeCDF
83704-55-6
1,3,4,6,9-PeCDF
70648-15-6
1,3,4,7,8-PeCDF
58802-16-7
1,3,4,7,9-PeCDF
70648-20-3
1,3,6,7,8-PeCDF
70648-21-4
1,4,6.7,8-PeCDF
83704-35-2
2,3,4,6,7-PeCDF
57117-43-8
2,3,4,6,8-PeCDF
67481-22-5
Melting
Point
C"
220-221
195-
195.5
168-170
201.5-
202
219-220
Melting
Point
Ref.
21
21
21
21
21
Water
Solubility
mg/1'
WS
Temp.
WS
Ref.
Vapor
Pressure
mm Hg*
(1.88E-09)
(2.70E-09)
(4.28E-09)
(2.40E-09)
(1.88E-09)
VP
Temp.
25
25
25
25
25
VP
Ref.
21
21
21
21
21
Henry's
Constant
atm-m'/mol'
Henry's
Constant
Ref.
Log
K»-
6.42
6.51
6.19
6.24
6.34
6.33
6.53
6.47
6.59
Log
K~
Ref.
10
10
10
10
10
10
10
10
10
Photo
Quantum
Yield*
Photo
Quantum
Yield
Ref.
A-15
-------
TABLE A-2 (continued)
Chemical
CAS No.
2,3,4,7.8-PeCDF
57117-31-4
Homologue Group Average
1.2.3,4.6,7-HxCDF
79060-60-9
1.2.3.4.6.8-HxCDF
69698-60-8
1.2.3.4.6.9-HxCDF
91538-83-9
1,2,3,4,7,8-HxCDF
70648-26-9
1,2.3,4,7,9-HxCDF
91538-84-0
1.2.3.4,8.9-HxCDF
92341-07-6
1.2.3.6,7,8-HxCDF
57117-44-9
1,2.3,6.7,9-HxCDF
92341-06-5
1,2,3.6, 8,9-HxCDF
75198-38-8
l,2.3.7.8.9-HxCDF
72918-21-9
Melting
Point
Co.
196-
196.5
227-228
233.5-
234
196-197
225.5-
226.5
216-217
232-234
206-207
246-249
Melting
Point
Ref.
21
21
21
21
21
21
21
21
21
Water
Solubility
rng/1'
2.36E-04
(2.36E-04)
ws
Temp.
22.7
22.7
WS
Ref.
11
20
Vapor
Pressure
mm Hg*
(2.63E-09)
(2.57E-09)
VP
Temp.
25
25
VP
Ref.
21
20
Henry's
Constant
atm-m'/inol'
(4.99E-06)
(4.99E-06)
Henry's
Constant
Ref.
19
20
Log
K..-
6.92
(6.44)
Log
K~
Ref.
10
20
Photo
Quantum
Yield*
Hexachlorodibenzofurans(MW=374.87)
8.25E-06
1.77E-05
22.7
22.7
11
11
(2.40E-10)
(2.18E-10)
(4.13E-10)
(2.40E-10)
(2 85E-10)
(2.18E-10)
(3.38E-10)
(1.80E-10)
25
25
25
25
25
25
25
25
21
21
21
21
21
21
21
21
(1.43E-05)
(6.08E-06)
19
19
6.96E-04b
Photo
Quantum
Yield
Ref.
25
A-16
-------
TABLE A-2 (continued)
Chemical
CAS No.
1.2,4.6,7,8-HxCDF
67562-40-7
1.2.4,6.7.9-HxCDF
75627-02-0
1. 2.4,6. 8,9-HxCDF
69698-59-5
1.3.4.6,7.8-HxCDF
71998-75-9
1.3,4.6,7,9-HxCDF
92341-05-4
2.3.4,6.7,8-HxCDF
60851-34-5
Homologue Group Average
Melting
Point
Co.
221-222
180-181
246-248
229-230
239-240
Melting
Point
Ref.
21
21
21
21
21
Water
Solubility
mg/T
(1.30E-05)
WS
Temp.
22.7
WS
Ref.
20
Vapor
Pressure
mm Hg*
(2.63E-10)
(5.70E-10)
(1.80E-10)
(2.33E-10)
(1.95E-10)
(2.83E-10)
VP
Temp.
25
25
25
25
25
25
VP
Ref.
21
21
21
21
21
20
Henry's
Constant
atm-m'/mol*
(1.02E-05)
Henry's
Constant
Ref.
20
Log
K.,-
Log
K~
Ref.
Photo
Quantum
Yield'
Photo
Quantum
Yield
Ref.
Heptachlorodibenzofurans (MW = 409 ,31)
1.2.3.4,6,7.8-HpCDF
67562-39-4
1,2.3.4,6,7,9-HpCDF
70648-25-8
1,2,3,4,6,8,9-HpCDF
69698-58-4
1.2,3,4.7.8,9-HpCDF
55673-89-7
Horaologue Group Average
236-237
211-212
221-223
21
21
21
1.35E-06
(1.35E-06)
22.7
22.7
11
20
(3.53E-11)
(5.78E-11)
(4.65E-11)
(4.65E-11)
25
25
25
25
21
21
21
20
(1.41E-05)
(1.41E-05)
19
20
7.92
(7.92)
10
20
A-17
-------
TABLE A-2 (continued)
Chemical
CAS No.
Melting
Point
c«.
Melting
Point
Ref.
Water
Solubility
mg/1"
WS
Temp.
WS
Ref.
Vapor
Pressure
mm Hg*
VP
Temp.
VP
Ref.
Henry's
Constant
atm-m'/fflol*
Henry's
Constant
Ref.
Log
K..'
Log
K~
Ref.
Photo
Quantum
Yield'
Photo
Quantum
Yield
Ref.
Octachlorodibenzofiirans (MW=444.76)
1.2,3,4.6.7.8.9-OCDF
39001-02-0
258-260
21
(1.16E-06)
25
11
3.75E-12
25
21
(1.89E-06)
19
(8.78)
8
Tetrachloro-PCB (MW=291.99)
3.3',4,4'-TeCB
32598-13-3
3.4,4',5-TeCB
70362-50-4
453
(410)
17
17
5.69E-04
(2.92E-03)
25
25
12
17
(1.34E-05)
(1.77E-05)
25
25
18
18
9.40E-05
(1.48E-04)
13
16
6.21
(6.36)
15
15
Pentachloro-PCB (MW=326.44)
2,3,3',4,4'-PeCB
32598-14-4
2,3,4,4',5-PeCB
74472-37-0
2,3',4,4',5-PeCB
31508-00-6
3,3',4,4',5-PeCB
57465-28-8
(398)
(392)
378
(398)
17
17
17
17
(2.06E-03)
(2.59E-03)
(2.06E-03)
(1.03E-03)
25
25
25
25
17
17
17
17
(5.86E-06)
(9.00E-06)
(8.50E-06)
(2.90E-06)
25
25
25
25
18
18
18
18
(6.00E-05)
(1.14E-04)
(1.16E-04)
(5.40E-05)
16
16
16
16
(6.65)
(6.65)
(6.74)
(6.89)
15
15
15
15
Hexachloro-PCB (MW=360.88)
2.3.3',4,4',5-HxCB
38380-08-4
2,3,3',4,4',5'-HxCB
69782-90-7
2,3',4,4',5,5'-HxCB
52663-72-6
(414)
(414)
(408)
17
17
17
(3.61E-04)
(3.61E-04)
(3.61E-04)
25
25
25
17
17
17
(1.39E-06)
(1.23E-06)
(1.95E-06)
25
25
25
18
18
18
(2.20E-05)
(6.60E-05)
(1.23E-04)
16
16
16
7.13
7.20
7.26
14
14
14
A-18
-------
TABLE A-2 (continued)
Chemical
CAS No.
3,3',4,4',5,5'-HxCB
32774-16-6
Melting
Point
c°*
(485)
Melting
Point
Ret.
17
Water
Solubility
rog/1'
(3.61E-05)
WS
Temp.
25
WS
Ref.
17
Vapor
Pressure
mm Hg*
(1.52E-06)
VP
Temp.
25
VP
Ref.
20
Henry's
Constant
atni-m'/mol"
(5.90E-05)
Henry's
Constant
Ref.
16
Log
K^
7.47
Log
K~
Ref.
14
Photo
Quantum
Yield'
Photo
Quantum
Yield
Ref.
Heptachloro-PCB (MW=396.33)
2,3,3',4,4',5,5'-HpCB
39635-31-9
(431)
17
(6.28E-05)
25
17
(3.04E-07)
25
18
3.00E-3
19
(7.71)
15
Footnote References
* Values are presented as they appeared in the referenced article. Values in ( ) are calculated.
b Tested in solution of wateracetonitrile (1:1 v/v) at 313 run.
0 Tested in solution of water:acetonitrile (2:3 v/v) at 313 run.
1. Marple et al. (1986a)
2. USEPA (1990)
3. Podoll et al. (1986)
4. Marple et al. (1986b)
5. Shiu et al. (1988)
6. Friesen et al. (1985)
7. Webster et al. (1985)
8. Burkhard and Kuehl (1986)
9. Rordorf(1987)
10. Sijm et al. (1989)
11. Friesen et al. (1990)
12. Dickhut et al. (1986)
13. Dunnivantand Elzerman (1988)
14. Risby et al. (1990)
15. Hawker and Connell (1988)
16. Sabljic and Gusten (1989)
17. Abramowitz and Yalkowsky (1990)
18. Foreman and Bidleman (1985)
19. Calculated by the VP/WS ratio technique
20. Average of all literature values (measured and calculated) within a homologue group
21. Rordorf(1989)
22. Dulin et al. (1986)
23. Choudhry and Webster (1987)
24. Choudhry and Webster (1989)
25. Choudhry et al. (1990)
A-19
-------
DRAFT-DO NOT QUOTE OR CITE
APPENDIX B. ENVIRONMENTAL CONCENTRATIONS
The tables in this appendix are discussed in Chapter 3. References listed at the end
of each table are included in the reference list at the end of Chapter 3. Following are the
tables included in this Appendix:
Table B-1.
Table B-2.
Table B-3.
Table B-4.
Table B-5.
Table B-6.
Table B-7.
Table B-8.
Table B-9.
Table B-10.
Table B-11.
Table B-12.
Table B-13.
Table B-14.
Table B-15.
Table B-1 6.
Table B-1 7.
Table B-18.
Table B-19.
Table B-20.
Table B-21.
Environmental Levels of Dioxins in Soil (ppt).
Environmental Levels of Dibenzofurans in Soil (ppt).
Environmental Levels of Dioxins in Water (ppq).
Environmental Levels of Dibenzofurans in Water (ppq).
Environmental Levels of Dioxins in Sediments (ppt).
Environmental Levels of Dibenzofurans in Sediments (ppt).
Environmental Levels of PCBs in Sediment (ppt).
Environmental Levels of Dioxins in Fish (ppt).
Environmental Levels of Dibenzofurans in Fish (ppt).
Environmental Levels of PCBs in Fish (ppt).
Levels of Dioxins in Food Products (ppt).
Levels of Dibenzofurans in Food Products (ppt).
Environmental Levels of Dioxins in Air (pg/m3).
Environmental Levels of Dibenzofurans in Air (pg/m3).
Mean Background Environmental Levels of Dioxins in Soil
(ppt).
Mean Background Environmental Levels of Dibenzofurans in
Soil (ppt).
Mean Background Levels of Dioxins in Water (ppq).
Mean Background Levels of Dibenzofurans in Water (ppq).
Mean Background Environmental Levels of Dioxins in
Sediments (ppt).
Mean Background Environmental Levels of Dibenzofurans in
Sediment (ppt).
Mean Background Environmental Levels of PCBs in Sediment
(ppt).
B-1
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table B-22.
Table B-23.
Table B-24.
Table B-25.
Table B-26.
Table B-27.
Mean Background Environmental Levels of Dioxins in Finfish
(ppt).
Mean Background Levels of Dibenzofurans in Finfish (ppt).
Mean Background Levels of Dioxins in Food Products (ppt).
Mean Background Levels of Dibenzofurans in Food Products
(ppt).
Mean Background Environmental Levels of Dioxins in Air
(pg/m3).
Mean Background Environmental Levels of Dibenzofurans in
Air (pg/m3).
B-2
7/31/92
-------
Table B-l. Environmental Levels of Dioxins in Soil (ppt)
Chemical
Number
samples
Number
positive
samples
Concentration
range
(PPO
Cone,
mean
Location
Location description
Sample
year
Ref.
no.
Comments
Tetrachlorodibenzo-p-dioxins (MW= 321 .98)
2,3,7,8-TCDD
TCDDs
77
3
2
23
62
13
22
4
NR
33
11
20
8
4
12
19
1
77
NR
NR
NR
NR
47
NR
0
2
23
59
1
6
0
NR
33
9
13
8
0
12
6
1
NR
NR
NR
NR
NR
11
<0.5-2.1
< 0.2- < 2.0
2.4-0.84
10-36000
ND-270
ND-2
ND-5
ND
NR
41-52000
ND-590
ND-9.4
0.6-3.1
ND
1-7
ND-4.2
3.2
<0.5-69
200
2.8
0.9
874
ND-430
<0.5
<1.43
1.62
2133
55
<1
1
NA
2
4300
145
2
2
NA
3
1
NA
9.4
NR
NR
NR
NR
40.3
British Isles
Various parts of Europe
Various parts of Europe
Midland, MI
Midland, MI
Henry, IL
Middletown, OH
MN
Finland
Midland, MI
Midland, MI
US
Sweden
Elk River, MN
England
England
Various parts of Europe
British Isles
Muggenburger st.
Hamburg,Germany
Kirchsteinbek,
Hamburg, Germany
Ochsenwerder
Landscheideweg,
Hamburg, Germany
Moorefleeter Brack
Hamburg,Germany
Ontario and U.S. Midwestern
States.
Background
Rural
Industrial
Industrial
Residential
Residential
Residential
Pristine
Industrial
Industrial
Inddustrial
Industrial
Urban
Rural
Residential
Urban
Rural
Background
Industrial
Industrial
Contaminated site
Contaminated site
Urban
NR
NR
NR
NR
1985
1985
1985
1985
11
10
10
1
1
1
1
1
2
3
3
3
5
6
7
8
10
11
12
12
12
12
13
Urban area
Near Stockholm
Agriculture
Rural
Maximum contents
reported
Maximum contents
reported
Maximum contents
reported
Maximum contents
reported
B-3
-------
Table B-l. Environmental Levels of Dioxins in Soils (ppt) (continued)
Chemical
TCDDs (continued)
Number
samples
2
1
7
5
3
NR
11
12
53
29
8
4
12
19
Number
positive
samples
2
1
5
0
0
NR
NR
NR
0
NR
8
0
12
19
Concentration
range
(PPO
11.2-55.5
320
ND-290
ND
ND
NR
ND-7
ND-430
ND
ND-1200
37-217
ND
17-120
9-160
Cone.
mean
33.35
NA
109
NA
NA
89
<1
69
NA
69
98
NA
42
65
Location
Various parts of Europe
Midland, MI
Midland, MI
Middletown, OH
MN
Finland
Canada
Canada
Canada
Canada
Sweden
Elk River, MN
England
England
Location description
Industrial
Industrial
Residential
Residential
Pristine
Industrial
Rural
Urban
Urban
Agriculture
Residential
Urban
Sample
year
Ref.
no.
10
1
1
1
1
2
4
4
4
4
5
6
7
8
Comments
Near Incinerator
Near Incinerator
Near Stockholm
Rural
PenUchlorodibenzo-p--dioxin5(MW=356.42}
1,2,3,7,8-PeCDD
PeCDDs
NR
8
4
19
77
1
2
77
47
1
6
5
NR
8
0
7
NR
1
2
NR
7
1
2
0
NR
2.6-18.3
ND
ND-11
<0.5-2.4
4.6
220-270
< 0.5-46
ND-580
240
ND-120
ND
15
10
NA
2
<0.5
NA
245
6.6
38.45
NA
37
NA
Finland
Sweden
Elk River, MN
England
British Isles
various parts of Europe
various parts of Europe
British Isles
Ontario and U.S. Midwestern
States.
Midland, MI
Midland, MI
Middletown, OH
Industrial
Urban
Rural
Urban
Background
Rural
Industrial
Background
Urban
Industrial
Residential
Residential
NR
NR
2
5
6
8
11
10
10
11
13
1
1
1
Near Stockholm
Agriculture
B-4
-------
Table B-l. Environmental Levels of Dioxins in Soils (ppt) (continued)
Chemical
PeCDDs (continued)
Number
samples
3
NR
11
12
53
29
g
4
12
19
Number
positive
samples
0
NR
NR
NR
0
NR
8
1
12
19
Concentration
range
(Ppt)
ND
NR
ND-580
ND-540
ND
ND-130
46-476
ND-38
4-50
6-190
Cone.
mean
NA
900
53
81
ND
23
159
10
20
69
Location
MN
Finland
Canada
Canada
Canada
Canada
Sweden
Elk River, MN
England
England
Location description
Pristine
Industrial
Rural
Urban
Urban
Rural
Residential
Urban
Sample
year
Ret
no.
1
2
4
4
4
4
5
6
7
8
Comments
Near Incinerator
Near Incinerator
Near Stockholm
Agriculture
Rural
Hexachlorodibenzo-p-dioxins (MW= 390 . 87)
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
HxCDDs
NR
8
4
NR
8
4
NR
8
4
1
2
77
47
20
1
7
NR
8
0
NR
8
1
NR
8
2
1
2
NR
13
8
1
5
NR
4.3-8.0
ND
NR
3.3-32.2
ND-14
NR
1.6-16.6
ND-9.9
4.7
200-330
2.8-165
ND-410
ND-240
4000
ND-410
<2
6
NA
2100
12
4
700
8
9
NA
265
38
38.12
44.1
NA
172
Finland
Sweden
Elk River, MN
Finland
Sweden
Elk River, MN
Finland
Sweden
Elk River, MN
Various parts of Europe
Various parts of Europe
British Isles
Ontario and U.S. Midwestern
States.
Canada and U.S.A.
Midland, MI
Midland, MI
Industrial
Urban
Rural
Industrial
Urban
Rural
Industrial
Urban
Rural
Rural
Industrial
Background
Urban
Industrial
Industrial
Residential
NR
2
5
6
2
5
6
2
5
6
10
10
11
13
13
1
1
Near Stockholm
Agriculture
Near Stockholm
AGriculture
Near Stockholm
Agriculture
B-5
-------
Table B-l. Environmental Levels of Dioxins in Soils (ppt) (continued)
Chemical
HxCDDs (continued)
Number
samples
5
3
NR
11
12
53
29
8
4
12
19
Number
positive
samples
1
0
NR
NA
NA
0
0
8
4
12
19
Concentration
range
(PPQ
ND-72
ND
NR
ND-170
ND-70
ND
ND
43-349
12-99
8-43
23-340
Cone.
mean
14
NA
7200
15
9
NA
NA
156
48
23
154
Location
Middletown, OH
MN
Finland
Canada
Canada
Canada
Canada
Sweden
Elk River, MN
England
England
Location description
Residential
Pristine
Industrial
Rural
Urban
Urban
Rural
Residential
Urban
Sample
year
Ref.
no.
1
1
2
4
4
4
4
5
6
7
8
Comments
Near Incinerator
Near Incinerator
Near Stockholm
Agriculture
Rural
Heptachlorodibenzo-p-dioxins (MW= 425.3 1 )
1,2,3,4,6,7,8-
HpCDD
1,2,3,4,6,7,9-
HpCDD
HpCDDs
NR
8
4
NR
3
2
77
30
47
20
1
7
5
3
11
12
53
29
8
NR
8
4
NR
1
2
NR
3
25
19
1
7
5
3
NR
NR
0
NR
8
NR
43-492
37-360
NR
< 10-17
370-1600
7.5-234
ND-91
ND-2400
ND-5000
75000
150-2400
23-200
25-91
ND-390
ND-300
ND
ND-1100
83-904
4700
144
194
7100
NA
985
66
5.4
211.8
1196.9
NA
930
113
54
90
43
NA
93
277
Finland
Sweden
Elk River, MN
Finland
Various parts of Europe
Various parts of Europe
British Isles
Ontario and U.S. Midwestern
States.
Ontario and U.S. Midwestern
States.
Canada and U.S.A.
Midland, MI
Midland, MI
Middletown, OH
MN
Canada
Canada
Canada
Canada
Sweden
Industrial
Urban
Rural
Industrial
Rural
Industrial
Background
Rural
Urban
Industrial
Industrial
Residential
Residential
Pristine
Rural
Urban
Urban
NR
2
5
6
2
10
10
11
13
13
13
1
1
1
1
4
4
4
4
5
Near Stockholm
Agriculture
Near Incinerator
Near Incinerator
Near Stockholm
R-fi
-------
Table B-l. Environmental Levels of Dioxins in Soils (ppt) (continued)
Chemical
HpCDDs (continued)
Number
samples
4
12
19
Number
positive
samples
4
12
19
Concentration
range
(PPO
62-640
20-130
77-5500
Cone,
mean
346
64
817
Location
Elk River, MN
England
England
Location description
Rural
Residential
Urban
Sample
year
Ref.
no.
6
7
8
Comments
Agriculture
Rural
Octachlorodibenzo-p-dioxin (MW=460.76)
OCDD
1
2
77
30
47
20
1
7
5
3
MR
11
12
53
29
8
4
12
19
1
2
MR
17
38
20
1
7
5
3
NR
NR
NR
NR
NR
8
4
12
19
14
140-160
29-832
44-810
ND-12000
15-26000
375000
330-12000
170-10600
25-91
NR
ND-3500
ND-1500
ND-100
ND-16000
113-2659
340-3300
20-150
176-99000
NA
160
191
67.33
1599.1
3442.25
NA
4473
2418
54
6200
663
570
31
2464
687
1655
58
9980
Various parts of Europe
Various parts of Europe
British Isles
Ontario and U.S. Midwestern
States.
Ontario and U.S. Midwestern
States.
Canada and U.S. A.
Midland, MI
Midland, MI
Middletown, OH
MN
Finland
Canada
Canada
Canada
Canada
Sweden
Elk River, MN
England
England
Rural
Industrial
Background
Rural
Urban
Industrial
Industrial
Residential
Residential
Pristine
Industrial
Rural
Urban
Urban
Rural
Residential
Urban
NR
10
10
11
13
13
13
1
1
1
1
2
4
4
4
4
5
6
7
8
Near Incinerator
Near Incinerator
Near Stockholm
Agriculture
Rural
Footnote References
NOTES: Summary statistics provided in or derived from references; means from references included non-detects counted as 0.0; when reference did not compute mean, it was computed with same strategy;
NA = Not applicable;
ND = Non-detect;
NR = Not reported.
Detection limits varied by study and as was different for different compounds, but generally were 1 to 5 ng/kg (ppt). Descriptions provided were those given by reference or surmised from study description
when not given.
B-7
-------
Table B-l. Environmental Levels of Dioxins in Soils (ppt) (continued)
Sources:
1. EPA (1985) 8. Creaser, et al. (1990)
2. Kitunen and Salkinoja-Salonen (1990). 10. Rappe and Kjeller (1987)
3. Nestrick, et al. (1986). 11. Creaser, et al. (1989)
4. Pearson, et al. (1990). 12. Sievers and Friesel (1989)
5. Broman, et al. (1990). 13. Birmingham (1990)
6. Reed, et al. (1990).
7. Stenhouse and Badsha (1990).
B-8
-------
Table B-2. Environmental Levels of Dibenzofurans in Soil (ppt)
Chemical
2,3,7,8-TCDF
2,3,4,8/2,3,7,8-TCDF
TCDFs
Number
samples
2
8
5
3
NR
4
12
8
3
2
77
47
20
1
1
NR
11
12
53
29
8
4
12
19
Number
positive
samples
2
3
2
0
NR
0
12
8
3
2
NR
13
3
0
0
NR
NR
0
1
NR
8
1
12
19
Cone, range
Cone.
mean
Location
Location
description
Sampley
ear
Ref.
no.
Comments
Tetrachlorodibenzofiirans (MW= 305 .98)
27-450
ND-15
ND-6
ND
NR
ND
3-50
8.4-57.5
7.7-11
320-370
<0.5-237
ND-120
ND-1850
ND
ND
NR
ND-71
ND
ND-280
ND-120
109-454
ND-1.2
20-300
29-950
238
5
2
NA
<2
NA
17
22
9.3
345
25
42.92
1016.66
NA
NA
300
10
NA
NA
15
237
<1
102
232
Midland, MI
Midland, MI
Middletown, OH
MN
Finland
Elk River, MN
England
Sweden
Various parts of
Europe
Various parts of
Europe
British Isles
Ontario and U.S.
Midwestern states
Ontario and U.S.
Midwestern states
Midland, MI
MN
Finland
Canada
Canada
Canada
Canada
Sweden
Elk River, MN
England
England
Industrial
Residential
Residential
Pristine
Industrial
Rural
Residential
Urban
Rural
Industrial
Background
Urban
Industrial
Residential
Pristine
Industrial
Rural
Urban
Uiban
Rural
Residential
Urban
1
1
1
1
2
6
7
5
10
10
11
13
13
1
1
2
4
4
4
4
5
6
7
8
Agriculture
Rural
Near Stockholm
Near Incinerator
Near Incinerator
Near Stockholm
Agriculture
Rural
B-9
-------
Table B-2. Environmental Levels of Dibenzofurans in Soil (ppt) (continued)
Chemical
Number
samples
Number
positive
samples
Cone, range
Cone.
mean
Location
Location
description
Sampley
ear
Ref.
no.
Comments
Pentachlorodibenzofiirans (MW=340.42)
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1 ,2,3,7,8/1 ,2,3 ,4,8-PeCDF
PeCDFs
4
12
NR
8
4
12
NR
8
3
77
47
20
2
1
8
5
NR
11
12
53
29
8
4
12
0
12
NR
8
0
12
NR
8
3
NR
4
5
2
1
2
0
NR
0
0
0
NR
8
3
12
ND
1-10
NR
3.1-26.5
ND
1-5
NR
7.4-32.1
6.7-14
<0.5-185
ND-110
ND-285
200-450
900
ND-110
ND
NR
ND
ND
ND
ND-160
36-457
18-45
6-70
NA
4
580
13
NA
2
82
16
11.23
23
41.5
86
325
NA
19
NA
27000
NA
NA
NA
35
149
35
31
Elk River, MN
England
Finland
Sweden
Elk River, MN
England
Finland
Sweden
Various parts of
Europe
British Isles
Ontario and U.S.
Midwestern states
Ontario and U.S.
Midwestern states
Various parts of
Europe
Midland, MI
Midland, MI
Middletown, OH
Finland
Canada
Canada
Canada
Canada
Sweden
Elk River, MN
England
Rural
Residential
Industrial
Urban
Rural
Residential
Industrial
Urban
Rural
Background
Urban
Industrial
Industrial
Industrial
Residential
Residential
Industrial
Rural
Urban
Urban
Rural
Residential
6
7
2
5
6
7
2
5
10
11
13
13
10
1
1
1
2
4
4
4
4
5
6
7
Agriculture
Rural
Near Stockholm
Agriculture
Rural
Near Stockholm
Near Incinerator
Near Incinerator
Near Stockholm
Agriculture
Rural
B-10
-------
Table B-2. Environmental Levels of Dibenzofurans in Soil (ppt) (continued)
Chemical
PeCDFs (continued)
Number
samples
19
Number
positive
samples
19
Cone, range
19-830
Cone.
mean
189
Location
England
Location
description
Urban
Sampley
ear
Ref.
no.
8
Comments
Hexachlorodibenzofurans (MW= 374.87)
1,2,3,4,7,8/1,2,3,4,7,9-
HxCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
HxCDF
NR
8
4
NR
8
4
NR
8
4
NR
4
3
77
47
20
2
1
8
5
3
NR
11
12
NR
8
0
NR
8
0
NR
8
0
NR
1
3
NR
6
7
2
1
3
0
0
NR
0
0
NR
7.8-29.1
ND
NR
7.7-28.9
ND
NR
0.5-3.8
ND
NR
ND-7.1
11-16
4.3-212
ND-260
ND-420
270-1900
15400
64-260
ND
ND
NR
ND
ND
920
16
NA
<2
14
NA
<2
1
NA
<2
2
13
41
94.83
177.85
1085
NA
62
NA
NA
7200
NA
NA
Finland
Sweden
Elk River, MN
Finland
Sweden
Elk River, MN
Finland
Sweden
Elk River, MN
Finland
Elk River, MN
Various parts of
Europe
British Isles
Ontario and U.S.
Midwestern states
Ontario and U.S.
Midwestern states
Various parts of
Europe
Midland, MI
Midland, MI
Middletown, OH
MN
Finland
Canada
Canada
Industrial
Urban
Rural
Industrial
Urban
Rural
Industrial
Urban
Rural
Industrial
Rural
Rural
Background
Urban
Industrial
Industrial
Industrial
Residential
Residential
Pristine
Industrial
2
5
6
2
5
6
2
5
6
2
6
10
11
13
13
10
1
1
1
1
2
4
4
Near Stockholm
Agriculture
Near Stockholm
Agriculture
Near Stockholm
Agriculture
Agriculture
Near Incinerator
Near Incinerator
B-ll
-------
Table B-2. Environmental Levels of Dibenzofurans in Soil (ppt) (continued)
Chemical
HxCDF (continued)
Number
samples
53
29
8
4
12
19
Number
positive
samples
0
NR
8
4
12
19
Cone, range
ND
ND-120
53-308
7-150
6-50
17-660
Cone.
mean
NA
9
145
66
24
156
Location
Canada
Canada
Sweden
Elk River, MN
England
England
Location
description
Rural
Urban
Urban
Rural
Residential
Urban
Sampley
ear
Ref,
no.
4
4
5
6
7
8
Comments
Near Stockholm
Agriculture
Rural
Heptachlorodibenzoftirans(MW=409.3 1)
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
HpCDFs
NR
8
4
NR
8
4
3
2
77
47
20
1
8
5
3
11
12
53
NR
8
4
NR
8
0
3
2
NR
10
15
1
6
1
0
NR
0
0
NR
31-134
11-80
NR
1.0-6.3
ND
14-22
260-4500
1.5-138
ND-120
ND-3750
75000
ND-820
ND-43
ND
ND-180
ND
ND
4700
73
47
<5
3
NA
18
2380
26
283.1
732.66
NA
300
9
NA
30
NA
NA
Finland
Sweden
Elk River, MN
Finland
Sweden
Elk River, MN
Various parts of
Europe
Various parts of
Europe
British Isles
Ontario and U.S.
Midwestern slates
Ontario and U.S.
Midwestern states
Midland, MI
Midland, MI
Middletown, OH
MN
Canada
Canada
Canada
Industrial
Urban
Rural
Industrial
Urban
Rural
Rural
Industrial
Background
Urban
Industrial
Industrial
Residential
Residential
Pristine
Rural
2
5
6
2
5
6
10
10
11
13
13
1
1
1
1
4
4
4
Near Stockholm
Agriculture
Near Stockholm
Agriculture
Near Incinerator
Near Incinerator
B-12
-------
Table B-2. Environmental Levels of Dibenzofurans in Soil (ppt) (continued)
Chemical
HpCDFs (continued)
Number
samples
29
8
4
12
19
Number
positive
samples
NR
8
4
12
19
Cone, range
ND-410
31-187
30-260
4-59
16-»58
Cone.
mean
29
81
100
20
152
Location
Canada
Sweden
Elk River, MN
England
England
Location
description
Urban
Urban
Rural
Residential
Urban
Sampley
ear
Ref.
no.
4
5
6
7
8
Comments
Near Stockholm
Agriculture
Rural
Octachlorodibenzofiirans(MW!= 444.76)
OCDF
1
77
47
20
2
1
8
5
3
NR
11
12
53
29
8
4
12
19
1
NR
15
15
2
1
6
1
0
NR
NR
NR
0
NR
8
3
12
19
5.7
<2.0-144
ND-
ND-5200
68-71
8600
ND-660
ND-50
ND
NR
ND-33
ND-230
ND
ND-600
2.9-19.0
60-270
10-90
7-1100
NA
27
184.8
842.53
69.5
NA
240
10
NA
3000
4
43
NA
50
8
113
30
196
Various parts of
Europe
British Isles
Canada and USA
Canada and USA
Various parts of
Europe
Midland, MI
Midland, MI
Middletown, OH
MN
Finland
Canada
Canada
Canada
Canada
Sweden
Elk River, MN
England
England
Rural
Background
Urban
Industrial
Industrial
Industrial
Residential
Residential
Pristine
Industrial
Rural
Urban
Urban
Rural
Residential
Urban
10
11
13
13
10
1
1
1
1
2
4
4
4
4
5
6
7
8
Near Incinerator
Near Incinerator
Near Stockholm
Agriculture
Rural
B-13
-------
Table B-2. Environmental Levels of Dibenzorurans in Soil (ppt) (continued)
Footnote References
NOTES: Summary statistics provided in or derived from references; means from references included non-detects counted as 0.0; when reference did not compute mean, it was computed with same
strategy;
NA = not available;
MR = not reported;
ND = Non-detect.
Detection limits varied by study and as was different for different compounds, but generally were 1 to 5 ng/kg (ppt). Descriptions provided were those given by reference or surmised from study
description when not given.
Sources: 1. EPA (1985)
2. Kitunen and Salkinoja-Salonen (1990.
3. Nestrick, etal. (1986).
4. Pearson, et al. (1990).
5. Broman, et al. (1990).
6. Reed, et al. (1990).
7. Stenhouse and Badsha (1990).
8. Greaser, et al. (1990)
10. Rappe and Kjeller (1987)
ll.Creaser, etal. (1989)
12. Sievers and Friesel (1989)
13. Birmingham (1990)
B-14
-------
Table B-3. Environmental Levels of Dioxins in Water (ppq)
Chemical
Number
samples
Number
positive
samples
Cone, range
Cone.
mean
Location
Location
description
Samp,
year
Ref.
:no.
Comments
Tetracnlorodibenzo-p-dioxins(MW=321.98)
2,3,7,8-TCDD
TCDDs
1
2
185
22
1
2
0
0
1
0
1
2
ND(0.7)
ND(.02-.024)
NEMO
ND(.4-2.6)
1.7
.05-.084
NA
NA
0.23
NA
1.7
.067
Lockport, New York
Eman river, Sweden
Ontario, Canada
New York State
Lockport, New York
Eman river, Sweden
NR
PCB contaminated
NR
NR
NR
PCB contaminated
88
NR
83-89
86-88
88
NR
2
3
1
2
2
3
raw surface drinking water
raw surface drinking water
raw surface drinking water
treated surface drinking water
raw surface drinking water
raw surface drinking water
Pentachlorodibenzo-p-dioxins(MW=356.42)
1,2,3,7,8-PeCDD
PeCDDs
1
2
1
22
2
0
0
0
0
2
ND(l.O)
ND(.025-.039)
ND(l.O)
ND(1. 2-7.4)
.067-.12
NA
NA
NA
NA
.094
Lockport, New York
Eman river, Sweden
Lockport, New York
New York State
Eman river, Sweden
NR
PCB contaminated
NR
NR
PCB contaminated
88
NR
88
86-88
NR
2
3
2
2
3
raw surface drinking water
raw surface drinking water
raw surface drinking water
treated surface drinking water
raw surface drinking water
Hexachlorodibenzo-p-dioxins(MW =390.87)
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
HxCDDs
1
2
1
2
1
2
1
22
0
1
0
1
0
1
0
0
ND(1.8)
N0-.054
ND(1.5)
ND-.12
ND(1.5)
ND-.075
ND(1.5)
ND(.4-4.7)
NA
.027
NA
.06
NA
.038
NA
NA
Lockport, New York
Eman river, Sweden
Lockport, New York
Eman river. Sweden
Lockport, New York
Eman river, Sweden
Lockport, New York
New York State
NR
PCB contaminated
NR
PCB contaminated
NR
PCB contaminated
NR
NR
88
NR
88
NR
88
NR
88
86-88
2
3
2
3
2
3
2
2
raw surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
treated surface drinking water
B-15
-------
Table B-3. Environmental Levels of Dioxins in Water (ppq) (continued)
Chemical
HxCDDs (continued)
1,2,3,4,6,7,8-HpCDD
HpCDDs
1,2,3,4,6,7,8,9-OCDD
Number
samples
2
1
2
1
22
2
Number
positive
samples
2
Cone, range
.13-.67
Cone,
mean
.4
Location
Eman river, Sweden
Location
description
PCB contaminated
Samp.
year
NR
Ref.
no.
3
Comments
raw surface drinking water
Heptachlorodibenzo-p-dioxins(MW = 425 .3 1 )
0
2
0
0
2
ND(2.8)
.15-.30
ND(2.8)
ND(.4-6.8)
.17-.64
NA
.22
NA
NA
.40
Lockport, New York
Eman river, Sweden
Lockport, New York
New York State
Eman river, Sweden
MR
PCB contaminated
NR
NR
PCB contaminated
88
NR
88
86-88
NR
2
3
2
2
3
raw surface drinking water
raw surface drinking water
raw surface drinking water
treated surface drinking water
raw surface drinking water
OcUchlorodibenzo-p-dioxin(MW== 460.76)
185
214
32
4
ND-175
ND-46
8.45
0.63
Ontario, Canada
Ontario, Canada
NR
NR
83-89
83-89
1
1
raw surface drinking water
treated surface drinking water
Footnote References
NOTES: Summary statistics provided in or derived from references; means from references included non-detects counted as 0.0; when reference did
NA = not applicable;
ND = non-detected (limit of detection);
NR = not reported.
Descriptions provided were those given by reference or surmised from study description when not given.
Sources: 1. Jobb, et al. (1990)
2. Meyer, et al. (1989)
3. Rappe, et al. (1989b)
not compute mean, it was computed with same strategy;
B-16
-------
Table B-4. Environmental Levels of Dibenzofurans in Water (ppq)
Chemical
Number
samples
Number
positive
samples
2,3,7,8-TCDF
TCDFs
22
2
1
22
1
2
1
2
0
1
1
2
Cone, range
Cone,
mean
Location
Location
description
Samp.
year
Ref.
no.
Comments :
Tetrachlorodibenzofiirans(MW=305*98)
ND-1.2
.022-.026
ND(0.7)
ND-2.6
18
.21-.23
0.05
.024
NA
0.12
NA
0.22
New York State
Eman river, Sweden
Lockport, New York
New York State
Lockport, New York
Eman river, Sweden
NR
PCB contaminated
NR
NR
NR
PCB contaminated
86-88
NR
88
86-88
88
NR
2
3
2
2
2
3
treated surface drinking water
raw surface drinking water
raw surface drinking water
treated surface drinking water
raw surface drinking water
raw surface drinking water
Pentachlorodibenzo(urans(MW=340.42>
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,8/1,2,3,7,8-PeCDF
PeCDFs
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1
2
1
2
1
22
2
1
2
0
2
1
0
2
1
1
2
1
2
1
2
1
1
2
0
1
0
1
2.0
.014- .019
ND(l.O)
.013-.025
27
ND(0.3-4.0)
.13-.21
NA
.016
NA
.019
NA
NA
.17
Lockport, New York
Eman river, Sweden
Lockport, New York
Eman river, Sweden
Lockport, New York
New York State
Eman river, Sweden
NR
PCB contaminated
NR
PCB contaminated
NR
NR
PCB contaminated
88
NR
88
NR
88
86-88
NR
2
3
2
3
2
2
3
Hexachlorod ibenzofiirans (MW=3 74 . 87)
39
9.2
.019-.025
ND(1.3)
ND-.027
ND(1.2)
ND-.022
NA
NA
.022
NA
.014
NA
.011
Lockport, New York
Lockport, New York
Eman river, Sweden
Lockport, New York
Eman river, Sweden
Lockport, New York
Eman river, Sweden
NR
NR
PCB contaminated
NR
PCB contaminated
NR
PCB contaminated
88
88
NR
88
NR
88
NR
2
2
3
2
3
2
3
raw surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
treated surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
B-17
-------
Table B-4. Environmental Levels of Dibenzofurans in Water (ppq) (continued)
Chemical
1 ,2.3,4,7.9/1 ,2.3,4,7.8-HxCDF
HxCDFs
1,2,3.4,6,7,8-HpCDF
1,2,3,4,7.8,9-HpCDF
HpCDFs
1,2,3,4,6,7,8,9-OCDF
Number
samples
2
1
22
2
Number
positive
samples
2
1
0
2
Cone, range
.021-.026
85
ND(.3-4.4)
.17-.19
Cone.
mean
.024
NA
NA
.18
Location
Eman river. Sweden
Lockport, New York
New York State
Eman river, Sweden
Location
description
PCB contaminated
NR
NR
PCB contaminated
Samp.
year
NR
88
86-88
NR
Ref.
no.
3
2
2
3
Heptachlorodibenzofurans (MW= 409 .3 1 )
1
2
2
1
22
2
1
2
2
1
0
2
210
.083-. 13
.03-. 058
210
ND(.8-6.6)
.18- .35
NA
.11
.044
NA
NA
.26
Lockport, New York
Eman river, Sweden
Eman river, Sweden
Lockport, New York
New York State
Eman river, Sweden
NR
PCB contaminated
PCB contaminated
NR
NR
PCB contaminated
88
NR
NR
88
86-88
NR
2
3
3
2
2
3
Comments
raw surface drinking water
raw surface drinking water
treated surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
raw surface drinking water
treated surface drinking water
raw surface drinking water
Octachlorodibenzofurans (MW=444.76)
22
1
2
2
1
2
ND-0.8
230
.15-.36
.07
NA
.26
New York State
Lockport, New York
Eman river, Sweden
NR
NR
PCB contaminated
86-88
88
NR
2
2
3
treated surface drinking water
raw surface drinking water
raw surface drinking water
Footnote References
NOTES: Summary statistics provided in or derived from references; means from references included non-detects counted as 0.0; when reference did not compute mean, it was computed with same strategy;
NA = not applicable;
ND = non-detected (limit of detection);
NR = not reported.
Descriptions provided were those given by reference or surmised from study description when not given.
Sources: 2. Meyer, et al. (1989)
3. Rappe, et al. (1989b)
B-18
-------
Table B-5. Environmental Levels of Dioxins in Sediments (ppt)
Chemical
Number
samples
Number
positive
samples
Concentration
range
Cone.
mean
Wt.
basis
Location
Location
description
Sample
year
Ref.
no.
Comments*
Tetrachlorodibenzo-p-dioxins (MW= 321 .98)
2,3,7,8-TCDD
18
9
4
2
1
1
2
3
3
1
2
1
1
12
4
4
2
3
4
4
6
0
8
4
2
1
1
2
1
3
1
1
1
0
6
0
0
2
2
4
1
6
N/A
ND-730
75-2500
190-1200
N/A
N/A
660-1100
ND-7600
390-2900
N/A
ND-7600
N/A
N/A
ND-57
N/A
N/A
1.0-1.4
ND-2.4
1.9-26
ND(.5-2)-3
1.2-32
N/A
236
1769
695
680
150
880
367
1227
93
3800
21000
ND
13
ND
ND
1.2
1.5
9.6
1.19
12.05
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
NR
Dry
NR
Dry
NR
NR
South Central Finland
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Long Island Sound
New England
Seattle, WA
Central Minnesota
Baltic Sea
Stockholm Sweden
Iggesund Sweden
Dala River, Sweden
Lake Vattern, Sweden
Various
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Reference Site
Industrial
Industrial
Rural
Reference Site
Various
Industrial
Industrial
Industrial
88/89
85/86
85/86
85/86
85/86
85/86
85/86
85/86
85/86
85/86
85/86
85/86
NR
NR
NR
NR
NR
NR
NR
88
88
1
2
2
2
2
2
2
2
2
2
2
2
4
4
4
5
13
12
13
17
17
AB
0-2% AC
2-4", AC
4-8", AC
12-16", AC
20-24", AC
24-28", AC
28-32", AC
32-36", AC
40-44", AC
48-52", AC
108-111", AC
A
A/3 sites
A
A
A
A
A/papermill
0-1 cm
0-1 cm
B-19
-------
Table B-5. Environmental Levels of Uioxins in Sediments (ppt) (continued)
Chemical
2,3,7,8-TCDD (continued)
TCDDs
Number
samples
5
4
1
1
5
18
1
1
1
4
25
12
3
4
2
4
1
Number
positive
samples
3
4
1
1
5
14
1
1
0
0
0
7
3
4
2
4
1
Concentration
range
ND(4-11)-110
0.03-0.11
N/A
N/A
3.4-1,500
ND-1,400
N/A
N/A
N/A
N/A
N/A
ND-44
21-69
21-66
19-35
1.4-6.7
N/A
Cone.
mean
28.08
0.06
0.04
0.01
375
372
26
12
ND
ND
ND
17
38
45
27
4.2
13
Wt.
basis
NR
Dry
Dry
Dry
NR
Dry
Dry
Dry
Dry
NR
NR
NR
NR
Dry
Dry
Dry
Dry
Location
Lake Vanem, Sweden
Fjord between Denmark,
Sweden, and Norway
Fjord between Denmark,
Sweden, and Norway
Fjord between Demark,
Sweden, and Norway
Hamburg Germany
South Central Finland
Siskiwit Lake, Isle Royale,
Lake Michigan
Siskiwit Lake, Isle Royale,
Lake Michigan
Siskiwit Lake, Isle Royale,
Lake Michigan
Central Minnesota
Ontario Canada
NY/Mass
Stockholm Sweden
Iggesund Sweden
Baltic Sea
Fjord between Denmark,
Sweden, and Norway
Fjord between Denmark,
Sweden, and Norway
Location
description
Industrial
Industrial
Industrial
Industrial
Urban
Various
Pristine
Pristine
Pristine
Rural
Industrial
NR
Various
Industrial
Reference Site
Industrial
Industrial
Sample
year
88
87
87
87
NR
88/89
NR
NR
NR
NR
88
NR
NR
NR
NR
87
87
Ref.
no.
17
14
14
14
15
1
3
3
3
5
8
10
12
13
13
14
14
Comments*
0-1 cm
0-1 cm, E
4-6 cm, E
9-13 cm, E
AB
0-0.5 cm, AD
5-6 cm, AD
8-9 cm, AD
A/4 sites
A/25 sites
A
A
A/paper mill
A
0-2 cm, E
4-6 cm, E
B-20
-------
Table B-5. Environmental Levels of Dioxins in Sediments (ppt) (continued)
Chemical
TCDDs
(continued)
Number
samples
1
5
Number
positive
samples
1
5
Concentration
range
N/A
80-1,700
Cone.
mean
5.0
564
Wt.
basis
Dry
NR
Location
Fjord between Denmark,
Sweden, and Norway
Hamburg Germany
Location
description
Industrial
Urban
Sample
year
87
NR
Ref.
no.
14
15
Comment^
9-13 cm, E
Pentachlorodibenzo-p-dioxins(MW=356 .42)
1,2,3,7,8-PeCDD
PeCDDs
4
6
5
1
1
1
4
25
12
3
4
2
4
1
1
3
6
4
1
1
0
0
0
7
3
4
2
4
1
1
ND(2)-25
7.4-95
ND(14)-100
N/A
N/A
N/A
N/A
N/A
ND-235
86-230
52-500
52-100
1.7-41
N/A
N/A
7.68
44.1
49.6
12
11
ND
ND
ND
50
138
209
76
19
6.6
15
NR
NR
NR
Dry
Dry
Dry
NR
NR
NR
NR
Dry
Dry
Dry
Do-
Dry
Data River, Sweden
Lake Vattern, Sweden
Lake Vanern, Sweden
Siskiwit Lake, Isle Royale,
Lake Michigan
Siskiwit Lake, Isle Royale,
Lake Michigan
Siskiwit Lake, Isle Royale,
Lake Michigan
Central Minnesota
Ontario Canada
NY/Mass
Stockholm Sweden
Iggesund Sweden
Baltic Sea
Fjord between Denmark,
Sweden, and Norway
Fjord between Denmark,
Sweden, and Norway
Fjord between Denmark,
Sweden, and Norway
Industrial
Industrial
Industrial
Pristine
Pristine
Pristine
Rural
Industrial
NR
Various
Industrial
Reference Site
Industrial
Industrial
Industrial
88
88
88
NR
NR
NR
NR
88
NR
NR
NR
NR
87
87
87
17
17
17
3
3
3
5
8
10
12
13
13
14
14
14
0-1 cm
0-1 cm
0-1 cm
0-0.5 cm, AD
5-6 cm, AD
8-9 cm, AD
A/4 sites
A/25 sites
A
A
A/paper mill
A
0-2 cm, E
4-6 cm, E
9-13 cm, E
B-21
-------
Table B-5. Environmental Levels of Dioxins in Sediments (ppt) (continued)
Chemical
PeCDDs
(continued)
Number
samples
5
Number
positive
samples
5
Concentration
range
260-2,700
Cone.
mean
1,112
Wt.
basis
MR
Location
Hamburg Germany
Location
description
Urban
Sample
year
NR
Ref.
no.
15
Comment^
Hexachlorodibenzo-p-dioxins(MW=390.87)
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
HxCDDs
4
6
5
4
6
5
4
6
5
1
1
1
4
25
12
3
2
4
3
6
5
4
6
5
4
6
5
1
1
0
2
6
10
3
2
4
ND(6)-19
6.1-33
14-94
4.9-120
21-450
36-600
3.9-51
18-200
18-330
N/A
N/A
N/A
ND-14
ND-5,700
ND-1,335
16-49
120-170
130-1,900
6.68
20.5
36.0
36.38
188
236
18.1
95.5
149
10
8
ND
5.2
1,157
399
28
145
608
NR
MR
NR
NR
NR
NR
NR
NR
NR
Do-
Dry
Dry
NR
NR
NR
NR
Dry
Dry
Dala River, Sweden
Lake Vattern, Sweden
Lake Vanern, Sweden
Dala River, Sweden
Lake Vattern, Sweden
Lake Vanem, Sweden
Dala River, Sweden
Lake Vattern, Sweden
Lake Vanem, Sweden
Siskiwit Lake, Isle Royale,
Lake Michigan
Siskiwit Lake, Isle Royale,
Lake Michigan
Siskiwit Lake, Isle Royale,
Lake Michigan
Central Minnesota
Ontario Canada
NY/Mass
Stockholm Sweden
Baltic Sea
Iggesund Sweden
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Pristine
Pristine
Pristine
Rural
Industrial
NR
Various
Reference Site
Industrial
88
88
88
88
88
88
88
88
88
NR
NR
NR
NR
88
NR
NR
NR
NR
17
17
17
17
17
17
17
17
17
3
3
3
5
8
10
12
13
13
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-0.5 cm, AD
5-6 cm, AD
8-9 cm, AD
A/4 sites
A/25 sites
A
A
A
A/paper mill
B-22
-------
Table B-5. Environmental Levels of Dioxins in Sediments (ppt) (continued)
Chemical
HxCDDs
(continued)
Number
samples
4
1
1
5
Number
positive
samples
4
1
1
5
Concentration
range
2.3-27.0
N/A
N/A
580-7,500
Cone.
mean
14
16
14
2,744
Wt.
basis
Do-
Dry
Dry
NR
Location
Fjord between Denmark,
Sweden, and Norway
Fjord between Denmark,
Sweden, and Norway
Fjord between Denmark,
Sweden, and Norway
Hamburg Germany
Location
description
Industrial
Industrial
Industrial
Urban
Sample
year
87
87
87
NR
Ref.
no.
14
14
14
15
Comments*
0-2 cm, E
4-6 cm, E
9-13 cm, E
Heptachlorodibenzo-p-dioxins (MW = 425 .31)
HpCDDs
4
25
12
3
2
4
4
1
1
5
4
20
11
3
2
4
4
1
1
5
7.3-110
ND-320,000
ND-18,950
880-5,700
79-210
90-340
2.2-19
N/A
N/A
1,300-8,600
71
51,680
4,168
2,233
145
190
12.4
7.2
7.1
4,040
MR
NR
NR
NR
Dry
Dry
Dry
Dry
Dry
NR
Central Minnesota
Ontario Canada
NY/Mass
Stockholm Sweden
Baltic Sea
Iggesund Sweden
Fjord between Denmark,
Sweden, and Norway
Fjord between Denmark,
Sweden, and Norway
Fjord between Denmark,
Sweden, and Norway
Hamburg Germany
Rural
Industrial
NR
Various
Reference Site
Industrial
Industrial
Industrial
Industrial
Urban
NR
88
NR
NR
NR
NR
87
87
87
NR
5
8
10
12
13
13
14
14
14
15
A/4 sites
A/25 sites
A
A
A
A/paper mill
0-2 cm, E
4-6 cm, E
9-13 cm, E
Octachlorodibenzo-p-d5oxin(MW=460.76) ;
OCDD
18
9
4
3
9
4
ND-42
3,100-14,000
5,300-23,000
6.1
8,100
14,100
Dry
Dry
Dry
South Central Finland
Newark, NJ
Newark, NJ
Various
Industrial
Industrial
88/89
85/86
85/86
1
2
2
AB
0-2", AC
2-4% AC
B-23
-------
Table B-5. Environmental Levels of Dioxins in Sediments (ppt) (continued)
Chemical
OCDD (continued)
Number
samples
2
1
1
2
2
3
1
2
1
1
1
1
4
25
12
7
3
2
4
Number
positive
samples
2
1
1
2
2
3
1
2
1
1
1
1
4
20
12
7
3
2
4
Concentration
range
10,000-31,000
N/A
N/A
11,000-19,000
5,500-22,000
5,600-42,000
N/A
4,400-24,000
N/A
N/A
N/A
N/A
450-600
ND-980,000
1,990-15,500
12-250
260-3,100
89-250
96-330
Cone.
mean
20,500
7,500
5,900
15,000
13,800
17,800
5,500
14,200
38,000
560
390
54
518
141,420
8,201
145
1,290
170
194
Wt.
basis
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Do-
Dry
Dry
Dry
Dry
NR
NR
NR
Dry
NR
Dry
Dry
Location
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Siskiwit Lake, Isle Royale,
Lake Superior
Siskiwit Lake, Isle Royale,
Lake Superior
Siskiwit Lake, Isle Royale,
Lake Superior
Central Minnesota
Ontario Canada
NY/Mass
Jackfish Bay, Lake
Superior
Stockholm Sweden
Baltic Sea
Iggesund Sweden
Location
description
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Pristine
Pristine
Pristine
Rural
Industrial
NR
Various
Various
Reference Site
Industrial
Sample
year
85/86
85/86
85/86
85/86
85/86
85/86
85/86
85/86
85/86
NR
NR
NR
NR
88
NR
NR
NR
NR
NR
Ref.
no.
2
2
2
2
2
2
2
2
2
3
3
3
5
8
10
11
12
13
13
Comments*
4-8", AC
12-16', AC
20-24", AC
24-28", AC
28-32", AC
32-36", AC
40-44", AC
48-52", AC
108-1 11", AC
0-0.5 cm, AD
5-6 cm, AD
8-9 cm, AD
A/4 sites
A/25 sites
A
Papermill,
atmospheric
contamination
A
A
A/paper mil!
B-24
-------
Table B-5. Environmental Levels of Dioxins in Sediments (ppt) (continued)
Chemical
OCDD (continued)
Number
samples
4
6
5
4
1
1
5
Number
positive
samples
4
6
5
4
1
1
5
Concentration
range
180-16,830
800-2200
1320-6090
3.6-45
N/A
N/A
2,800-15,000
Cone.
mean
5195
1775
3040
24
10
6.9
7,560
Wt,
basis
NR
NR
NR
Dry
Dry
Dry
NR
Location
Dala River, Sweden
Lake Vattern, Sweden
Lake Vanern, Sweden
Fjord between Denmark,
Sweden, and Norway
Fjord between Denmark,
Sweden, and Norway
Fjord between Denmark,
Sweden, and Norway
Hamburg Germany
Location
description
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Urban
Sample
year
88
88
88
87
87
87
NR
Ref.
no.
17
17
17
14
14
14
15
Comments1
0-1 cm
0-1 cm
0-1 cm
0-2 cm, E
4-6 cm, E
9-13 cm, E
•Key A LOD (Ref 1) = 20 to 50 ppt, LOD (Ref 2) = 22 to 60 ppt, LOD (Ref 3) - 0.4 ppt, LOD (Ref 4)
= 3 ppt, LOD (Ref 12) = 1 to 20 ppt, LOD (Ref 13) = 1 to 6.6 ppt.
B Dry surface sediments from 18 lakes.
C Industry produced 2,4,5-Trichlorophenate (2,4,5-T precursor).
D No anthropogenic inputs into drainage basin—only atmospheric sources into lake.
E Mg-Production Facility.
0.7 to 12.0 ppt, LOD (Ref 5) = 0.61 to 4.1 ppt, LOD (Ref 8) = 10 to 500 ppt, LOD (Ref 10)
B-25
-------
Notes
NR = Not Reported
N/A = Not Applicable
ND = Not Detected
ppt = Parts per trillion
Table B-5. Environmental Levels of Dioxins in Sediments (ppt) (continued)
Sources:
1. Koistinen et al. (1990)
2. Bopp et al. (1991)
3. Czuczwa et al. (1984)
4. Norwood et al. (1989)
5. Reed et al. (1990)
6. Sonzongiet al. (1991)
7. Oliver and Nilmi (1988)
8. McKee et al. (1990)
9. Huckins et al. (1988)
10. Petty et al. (1982)
11. Sherman et al. (1990)
12. Rappe and Kjeller (1987)
13. Rappe et al. (1989a)
14. Oehme et al. (1989)
15. Gotz and Schumacher (1990)
16. Smith et al. (1990)
17. Kjeller et al. (1990)
B-26
-------
Table B-6. Environmental Levels of Dibenzofurans in Sediment (ppt)
Chemical
Number
samples
Number
positive
samples
Concentration
range
Cone.
mean
Weight
basis
Location
Location description
Sample
year
Ref.
no.
Comments'
Tetrachlorodibenzofurans(MW=305,98)
2,3,7,8-TCDF
IS
9
4
2
1
1
2
3
3
1
2
1
1
12
4
4
2
4
4
6
5
0
9
4
2
1
1
2
2
3
1
2
1
1
12
0
1
2
4
4
6
5
N/A
24-490
150-1,400
580-1,200
N/A
N/A
300-390
N/D-530
190-730
140
80-3,100
4,500
N/A
8.8-1,400
N/A
ND-0.31
8.3-14
11-210
4.9-170
41-320
54-810
ND
232
855
890
370
300
345
243
370
140
1,590
4,500
15
566
ND
0.08
11
72
46.5
175.5
241.2
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
NR
Dry
Dry
NR
NR
NR
South Central Finland
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Newark, NJ
Long Island Sound
New England
Seattle, WA
Central Minnesota
Baltic Sea
Iggesund Sweden
Dala River, Sweden
Lake Vattern, Sweden
Lake Vanern, Sweden
Various
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Reference Site
Industrial
Industrial
Rural
Reference Site
Industrial
Industrial
Industrial
Industrial
88/89
85/86
85/86
85/86
85/86
85/86
85/86
85/86
85/86
85/86
85/86
85/86
NR
NR
NR
NR
NR
NR
88
88
88
1
2
2
2
2
2
2
2
2
2
2
2
4
4
4
5
13
13
17
17
17
AB
0-2", AC
2-4", AC
4-8", AC
12-16", AC
20-24", AC
24-28", AC
28-32", AC
32-36", AC
40-44", AC
48-52", AC
108-1 11", AC
A
A/3 sites
A
A/4 sites
A
A/papermSll
0-1 cm
0-1 cm
0-1 cm
B-27
-------
Table B-6. Environmental Levels of Dibenzofurans in Sediment (ppt) (continued)
Chemical
2,3,7,8-TCDF (continued)
TCDFs
Number
samples
4
1
1
1
1
1
4
25
12
7
3
2
4
4
1
1
Number
positive
samples
4
1
1
1
1
0
2
0
11
7
3
2
4
4
1
1
Concentration
range
0.7-9.6
N/A
N/A
N/A
N/A
N/A
ND-0.54
N/A
ND-200
2.4-6,223
120-290
87-130
79-360
6.7-54
N/A
N/A
Cone.
mean
5.2
13
5.2
15
18
ND
0.21
ND
58
1,260
187
109
180
30
63
23
Weight
basis
Dry
Dry
Dry
Dry
Dry
Dry
NR
NR
NR
Dry
NR
Dry
Dry
Dry
Dry
Dry
Location
Fjord between Sweden,
Norway and Denmark
Fjord between Sweden,
Norway and Denmark
Fjord between Sweden,
Norway and Denmark
Siskiwit Lake, Isle
Royale, Lake Superior
Siskiwit Lake, Isle
Royale, Lake Superior
Siskiwit Lake, Isle
Royale, Lake Superior
Central Minnesota
Ontario Canada
NY/MASS
Jackfish Bay, Lake
Superior
Stockholm Sweden
Baltic Sea
Iggesund Sweden
Fjord between Sweden,
Norway and Denmark
Fjord between Sweden,
Norway and Denmark
Fjord between Sweden,
Norway and Denmark
Location description
Industrial
Industrial
Industrial
Pristine
Pristine
Pristine
Rural
Industrial
NR
Various
Various
Reference Site
Industrial
Industrial
Industrial
Industrial
Sample
year
87
87
87
NR
NR
NR
NR
88
NR
NR
NR
NR
NR
87
87
87
Ref.
no.
14
14
14
3
3
3
5
8
10
11
12
13
13
14
14
14
Comments*
0-2cm, E
4-6cm, E
9-13cm, E
0-0.5cm, AD
5-6cm, AD
8-9cm, AD
A/4 sites
A/25 sites
A
Papermill/
atmospheric
contamination
A
A
A/papermill
0-2cm, E
4-6cm, E
9-13cm, E
B-28
-------
Table B-6. Environmental Levels of Dibenzofurans in Sediment (ppt) (continued)
Chemical
TCDFs
(continued)
Number
samples
5
Number
positive
samples
5
Concentration
range
170-1070
Cone,
mean
526
Weight
basis
NR
Location
Hamburg Germany
Location description
Urban
Sample
year
NR
Ref.
no.
15
Comments*
Pentachlorodibenzofurans(MW=340.42) :
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
PeCDFs
4
6
5
4
6
5
1
1
1
4
25
12
3
2
4
4
1
4
6
5
4
6
5
1
1
0
2
0
9
3
2
4
4
1
5.6-110
27-120
50-300
7.8-99
25-110
36-250
N/A
N/A
N/A
ND-25
N/A
ND-193
130-260
66-125
48-58
7.7-81
N/A
34.35
74.3
120
34.7
74.2
108
5.0
2.0
ND
7.4
ND
64
177
96
55
47
24
NR
NR
NR
NR
NR
NR
Dry
Dry
Dry
NR
NR
NR
NR
Dry
Dry
Dry
Dry
Dala River, Sweden
Lake Vattern, Sweden
Lake Vanern. Sweden
Dala River, Sweden
Lake Vattem, Sweden
Lake Vanern, Sweden
Siskiwit Lake, Isle
Royale, Lake Michigan
Siskiwit Lake, Isle
Royale, Lake Michigan
Siskiwit Lake, Isle
Royale, Lake Michigan
Central Minnesota
Ontario Canada
NY/MASS
Stockholm Sweden
Baltic Sea
Iggesund Sweden
Fjord between Sweden,
Norway and Denmark
Fjord between Sweden,
Norway and Denmark
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Pristine
Pristine
Pristine
Rural
Industrial
NR
Various
Reference Site
Industrial
Industrial
Industrial
88
88
88
88
88
88
NR
NR
NR
NR
88
NR
NR
NR
NR
87
87
17
17
17
17
17
17
3
3
3
5
8
10
12
13
13
14
14
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-0.5cm AD
5-6cm AD
8-9cm AD
A/4 sites
A/25 sites
A
A
A
A/papermill
0-2cm, E
4-6cm, E
B-29
-------
Table B-6. Environmental Levels of Dibenzofurans in Sediment (ppt) (continued)
Chemical
PeCDFs (continued)
Number
samples
1
5
Number
positive
samples
1
5
Concentration
range
N/A
1,300-5,200
Cone.
mean
44
2,980
Weight
basis
Dry
NR
Location
Fjord between Sweden,
Norway and Denmark
Hamburg Germany
Location description
Industrial
Urban
Sample
year
87
NR
Ref.
no.
14
15
Comments*
9-13cm, E
Hexachlorodibenzofurans (MW=374,87)
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
HxCDFs
4
6
5
4
6
5
4
6
5
4
6
5
1
1
1
4
25
4
6
5
4
6
5
3
2
2
4
6
5
1
1
0
1
17
9.3-120
28-170
32-460
3.7-73
15-110
25-140
ND(2)-25
ND(.9-3)-4.4
ND(5-14)-14
1.8-78
32-130
36-110
N/A
N/A
N/A
ND-12
ND-6,500
44.3
89.5
163
26.7
64.7
73.4
9.38
1.58
5.68
26.2
77.7
70.8
2.0
2.0
ND
3.0
1,339
MR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
Dry
Dry
Dry
NR
NR
Dala River, Sweden
Lake Vattern, Sweden
Lake Vanern, Sweden
Dala River, Sweden
Lake Vattern, Sweden
Lake Vanern, Sweden
Dala River, Sweden
Lake Vattern, Sweden
Lake Vanern, Sweden
Data River, Sweden
Lake Vattern, Sweden
Lake Vanern, Sweden
Siskiwit Lake, Isle
Royale, Lake Michigan
Siskiwit Lake, Isle
Royale, Lake Michigan
Siskiwit Lake, Isle
Royale, Lake Michigan
Central Minnesota
Ontario Canada
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Pristine
Pristine
Pristine
Rural
Industrial
88
88
88
88
88
88
88
88
88
88
88
88
NR
NR
NR
NR
88
17
17
17
17
17
17
17
17
17
17
17
17
3
3
3
5
8
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-0.5cm, AD
5-6cm, AD
8-9cm, AD
A/4 sites
A/25 sites
B-30
-------
Table B-6. Environmental Levels of Dibenzofurans in Sediment (ppt) (continued)
Chemical
HxCDFs (continued)
Number
samples
12
3
2
4
4
1
1
5
Number
positive
samples
10
3
2
4
4
1
1
5
Concentration
range
ND-377
92-250
78-150
59-150
23-283
N/A
N/A
930-8,600
Cone,
mean
133
187
114
104
148
166
131
4,106
Weight
basis
NR
NR
Dry
Dry
Dry
Do-
Dry
NR
Location
NY/MASS
Stockholm Sweden
Baltic Sea
Iggesund Sweden
Fjord between Sweden,
Norway and Denmark
Fjord between Sweden,
Norway and Denmark
Fjord between Sweden,
Norway and Denmark
Hamburg Germany
Location description
NR
Various
Reference Site
Industrial
Industrial
Industrial
Industrial
Urban
Sample
year
NR
NR
NR
NR
87
87
87
NR
Ref.
no.
10
12
13
13
14
14
14
15
Comments*
A
A
A
A/papermill
0-2cm, E
4-6cm, E
9-13cm, E
Heptachlorodibenzofurans(MW=409.31)
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
HpCDFs
8
12
10
4
6
5
4
25
12
3
2
8
12
10
3
6
4
3
20
10
3
2
59-2180
130-1030
21-8400
ND(4)-42
4.3-91
ND(21)-260
ND-30
ND-53,000
ND-2,436
190-1,500
79-180
558
584
1764
14.25
33.2
78.3
16
11,715
1,039
997
130
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
Dry
Dala River, Sweden
Lake Vattern, Sweden
Lake Vanern, Sweden
Dala River, Sweden
Lake Vattern, Sweden
Lake Vanern, Sweden
Central Minnesota
Ontario Canada
NY/MASS
Stockholm Sweden
Baltic Sea
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Rural
Industrial
NR
Various
Reference Site
88
88
88
88
88
88
NR
88
NR
NR
NR
17
17
17
17
17
17
5
8
10
12
13
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
0-1 cm
A/4 sites
A/25 sites
A
A
A
B-31
-------
Table 13-6. Environmental Levels of Dibenzofurans in Sediment (ppt) (continued)
Chemical
HpCDFs (continued)
Number
samples
4
4
1
1
5
Number
positive
samples
4
4
1
1
5
Concentration
range
11-410
20-158
N/A
N/A
560-4,300
Cone,
mean
178
100
43
192
2,358
Weight
basis
Dry
Do-
Dry
Dry
NR
Location
Iggesund Sweden
Fjord between Sweden,
Norway and Denmark
Fjord between Sweden,
Norway and Denmark
Fjord between Sweden,
Norway and Denmark
Hamburg Germany
Location description
Industrial
Industrial
Industrial
Industrial
Urban
Sample
year
NR
87
87
87
NR
Ref.
no.
13
14
14
14
15
Comments*
A/papermill
0-2cm, E
4-6cm, E
9-13cm, E
Octachlorodibenzofurans (MW==444.76)
OCDF
18
1
1
1
4
25
12
3
2
4
4
6
5
3
1
1
1
1
21
11
1
1
2
4
6
5
ND-160
N/A
N/A
N/A
ND-23
ND-400,000
ND-1,010
ND-39
ND-3.8
ND-15
150-4250
170-1310
230-79,250
14
4
3.2
1.1
5.8
34,912
460
14
1.9
5
1212
602
19,356
Dry
Dry
Dry
Dry
NR
NR
NR
NR
Dry
Dry
NR
NR
NR
South Central Finland
Siskiwit Lake, Isle
Royale, Lake Michigan
Siskiwit Lake, Isle
Royale, Lake Michigan
Siskiwit Lake, Isle
Royale, Lake Michigan
Central Minnesota
Ontario Canada
NY/MASS
Stockholm Sweden
Baltic Sea
Iggesund Sweden
Dala River, Sweden
Lake Vattern, Sweden
Lake Vanern, Sweden
Various
Pristine
Pristine
Pristine
Rural
Industrial
NR
Various
Reference Site
Industrial
Industrial
Industrial
Industrial
88/89
NR
NR
NR
NR
88
NR
NR
NR
NR
88
88
88
1
3
3
3
5
8
10
12
13
13
17
17
17
AB
0-0.5cm, AD
5-6cm, AD
8-9cm, AD
A/4 sites
A/25 sites
A
A
A
A/Papermill
0-1 cm
0-1 cm
0-1 cm
B-32
-------
Table R-6. Environmental Levels of Dibenzofurans in Sediment (ppt) (continued)
Chemical
OCDF (continued)
Number
samples
4
1
1
5
Number
positive
samples
4
1
1
5
Concentration
range
58-151
N/A
N/A
660-5,200
Cone,
mean
96
43
192
2,712
Weight
basis
Dry
Dry
Dry
MR
Location
Fjord between Sweden,
Norway and Denmark
Fjord between Sweden,
Norway and Denmark
Fjord between Sweden,
Norway and Denmark
Hamburg Germany
Location description
Industrial
Industrial
Industrial
Urban
Sample
year
87
87
87
NR
Ref,
no.
14
14
14
15
Comments'
0-2cm, E
4-6cm, E
9-13cm, E
•Key A LOD (Ref. 1) = 20 to 50 ppt, LOD (Ref. 2) = 32 ppt, LOD (Ref. 3) = 0.4 ppt, LOD (Ref. 4) = 0.7 to 12.0 ppt, LOD (Ref. 5) = 0.61 to 4.1 ppt, LOD (Ref. 8) = 10 to 700 ppt, LOD (Ref.
10) = 3 ppt, LOD (ref. 12) = 1 to 20 ppt, LOD (Ref. 13) = 1 to 6.6 ppt.
B Dry surface sediments from 18 lakes.
C Industry produced 2,4,5 trichlorophenate (2,4,5T precursor).
D No anthropogenic inputs into drainage basin — only atmospheric sources into lake.
E Mg-production facility.
Notes
NR
N/A
ND
ppt
= Not Reported
= Not Applicable
= Not Detected
= Parts per trillion
Sources:
1. Koistinen et al. (1990)
2. Bopp et al. (1991)
3. Czuczwa et al. (1984)
4. Norwood et al. (1989)
5. Reed et al. (1990)
6. Sonzongi et al. (1991)
7. Oliver and Nilmi (1988)
8. McKee et al. (1990)
9. Huckinsetal. (1988)
10. Petty et al. (1982)
11. Sherman et al. (1990)
12. Rappe and Kjeller (1987)
13. Rappe et al. (1989a)
14. Oehme et al. (1989)
15. Gotz and Schumacher (1990)
16. Smith et al. (1990)
17. Kjeller et al. (1990)
B-33
-------
Table B-7. Environmental Levels of PCBs in Sediment (ppt)
IUPAC
number Chemical
Number
samples
Number
positive
samples
Concentration
range
Cone.
mean
Wt. basis
Location
Location description
Sample
year
Ref.
no.
Comments*
Tetrachloro-PCB (MW=291.99)
77 3,3',4,4'-TeCB
81 3,4,4',5-TeCB
18
8
5
NR
8
NR
13
8
5
NR
3
NR
N/D-550
500-360,000
5,000-27.5M
NR
ND-90,000
NR
137.7
64,250
9.47M
ND
26,880
ND
Dry
Dry
NR
Dry
Dry
Dry
South Central
Finland
Eastern Wisconsin
Waukegan, Illinois
Green Bay, Lake
Michigan
Eastern Wisconsin
Green Bay, Lake
Michigan
Various
Industrial
Urban
Various
Industrial
Various
88/89
88/89
78
NR
88/89
NR
1
6
9
16
6
16
AB
AD
5 sites
A
AD
A
Pentachloro-PCB (MW=326.44)
126 3,3',4,4',5-PeCDD
105 2,3,3',4,4'-PeCDD
114 2,3,4,4',5-PeCDD
18
8
NR
10
8
38
5
NR
8
1
2
NR
10
8
NR
5
NR
1
N/D-110
ND-10,000
NR
52-120
6,000-
490,000
NR
102,000-
131M
NR
ND-1 10,000
6.1
2,380
ND
96.4
85,120
10,000
35.14M
5,800
13,750
Dry
Dry
Dry
Dry
Dry
Dry
NR
Dry
Dry
South Central
Finland
Eastern Wisconsin
Green Bay, Lake
Michigan
South Central
Finland
Eastern Wisconsin
Lake Ontario
Waukegan, Illinois
Green Bay, Lake
Michigan
Eastern Wisconsin
Various
Industrial
Various
Various
Industrial
Various
Urban
Various
Industrial
88/89
88/89
NR
88/89
88/89
81
78
NR
88/89
1
6
16
1
6
7
9
16
6
AB
AD
A
ABC
AD
Bottom Sediment
5 Locations
A
AD
B-34
-------
Table B-7. Environmental Levels of PCBs in Sediment (ppt) (continued)
IUPAC
number Chemical
114 2,3,4,4',5-PeCDD
(contmued)
118 2,3',4,4',5-PeCDD
Number
samples
MR
8
38
NR
Number
positive
samples
NR
8
NR
NR
Concentration
range
NR
86,000-
1.48M
NR
NR
Cone.
mean
1,000
351,620
15,000
11,000
Wt, basis
Dry
Dry
Do-
Dry
Location
Green Bay, Lake
Michigan
Eastern Wisconsin
Lake Ontario
Green Bay, Lake
Michigan
Location description
Various
Industrial
Various
Various
Sample
year
NR
88/89
81
NR
Ref.
no.
16
6
7
16
Comments*
A
AD
Bottom Sediment
A
Hexachloro-PCB (MW=360.88)
156 2,3,3',4,4',5-HxCDB
167 2,3',4,4',5,5'-HxCDB
169 3,3',4,4',5,5'-HxCDB
38
NR
8
NR
18
8
NR
NR
NR
2
NR
0
3
NR
NR
NR
ND-80,000
NR
N/A
ND-19,000
NR
2,100
1,700
15,500
ND
ND
4,800
ND
Dry
Dry
Dry
Dry
Dry
Dry
Do-
Lake Ontario
Green Bay, Lake
Michigan
Eastern Wisconsin
Green Bay, Lake
Michigan
South Central
Finland
Eastern Wisconsin
Green Bay, Lake
Michigan
Various
Various
Industrial
Various
Various
Industrial
Various
81
NR
88/89
NR
88/89
88/89
NR
7
16
6
16
1
6
16
Bottom Sediment
A
AD
A
AB
AD
A
Heptachloro-PCB (MW=396.33)
189 2,3,3',4,4',5,5'HpCDB
NR
NR
NR
ND
Dry
Green Bay, Lake
Michigan
Various
NR
16
A
•Key
A LOD (Ref. l)=20to 50 ppt, LOD (Ref. 6) = l,000ppt, LOD (Ref. 16)=500ppt.
B Dry Surface Sediments from 18 lakes.
C All collected samples not analyzed.
D Superfund/Michigan "Area of Concern" Site.
B-35
-------
Table B-7. Environmental Levels of PCBs in Sediment (ppt) (continued)
Notes
MR = Not Reported
N/A = Not Applicable
ND = Not Detected
ppt = Parts per trillion
Sources:
1. Koistinen et al. (1990) 9. Huckins et al. (1988)
6. Sonzongi et al. (1991) 16. Smith et al. (1990)
7. Oliver and Nilmi (1988)
B-36
-------
Table B-8. Environmental Levels of Dioxins in Fish (ppt)
Chemical
2,3,7,8-TCDD
Fish
specie^
Eel
Eel
Trout
Grayling
Barbel
Carp
Chub
Eel
Bream
Perch
Herring
Herring
Herring
Salmon
Salmon
Perch
Artie Char
Pike
Pike
Ch. Catfish
Carp
Tissue
Liver
Fillet
Fillet
Fillet
Fillet
Fillet
Fillet
Fillet
Whole
Whole
Whole
Muscle
Muscle
NR
NR
Muscle
Muscle
Fillets
Fillets
Number
samples
6
5
1
1
1
1
1
5
14
3
1
2
2
2
2
3
5
8
1
8
14
Number
positive
samples
Cone, range
Cone.
mean
Wt.
basis
Location*
Location
description
Samp.
year
Ref.
no.
Comments
Tetrachlorodibenzo-p-dloxins(MW=321 .98)
6
5
1
1
1
1
0
5
14
3
0
0
0
2
2
3
5
8
1
8
10
1.2-9.1
2.4-3.3
1.4
3.8
5.1
2.5
ND(2.3)
0.9-1.5
1.4-94.4
1.8-8.1
ND(0.1)
ND(0.1)
ND(0.1)
4.6-19.0
0.2-0.3
2.6-19
6.5-25
40-833
78
28-695
20-153
3.32
3.04
NA
NA
NA
NA
NA
1.3
18.0
5.9
NA
NA
NA
11.8
0.25
11.5
14.3
186
NA
157
55
NR
Fat
Fat
Fat
Fat
Fat
Fat
Fat
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fat
Fat
NR
NR
various, Netherlands
Rhine river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Hamburg, Germany
Hamburg, Germany
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Gulf of Bothnia, Sweden
Gulf of Bothnia, Sweden
Gulf of Bothnia, Sweden
Lake Vattern, Sweden
Lake Vanern, Sweden
Hedesunda Bay, Sweden
various, Michigan
various, Michigan
NR
NR
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Pristine
Urban
Industrial
NR
NR
NR
NR
Industrial
Industrial
NR
NR
NR
88
88
88
88
88
88
87-88
84
84
NR
NR
NR
NR
NR
NR
NR
88
88
78
78
1
2
2
2
2
2
2
2
4
4
5
5
5
5
5
5
5
19
19
7
7
one sample near dump site
up & downstream Basal
composite sample
composite sample
composite 5-10 fish
composite 5-10 fish
composite 5-10 fish
wild salmon
hatched salmon
caught near pulp mill
samples composite 2-5 fish
composite 5 fish
B-37
-------
Table B-8. Environmental Levels of Dioxins in Fish (ppt) (continued)
Chemical
2,3,7,8-TCDD
(continued)
Fish
species*
Y. Perch
Sm. M. Basss
Sucker
Lake Trout
Carp
Carp
Carp
Blue Crab
Lobster
Str. Bass
Lake Trout
Lake Trout
Lake Trout
Walleye
Walleye
Lake Trout
Lake Trout
Br. Trout
Br. Trout
Rb. Trout
Rb. Trout
Lake Trout
Tissue
Fillets
Fillets
Fillets
Fillets
Whole
Whole
Whole
Meat
Meat
Fillets
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Fillets
Whole
Fillets
Whole
Number
samples
6
2
4
2
3
3
2
2
2
2
1
1
3
1
1
1
10
1
1
1
1
1
Number
positive
samples
3
2
3
0
0
0
2
2
2
2
1
1
3
1
1
1
10
1
1
1
1
1
Cone, range
10-20
7-8
4-21
ND(5.0)
ND(6.6)
ND(6.6)
3-28
105.7-116.1
4.7-6.3
83.9-733.9
1.0
8.6
3.5-5.8
1.8
6.6
48.9
3.0-8.7
6
5
14
5
18
Cone.
mean
13
8
10
NA
NA
NA
16
110.9
5.5
408.9
NA
NA
4.4
NA
NA
NA
4.2
NA
NA
NA
NA
NA
Wt.
basis
NR
NR
NR
NR
NR
NR
NR
Wet
Wet
Wet
Wet
Wet
Wet
Wet
Wet
Wet
Wet
NR
NR
NR
NR
NR
Location*
various, Michigan
various, Michigan
various, Michigan
various, Michigan
Mississippi river, MN
Lake Orono, MN
Lake Huron
Passaic river, NJ
New York Bight
Newark Bay, NJ
Lake Superior
Lake Huron
Lake Michigan
Lake Erie
Lake St. Clair
Lake Ontario
Lake Michigan
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Location
description
NR
NR
NR
NR
Industrial
Industrial
NR
Urban
Dump Site
Urban
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
Samp.
year
78
78
78
78
NR
NR
NR
NR
NR
NR
84
84
84
84
84
84
84
NR
NR
NR
NR
NR
Ref.
no.
7
7
7
7
8
8
18
10
10
10
11
11
11
11
11
11
11
12
12
12
12
12
Comments
Elk river power station
Elk river power station
samples composite 3-5 fish
former sewage sludge
mean 5 samples
mean 5 samples
range 3 sample sites
mean 5 samples
mean 5 samples
mean 5 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
B-38
-------
Table B-8. Environmental Levels of Dioxins in Fish (ppt) (continued)
Chemical
2,3,7,8-TCDD
(continued)
TCDDs
Fish
species*
Lake Trout
Coho Salmon
Coho Salmon
Cod
Haddock
P. Flounder
Plaice
Flounder
Eel
Mussel
Shrimp
Cod
Various"
Various"
Bream
Perch
Carp
Carp
Blue Crab
Lobster
Str. Bass
Br. Trout
Tissue
Fillets
Whole
Fillets
Fillets
Fillets
Fillets
Fillets
Fillets
Fillets
Muscle
Muscle
Fillets
Mixed4
Mixed1
Whole
Whole
Meat
Meat
Fillets
Whole
Number
samples
1
1
1
4
1
1
1
1
4
3
2
6
314
34
13
2
3
3
2
2
2
1
Number
positive
samples
1
1
1
0
0
0
0
0
1
0
0
NR
220
34
13
2
1
0
2
2
2
1
Cone, range
8
20
6
ND(l.O)
ND(0.2)
ND(0.2)
ND(0.5)
ND(0.5)
ND-1.4
ND(0.5)
ND(2.0)
ND-3.8
NR
0.06-2.26
2.5-102.4
9.0-10.5
ND-3.9
ND(6.6)
117.6-149.5
6.6-8.3
85.4-733.9
11
Cone.
mean
NA
NA
NA
NA
NA
NA
NA
NA
0.35
NA
NA
NR
6.84
0.56
17.6
9.8
1.3
NA
133.6
7.4
409.6
NA
Wt,
basis
NR
NR
NR
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Wet
Wet
Fresh
Fresh
NR
NR
Wet
Wet
Wet
NR
Location11
Lake Ontario
Lake Ontario
Lake Ontario
various, Sweden
various, Sweden
various, Sweden
various, Sweden
various, Sweden
various, Sweden
Grenlandsfjord.Sweden
G re nlandsfjord, Sweden
Frierfjord, Sweden
Various, US
Various, US
Hamburg, Germany
Hamburg, Germany
Mississippi river, MN
Lake Orono, MN
Passaic river, NJ
New York Bight
Newark Bay, NJ
Lake Ontario
Location
description
NR
NR
NR
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Various
Background'
Urban
Urban
Industrial
Industrial
Urban
Dump Site
Urban
NR
Samp.
year
NR
NR
NR
88
88
88
88
88
87-88
87
88
87
86-89
86-89
84
84
NR
NR
NR
NR
NR
NR
Ref.
no.
12
12
12
17
17
17
17
17
17
17
17
17
20
20
4
4
8
8
10
10
10
12
Comments
composite 3 samples
composite 3 samples
composite 3 samples
composite 10 samples
composite 10 samples
composite 10 samples
composite 10 samples
only cone . range given
samples composite 3-5 fish
samples composite 3-5 fish
Elk river power station
Elk river power station
former sewage sludge
composite 3 samples
B-39
-------
Table B-8. Environmental Levels of Dioxins in Fish (ppt) (continued)
Chemical
TCDDs (continued)
Fish
species*
Br. Trout
Rb. Trout
Rb. Trout
Lake Trout
Lake Trout
Coho Salmon
Coho Salmon
Tissue
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Number
samples
1
1
1
1
1
1
1
Number
positive
samples
1
1
1
1
1
1
1
Cone, range
9
29
11
32
12
22
9
Cone.
mean
NA
NA
NA
NA
NA
NA
NA
Wt.
basis
NR
NR
NR
NR
NR
NR
NR
Location*
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Location
description
NR
NR
NR
NR
NR
NR
NR
Samp.
year
NR
NR
NR
NR
NR
NR
NR
Ref.
no.
12
12
12
12
12
12
12
Comments
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
compositeS samples
composite 3 samples
composite3 samples
Pentachlorodibenzo-p-dioxins(MW=356.42)
1,2,3,7,8-PeCDD
PeCDDs
Carp
Pike
Pike
Herring
Herring
Herring
Various0
Various'
Bream
Perch
Carp
Carp
Blue Crab
Lobster
Whole
Muscle
Muscle
Whole
Whole
Whole
Mixed1
Mixed1
Whole
Whole
Meat
Meat
2
8
1
1
2
2
314
34
13
2
3
3
2
2
2
8
1
1
2
2
170
34
13
2
3
0
2
2
2-11
70-250
39
0.6
1.1-2.8
2.0-4.7
NR
0.15-2.67
3.2-27.8
9.8-29.8
3.5-4.5
ND(6.6)
17.2-18.0
10.0-11.0
6
129
NA
NA
1.95
3.35
2.38
0.77
12.1
19.8
3.9
NA
17.6
10.5
NR
Fat
Fat
Fresh
Fresh
Fresh
Wet
Wet
Fresh
Fresh
NR
NR
Wet
Wet
Lake Huron
Lake Vanern, Sweden
Hedesunda Bay, Sweden
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Various, US
Various, US
Hamburg, Germany
Hamburg, Germany
Mississippi river, MN
Lake Orono, MN
Passaic river, NJ
New York Bight
NR
Industrial
Industrial
Pristine
Urban
Industrial
Various
Background*
Urban
Urban
Industrial
Industrial
Urban
Dump Site
NR
88
88
NR
NR
NR
86-89
86-89
84
84
NR
NR
NR
NR
18
19
19
5
5
5
20
20
4
4
8
8
10
10
samples composite 3-5 fish
samples composite 2-5 fish
composite 5 fish
composite 5-10 fish
composite 5-10 fish
composite 5-10 fish
samples composite 3-5 fish
samples composite 3-5 fish
Elk river power station
Elk river power station
former sewage sludge
B-40
-------
Table B-8. Environmental Levels of Dioxins in Fish (ppt) (continued)
Chemical
PeCDDs (continued)
Fish
species*
Sir. Bass
Br. Trout
Br. Trout
Rb. Trout
Rb. Trout
Lake Trout
Lake Trout
Coho Salmon
Coho Salmon
Tissue
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Number
samples
2
1
1
1
1
1
1
1
1
Number
positive
samples
2
1
1
1
1
1
1
1
1
Cone, range
5.2-10.6
8
6
31
15
39
9
6
5
Cone.
mean
7.9
NA
NA
NA
NA
NA
NA
NA
NA
Wt.
basis
Wet
NR
NR
NR
NR
NR
NR
NR
NR
Location1'
Newark Bay, NJ
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Location
description
Urban
NR
NR
NR
NR
NR
NR
NR
NR
Samp.
year
NR
NR
NR
NR
NR
NR
NR
NR
NR
Ref.
no.
10
12
12
12
12
12
12
12
12
Comments
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
Hexachlorodibenzo-p-dioxins(MW=390.87)
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
Carp
Pike
Pike
Herring
Herring
Herring
Various'
Pike
Pike
Herring
Herring
Whole
Muscle
Muscle
Whole
Whole
Whole
Mixed*
Muscle
Muscle
Whole
Whole
2
8
1
1
2
2
314
8
1
1
2
2
8
1
0
2
0
100
8
1
0
1
3-5
6.7-22
11
ND(0.2)
0.2-0.3
ND(0.2)
NR
30-100
22
ND(0.2)
ND-2.4
4
12.8
NA
NA
0.25
NA
1.67
50.5
NA
NA
1.25
NR
Fat
Fat
Fresh
Fresh
Fresh
Wet
Fat
Fat
Fresh
Fresh
Lake Huron
Lake Vanern, Sweden
Hedesunda Bay, Sweden
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Various, US
Lake Vanern, Sweden
Hedesunda Bay, Sweden
Atlantic Coast, Sweden
Baltic Sea, Sweden
NR
Industrial
Industrial
Pristine
Urban
Industrial
Various
Industrial
Industrial
Pristine
Urban
NR
88
88
NR
NR
NR
86-89
88
88
NR
NR
18
19
19
5
5
5
20
19
19
5
5
samples composite 3-5 fish
samples composite 2-5 fish
composite 5 fish
composite 5-10 fish
composite 5-10 fish
composite 5-10 fish
samples composite 3-5 fish
samples composite 2-5 fish
composite 5 fish
composite 5-10 fish
composite 5-10 fish
B-41
-------
Table B-8. Environmental Levels of Dioxins in Fish (ppt) (continued)
Chemical
1,2,3,6,7,8-HxCDD
(continued)
1,2,3,7,8,9-HxCDD
HxCDDs
Fish
species"
Herring
Various0
Pike
Pike
Herring
Herring
Herring
Various0
Bream
Perch
Carp
Carp
Blue Crab
Lobster
Str. Bass
Br. Trout
Br. Trout
Rb. Trout
Rb. Trout
Lake Trout
Lake Trout
Coho Salmon
Tissue
Whole
Mixed1
Muscle
Muscle
Whole
Whole
Whole
Mixed1
Whole
Whole
Meat
Meat
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Whole
Number
samples
2
314
8
1
1
2
2
314
13
2
3
3
2
2
2
1
1
1
1
1
1
1
Number
positive
samples
2
217
0
0
0
0
0
119
13
2
3
1
2
2
2
1
1
1
1
1
1
1
Cone, range
2.2-8.1
NR
ND(3-1 1)
ND(6)
ND(0.2)
ND(0.2)
ND(0.2)
NR
4.3-46.4
18.6-21.5
2.3-11
ND-3.0
0.3-1.5
3.0-3.4
0.6-0.7
20
25
67
37
114
27
16
Cone.
mean
5.15
4.29
NA
NA
NA
NA
NA
1.15
17.8
20.0
6.9
1.0
0.9
3.2
0.65
NA
NA
NA
NA
NA
NA
NA
Wt.
basis
Fresh
Wet
Fat
Fat
Fresh
Fresh
Fresh
Wet
Fresh
Fresh
NR
NR
Wet
Wet
Wet
NR
NR
NR
NR
NR
NR
NR
Location'
Gulf of Bothnia, Sweden
Various, US
Hedesunda Bay, Sweden
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Various, US
Hamburg, Germany
Hamburg, Germany
Mississippi river, MN
Lake Orono, MN
Passaic river, NJ
New York Bight
Newark Bay, NJ
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Location
description
Industrial
Various
Industrial
Pristine
Urban
Industrial
Various
Urban
Urban
Industrial
Industrial
Urban
Dump Site
Urban
NR
NR
NR
NR
NR
NR
NR
Samp.
year
NR
86-89
88
88
NR
NR
NR
86-89
84
84
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
Ref.
no.
5
20
19
19
5
5
5
20
4
4
8
8
10
10
10
12
12
12
12
12
12
12
Comments
composite 5-10 fish
samples composite 3-5 fish
.
composite 5 fish
composite 5-10 fish
composite 5-10 fish
composite 5-10 fish
samples composite 3-5 fish
Elk river power station
Elk river power station
former sewage sludge
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
B-42
-------
Table B-8. Environmental Levels of Dioxins in Fish (ppt) (continued)
Chemical
HxCDDs (continued)
1,2,3,4,6,7,8-HpCDD
HpCDDs
Fish
species*
Coho Salmon
Various"
Tissue
Fillets
Mixed1
Number
samples
1
34
Number
positive
samples
1
NR
Cone, range
22
ND-3.57
Cone.
mean
NA
0.39
Wt.
basis
NR
Wet
Location'
Lake Ontario
Various, US
Location
description
NR
Background'
Samp.
year
NR
86-89
Ref.
no.
12
20
Comments
composite 3 samples
samples composite 3-5 fish
Heptachlorodibenzo-p-dioxins (MW = 425 .3 1)
Carp
Herring
Herring
Herring
Various"
Bream
Perch
Carp
Carp
Blue Crab
Lobster
Str. Bass
Br. Trout
Br. Trout
Rb. Trout
Rb. Trout
Lake Trout
Lake Trout
Coho Salmon
Whole
Whole
Whole
Whole
Mixed*
Whole
Whole
Meat
Meat
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Whole
2
1
2
2
314
13
2
3
3
2
2
2
1
1
1
1
1
1
1
2
0
1
0
279
13
2
3
2
0
1
2
1
1
1
0
1
0
1
3-4
ND(0.2)
ND-0.6
ND(0.2)
NR
1.5-14.4
3.5-7.1
15-22
ND-11
ND(l.l)
ND-8.5
4.0-11.4
7
9
12
ND(7)
16
ND(7)
30
3.5
NA
0.35
NA
10.5
6.7
5.3
19.3
7.0
NA
4.25
7.7
NA
NA
NA
NA
NA
NA
NA
NR
Fresh
Fresh
Fresh
Wet
Fresh
Fresh
NR
NR
Wet
Wet
Wet
NR
NR
NR
NR
NR
NR
NR
Lake Huron
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Various, US
Hamburg, Germany
Hamburg, Germany
Mississippi river, MN
Lake Orono, MN
Passaic river, NJ
New York Bight
Newark Bay, NJ
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
NR
Pristine
Urban
Industrial
Various
Urban
Urban
Industrial
Industrial
Urban
Dump Site
Urban
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
86-89
84
84
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
18
5
5
5
20
4
4
8
8
10
10
10
12
12
12
12
12
12
12
samples composite 3-5 fish
composite 5-10 fish
composite 5-10 fish
composite 5-10 fish
samples composite 3-5 fish
Elk river power station
Elk river power station
former sewage sludge
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
B-43
-------
Table B-8. Environmental Levels of Dioxins in Fish (ppt) (continued)
Chemical
HpCDDs (continued)
Fish
species"
Coho Salmon
Tissue
Fillets
Number
samples
1
Number
positive
samples
1
Cone, range
50
Cone.
mean
NA
Wt.
basis
MR
Location'
Lake Ontario
Location
description
NR
Samp.
year
NR
Ref.
no.
12
Comments
composite3 samples
Octachlorodibenzo-p-dioxin(MW=460,76)
1,2,3,4,6,7,8,9-OCDD
Eel
Trout
Grayling
Barbel
Carp
Chub
Eel
Bream
Perch
Herring
Herring
Herring
Salmon
Salmon
Perch
Pike
Pike
Carp
Carp
Carp
Fillet
Fillet
Fillet
Fillet
Fillet
Fillet
Fillet
Whole
Whole
Whole
Muscle
Muscle
MR
Muscle
Muscle
Whole
Whole
Whole
5
1
1
1
1
1
5
14
3
1
2
2
2
2
3
8
1
3
3
2
5
0
1
1
1
1
5
14
3
1
1
1
1
2
3
8
1
3
3
2
28-60
ND(5.0)
47
9.0
23
15
25-40
1.4-5.1
2.3-10.5
1.1
ND-0.7
ND-0.3
ND-1.5
0.8-1.9
0.6-0.8
10-17
22
56-62
35-43
3-5
44.4
NA
NA
NA
NA
NA
30
2.5
5.2
NA
0.4
0.2
0.75
1.35
0.73
14.75
NA
59
39
4
Fat
Fat
Fat
Fat
Fat
Fat
Fat
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fat
Fat
NR
NR
NR
Rhine river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Hamburg, Germany
Hamburg, Germany
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Gulf of Bothnia, Sweden
Gulf of Bothnia, Sweden
Gulf of Bothnia, Sweden
Lake Vanern, Sweden
Hedesunda Bay, Sweden
Mississippi river, MN
Lake Orono, MN
Lake Huron
NR
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Pristine
Urban
Industrial
NR
NR
NR
Industrial
Industrial
Industrial
Industrial
NR
88
88
88
88
88
88
87-88
84
84
NR
NR
NR
NR
NR
NR
88
88
NR
NR
NR
2
2
2
2
2
2
2
4
4
5
5
5
5
5
5
19
19
8
8
18
up & downstream Basal
composite sample
composite sample
composite 5-10 fish
composite 5-10 fish
composite 5-10 fish
wild salmon
hatched salmon
caught near pulp mill
samples composite 2-5 fish
composite 5 fish
Elk river power station
Elk river power station
samples composite 3-5 fish
B-44
-------
Table B-8. Environmental Levels of Dioxins in Fish (ppt) (continued)
Chemical
1,2,3,4,6,7,8,9-OCDD
(continued)
Fish
species*
Blue Crab
Lobster
Str. Bass
Lake Trout
Lake Trout
Lake Trout
Walleye
Walleye
Lake Trout
Lake Trout
Br. Trout
Br. Trout
Rb. Trout
Rb. Trout
Lake Trout
Lake Trout
Coho Salmon
Coho Salmon
Cod
Haddock
P. Flounder
Plaice
Tissue
Meat
Meat
Fillets
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Fillets
Fillets
Fillets
Fillets
Number
samples
2
2
2
1
1
3
1
1
1
10
1
1
1
1
1
1
1
1
4
1
1
1
Number
positive
samples
2
2
2
1
1
3
1
1
1
10
0
1
0
0
1
I
1
1
3
0
1
1
Cone, range
34.3-78.8
6.3-10.9
5.M9.5
1.0
0.7
1.1-2.5
2.8
1.8
1.2
0.8-3.7
ND<7)
11
ND(7)
ND(7)
89
28
160
280
ND-11
ND(3.6)
3.4
424
Cone.
mean
56.6
8.6
27.3
NA
NA
1.8
NA
NA
NA
1.6
NA
NA
NA
NA
NA
NA
NA
NA
4.95
NA
NA
NA
Wt.
basis
Wet
Wet
Wet
Wet
Wet
Wet
Wet
Wet
Wet
Wet
NR
NR
NR
NR
NR
NR
NR
NR
Fresh
Fresh
Fresh
Fresh
Location1"
Passaic river, NJ
New York Bight
Newark Bay, NJ
Lake Superior
Lake Huron
Lake Michigan
Lake Erie
Lake St. Clair
Lake Ontario
Lake Michigan
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
various, Sweden
various, Sweden
various, Sweden
various, Sweden
Location
description
Urban
Dump Site
Urban
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
Industrial
Industrial
Industrial
Industrial
Samp.
year
NR
NR
NR
84
84
84
84
84
84
84
NR
NR
NR
NR
NR
NR
NR
NR
88
88
88
88
Ref.
no.
10
10
10
11
11
11
11
11
11
11
12
12
12
12
12
12
12
12
17
17
17
17
Comments
former sewage sludge
mean 5 samples
mean 5 samples
range 3 sample sites
mean 5 samples
mean 5 samples
mean 5 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 10 samples
composite 10 samples
composite 10 samples
B-45
-------
Table B-8. Environmental Levels of Dioxins in Fish (ppt) (continued)
Chemical
1,2,3,4,6,7,8,9-OCDD
(continued)
Fish
species*
Flounder
Eel
Mussel
Shrimp
Cod
Tissue
Fillets
Fillets
Muscle
Muscle
Fillets
Number
samples
1
4
3
2
6
Number
positive
samples
1
3
3
1
MR
Cons, range
2.4
ND-770
12-140
ND-18
0.63-2.2
Cone,
mean
NA
204.2
62.0
9.0
NR
Wt,
basis
Fresh
Fresh
Fresh
Fresh
Fresh
Location!1
various, Sweden
various, Sweden
G re nlandsfjord, Sweden
Grenlandsfjord .Sweden
Frierfjord, Sweden
Location
description
Industrial
Industrial
Industrial
Industrial
Industrial
Samp.
year
88
87-88
87
88
87
Ref,
no.
17
17
17
17
17
Comment*
composite 10 samples
only cone, range given
Footnote References
• Ch. = Channel; Y. = Yellow; Sm. M. = Small Mouth; Str. = Striped; Br. = Brown; Rb. = Rainbow; P. = Pole.
b Various, Netherlands = samples taken from six locations around Ijsselmeer Lake; Various, Michigan = samples taken from Tittabawassee River, Grand River, Saginaw River, Saginaw Bay, and Lake Michigan; Various, Sweden
— samples taken from Grenlandsfjord and Frierfjord, US — samples taken from 314 sites across the US, including industrial and background sites.
' Species were taken from both bottom feeders and open water feeders, and then composited.
' Whole fish samples and fillet samples were combined during analysis.
* The 34 background sites are a subset of the 314 US sites.
NOTES: Summary statistics provided in or derived from references; means from references included non-detects counted as 0.0; when reference did not compute mean, it was computed with same strategy;
NA = not applicable;
ND = non-detected (limit of detection);
NR = not reported.
Descriptions provided were those given by reference or surmised from study description when not given.
Sources: 1. Van den Berg, et al. (1987)
2. Frommberger (1991)
4. Gotz, et al. (1990)
5. Rappe, et al. (1989)
7. Harless and Lewis (1982)
8. Reed, et al. (1990)
10. Rappe, et al. (1991)
11. Vault, etal. (1989)
12. Niimi and Oliver (1989a)
17. Oehme, et al. (1989)
18. Stalling, et al. (1983)
19. Kjeller, et al. (1990)
20. USEPA(1992)
B-46
-------
Table B-9. Environmental Levels of Dibenzofurans in Fish (ppt)
Chemical
Fish
species*
2,3,7,8-TCDF
Eel
Eel
Br. Trout
Grayling
Barbel
Carp
Chub
Eel
Herring
Herring
Herring
Herring
Salmon
Salmon
Perch
Artie Char
Pike
Pike
Carp
Carp
Carp
Tissue
Number
samples
Number
positive
samples
Liver
Fillets
Fillets
Fillets
Fillets
Fillets
Fillets
Fillets
Adipose
Whole
Whole
Whole
Muscle
Muscle
NR
MR
Muscle
Muscle
Whole
Whole
Whole
6
5
1
1
1
1
1
5
2
1
2
2
2
2
3
5
8
1
2
3
3
0
5
1
1
1
1
1
5
2
1
2
2
2
2
3
5
8
1
2
3
3
Cone, range
Cone.
mean
Wt.
basis
Location*1
Tetrachlorodibenzofiirans (MW=305.98)
ND
2.1-12
45
108
57
58
128
0.9-2.0
3-4
1.7
5.3-5.5
3.0-6.2
28-35
7.8-9.0
2.1-8.7
21-75
330-3000
430
11
1.8-3.0
1.0-1.3
NA
6.98
NA
NA
NA
NA
NA
1.35
3.5
NA
5.4
4.6
31.5
8.4
5.4
55
774
NA
NA
2.6
1.1
NR
Fat
Fat
Fat
Fat
Fat
Fat
Fat
Fat
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fat
Fat
NR
NR
NR
various, Netherlands
Rhine river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
South Baltic Sea
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Gulf of Bothnia, Sweden
Gulf of Bothnia, Sweden
Gulf of Bothnia, Sweden
Lake Vattem, Sweden
Lake Vanern, Sweden
Hedesunda Bay, Sweden
Lake Huron
Mississippi river, MN
Lake Orono, MN
Location
description
NR
NR
Urban
Urban
Urban
Urban
Urban
Urban
NR
Pristine
Urban
Industrial
NR
NR
NR
NR
Industrial
Industrial
NR
Industrial
Industrial
Samp.
year
NR
88
88
88
88
88
88
87-88
NR
NR
NR
NR
NR
NR
NR
NR
88
88
NR
NR
NR
Ref.
no.
1
2
2
2
2
2
2
2
3
5
5
5
5
5
5
5
19
19
18
8
8
Comments
one sample near dump site
up & downstream Basal
Composite sample
Composite sample
composite 2-5 fish
composite 2-5 fish
composite 2-5 fish
wild salmon
hatched salmon
caught near pulp mill
samples composite 2-5 fish
composite 5 fish
samples composite 3-5 fish
Elk river power station
Elk river power station
B-47
-------
Table B-9. Environmental Levels of Dibenzofurans in Fish (ppt) (continued)
Chemical
2,3,7,8-TCDF
(continued)
Fish
species*
Carp
Catfish
Smk. Chub
Str. Bass
Lg. M. Bass
Lake Trout
Blue Crab
Lobster
Str. Bass
Lake Trout
Lake Trout
Lake Trout
Walleye
Walleye
Lake Trout
Lake Trout
Br. Trout
Br. Trout
Rb. Trout
Rb. Trout
Lake Trout
Lake Trout
Tissue
Fillets
Fillets
Fillets
Fillets
Fillets
Fillets
Meat
Meat
Fillets
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Fillets
Whole
Fillets
Whole
Fillets
Number
samples
1
1
1
5
1
3
2
2
2
1
1
3
1
1
1
10
1
1
1
1
1
1
Number
positive
samples
1
1
1
5
1
3
2
2
2
1
1
3
1
1
1
10
1
1
1
1
1
1
Cone, range
49
6
3
7-93
10
11-56
11.0-15.5
3.5-4.1
51.9-85.5
14.8
22.8
34.8-42.3
11.3
24.8
18.5
27.0-56.0
11
8
19
6
15
6
Cone.
mean
NA
NA
NA
28.0
NA
31.7
13.25
3.8
68.7
NA
NA
39.5
NA
NA
NA
38.4
NA
NA
NA
NA
NA
NA
Wt.
basis
NR
NR
NR
NR
NR
NR
Wet
Wet
Wet
Wet
Wet
Wet
Wet
Wet
Wet
Wet
NR
NR
NR
NR
NR
NR
Location11
NR
Saginaw river
NR
Hudson river
Hudson river
Lake Superior
Passaic river, NJ
New York Bight
Newark Bay, NJ
Lake Superior
Lake Huron
Lake Michigan
Lake Erie
Lake St. Clair
Lake Ontario
Lake Michigan
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Location
description
NR
NR
NR
NR
NR
NR
Urban
Dump Site
Urban
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
Samp.
year
78
84
85
85
85
85
NR
NR
NR
84
84
84
84
84
84
84
NR
NR
NR
NR
NR
NR
Ref.
no.
9
9
9
9
9
9
10
10
10
11
11
11
11
11
11
11
12
12
12
12
12
12
Comments
contamimated site
contaminated site
contaminated site
contaminated site
contaminated site
contaminted site
former sewage sludge
mean 5 samples
mean 5 samples
range 3 sample sites
mean 5 samples
mean 5 samples
mean 5 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
B-48
-------
Table B-9. Environmental Levels of Dibenzofurans in Fish (ppt) (continued)
Chemical
2,3,7,8-TCDF
(continued)
TCDFs
Fish
species*
Coho Salmon
Coho Salmon
Cod
Haddock
P. Flounder
Plaice
Flounder
Eel
Mussel
Shnmp
Cod
Various'
Various0
Bream
Perch
Y. Perch
Carp
Carp
Blue Crab
Lobster
Str. Bass
Br. Trout
Tissue
Whole
Fillets
Fillets
Fillets
Fillets
Fillets
Fillets
Fillets
Muscle
Muscle
Fillets
Mixed1
Mixed1
Whole
Whole
Whole
Meat
Meat
Fillets
Whole
Number
samples
1
1
4
1
1
1
1
4
3
2
6
314
34
13
2
1
3
3
2
2
2
1
Number
positive
samples
1
1
4
1
1
1
1
3
3
2
MR
279
34
13
2
1
3
3
2
2
2
1
Cone, range
20
6
0.2-1.4
0.75
0.28
1.4
1.2
ND-68
16.1-61
6.1-37
0.49-14.3
NR
0.1-13.73
7.8-86.5
10.7-41.1
1060
2.4-4.0
1.0-1.3
132.9-164.5
23.0-31.2
77.2-107.7
11
Cone.
mean
NA
NA
0.62
NA
NA
NA
NA
17.1
32.4
21.6
NR
13.6
1.61
44.3
25.9
NA
3.1
1.2
148.7
27.1
92.4
NA
Wi.
basis
NR
NR
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Wet
Wet
Fresh
Fresh
NR
NR
NR
Wet
Wet
Wet
NR
Location"
Lake Ontario
Lake Ontario
various, Sweden
various, Sweden
various, Sweden
various, Sweden
various, Sweden
various, Sweden
G re nlandsfjord, Sweden
Grenlandsfjord,Sweden
Frierfjord, Sweden
Various, US
Various, US
Hamburg, Germany
Hamburg, Germany
Housatonic river, MA
Mississippi river, MN
Lake Orono, MN
Passaic river, NJ
New York Bight
Newark Bay, NJ
Lake Ontario
Location
description
NR
NR
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Industrial
Various
Background*
Urban
Urban
NR
Industrial
Industrial
Urban
Dump Site
Urban
NR
Samp*
year
NR
NR
88
88
88
88
88
88
87
88
87
86-89
86-89
84
84
NR
NR
NR
NR
NR
NR
NR
Refc
no.
12
12
17
17
17
17
17
17
17
17
17
20
20
4
4
6
8
8
10
10
10
12
Comments
composite 3 samples
composite 3 samples
composite 10 samples
composite 10 samples
composite 10 samples
composite 10 samples
only cone, range given
samples composite 3-5 fish
samples composite 3-5 fish
Elk river power station
Elk river power station
former sewage sludge
composite 3 samples
B-49
-------
Table B-9. Eaviromnental Levels of Dibenzofurans in Fish (ppt) (continued)
Chemical
TCDFs
(continued
TCDFs
other than
2,3,7,8-TCDF
Fish
species*
Br. Trout
Rb. Trout
Rb. Trout
Lake Trout
Lake Trout
Coho Salmon
Coho Salmon
Eel
Grayling
Barbel
Carp
Chub
Eel
Tissue
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Fillets
Fillets
Fillets
Fillets
Fillets
Fillets
Number
samples
1
1
1
1
1
1
1
5
1
1
1
5
2
Number
positive
samples
1
1
1
1
1
1
1
5
1
1
1
5
2
Cone, range
8
19
6
18
6
20
6
4.6-13
142
77
14
17
1.5
Cone.
mean
NA
NA
NA
NA
NA
NA
NA
8.8
NA
NA
NA
NA
NA
Wt.
basis
NR
NR
NR
NR
NR
NR
NR
Fat
Fat
Fat
Fat
Fat
Fat
Location*
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Rhine river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Neckar river, Germany
Location
description
NR
NR
NR
NR
NR
NR
NR
NR
Urban
Urban
Urban
Urban
Urban
Samp.
year
NR
NR
NR
NR
NR
NR
NR
88
88
88
88
88
88
Ref.
no.
12
12
12
12
12
12
12
2
2
2
2
2
2
Comments
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
up & downstream Basal
Composite sample
composite sample
PentachIorodibenzofiirans(MW=340.42)
1,2,3,7,8-PeCDF
Carp
Pike
Pike
Herring
Herring
Herring
Various0
Various'
Whole
Muscle
Muscle
Whole
Whole
Whole
Mixed4
Mixed4
2
8
1
1
2
2
314
34
2
8
1
1
2
2
151
34
1-5
43-140
39
0.4
1.4-2.5
0.8-0.9
NR
0.1-1.90
3
73.2
NA
NA
1.95
0.85
1.71
0.43
NR
Fat
Fat
Fresh
Fresh
Fresh
Wet
Wet
Lake Huron
Lake Vanern, Sweden
Hedesunda Bay, Sweden
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Various, US
Various, US
NR
Industrial
Industrial
Pristine
Urban
Industrial
Various
Background*
NR
88
88
NR
NR
NR
86-89
86-89
18
19
19
5
5
5
20
20
samples composite 3-5 fish
samples composite 2-5 fish
composite 5 fish
composite 2-5 fish
composite 2-5 fish
composite 2-5 fish
samples composite 3-5 fish
samples composite 3-5 fish
B-50
-------
Table B-9. Environmental Levels of Dibenzofurans in Fish (ppt) (continued)
Chemical
2,3,4,7,8-PeCDF
PeCDFs
Fish
species*
Carp
Pike
Pike
Herring
Herring
Herring
Various'
Various0
Bream
Perch
Y. Perch
Carp
Carp
Blue Crab
Lobster
Str. Bass
Br. Trout
Br. Trout
Rb. Trout
Rb. Trout
Lake Trout
Lake Trout
Tissue
Whole
Muscle
Muscle
Whole
Whole
Whole
Mixed"
Mixed*
Whole
Whole
Whole
Meat
Meat
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Number
samples
2
8
1
1
2
2
314
34
13
2
1
3
3
2
2
2
1
1
1
1
1
1
Number
positive
samples
2
8
1
1
2
2
201
34
13
2
1
3
2
2
2
2
0
1
1
1
1
1
Cone, range
4-11
120-290
110
3.0
6.8-19.0
8.8-8.9
NR
0.1-1.39
13.9-114
62.3-153.3
640
15-45
ND-13
89.7-94.1
29.8-37.3
34.3-82.6
ND(7)
3
8
7
39
8
Cone.
mean
7.5
189
NA
NA
12.9
8.85
3.06
0.5
65.4
107.8
NA
26.3
9.0
91.9
33.6
58.4
NA
NA
NA
NA
NA
NA
Wt.
basis
NR
Fat
Fat
Fresh
Fresh
Fresh
Wet
Wet
Fresh
Fresh
NR
NR
NR
Wet
Wet
Wet
NR
NR
NR
NR
NR
NR
Locationb
Lake Huron
Lake Vanern, Sweden
Hedesunda Bay, Sweden
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Various, US
Various, US
Hamburg, Germany
Hamburg, Germany
Housatonic river, MA
Mississippi river, MN
Lake Orono, MN
Passaic river, NJ
New York Bight
Newark Bay, NJ
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Location
description
NR
Industrial
Industrial
Pristine
Urban
Industrial
Various
Background*
Urban
Urban
NR
Industrial
Industrial
Urban
Dump Site
Urban
NR
NR
NR
NR
NR
NR
Samp.
year
NR
88
88
NR
NR
NR
86-89
86-89
84
84
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
Ref.
no.
18
19
19
5
5
5
20
20
4
4
6
8
8
10
10
10
12
12
12
12
12
12
Comments
samples composite 3-5 fish
samples composite 2-5 fish
composite 5 fish
composite 2-5 fish
composite 2-5 fish
composite 2-5 fish
samples composite 3-5 fish
samples composite 3-5 fish
Elk river power station
Elk river power station
former sewage sludge
composite 3 samples
composite3 samples
composite3 samples
composite 3 samples
compositeS samples
compositeS samples
B-51
-------
Table B-9. Environmental Levels of Dibenzofurans in Fish (ppt) (continued)
Chemical
PeCDFs (continued)
PeCDFs
other than
1,2,3,7,8-PeCDF
and
2,3,4,7,8-PeCDF
Fish
species*
Coho Salmon
Coho Salmon
Grayling
Barbel
Carp
Tissue
Whole
Fillets
Fillets
Fillets
Fillets
Number
samples
1
1
1
1
1
Number
positive
samples
1
0
1
1
1
Cone, range
13
ND(7)
22
21
17
Cone.
mean
NA
NA
NA
NA
NA
Wt.
basis
NR
NR
Fat
Fat
Fat
Location*
Lake Ontario
Lake Ontario
Necfcar river, Germany
Neckar river, Germany
Neckar river, Germany
Location
description
NR
NR
Urban
Urban
Urban
Samp,
year
NR
NR
88
88
88
Ref.
no.
12
12
2
2
2
Comments
composite3 samples
composite 3 samples
Hexachlorodibenzofurans(MW=374.87)
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
Carp
Pike
Pike
Herring
Herring
Herring
Various'
Pike
Pike
Herring
Herring
Herring
Various1
Pike
Pike
Various0
Whole
Muscle
Muscle
Whole
Whole
Whole
Mixed1
Muscle
Muscle
Whole
Whole
Whole
Mixed1
Muscle
Muscle
Mixed1
2
8
1
1
2
2
314
8
1
1
2
2
314
8
1
314
2
8
1
1
2
2
132
8
1
1
2
2
66
0
0
3
2-5
10-33
11
0.2
0.4-0.7
0.3
NR
5.6-22
5.6
0.1
0.4-0.8
0.2
NR
ND(3-6)
ND(6)
NR
3.5
14.4
NA
NA
0.55
NA
2.35
11.9
NA
NA
0.6
NA
1.74
NA
NA
1.22
NR
Fat
Fat
Fresh
Fresh
Fresh
Wet
Fat
Fat
Fresh
Fresh
Fresh
Wet
Fat
Fat
Wet
Lake Huron
Lake Vanern, Sweden
Hedesunda Bay, Sweden
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Various, US
Lake Vanem, Sweden
Hedesunda Bay, Sweden
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Various, US
Lake Vanern, Sweden
Hedesunda Bay, Sweden
Various, US
NR
Industrial
Industrial
Pristine
Urban
Industrial
Various
Industrial
Industrial
Pristine
Urban
Industrial
Various
Industrial
Industrial
Various
NR
88
88
NR
NR
NR
86-89
88
88
NR
NR
NR
86-89
88
88
86-89
18
19
19
5
5
5
20
19
19
5
5
5
20
19
19
20
samples composite 3-5 fish
samples composite 2-5 fish
composite 5 fish
composite 2-5 fish
composite 2-5 fish
composite 2-5 fish
samples composite 3-5 fish
samples composite 2-5 fish
composite 5 fish
composite 2-5 fish
composite 2-5 fish
composite 2-5 fish
samples composite 3-5 fish
samples composite 2-5 fish
compositeS fish
samples composite 3-5 fish
B-52
-------
Table B-9. Environmental Levels of Dibenzofurans La Fish (ppt) (continued)
Chemical
2,3,4,6,7,8-HxCDF
HxCDFs
Fish
species?
Pike
Pike
Herring
Herring
Herring
Various'
Bream
Perch
Y. Perch
Carp
Carp
Blue Crab
Lobster
Str. Bass
Br. Trout
Br. Trout
Kb. Trout
Kb. Trout
Lake Trout
Lake Trout
Coho Salmon
Coho Salmon
Tissue
Muscle
Muscle
Whole
Whole
Whole
Mixed1
Whole
Whole
Whole
Meat
Meat
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Number
samples
8
1
1
2
2
314
13
2
1
3
3
2
2
2
1
1
1
1
1
1
1
1
Number
positive
samples
7
0
0
2
2
100
13
2
1
2
3
2
2
2
1
0
1
1
1
0
0
0
Cone, range
ND(3)-17
ND(6)
ND(0.2)
0.4-0.8
0.3
NR
5.7-17.4
21.9-53.8
440
ND-24
2.7-5.1
9.3-9.4
7.7-7.9
2.0-4.4
2
ND<7)
8
2
16
ND(7)
ND(7)
ND(7)
Cone.
mean
7.96
NA
NA
0.6
NA
1.24
25.2
37.8
NA
10.0
3.5
9.35
7.8
3.2
NA
NA
NA
NA
NA
NA
NA
NA
Wt.
basis
Fat
Fat
Fresh
Fresh
Fresh
Wet
Fresh
Fresh
NR
NR
NR
Wet
Wet
Wet
NR
NR
NR
NR
NR
NR
NR
NR
Location"
Lake Vanern, Sweden
Hedesunda Bay, Sweden
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Various, US
Hamburg, Germany
Hamburg, Germany
Housatonic river, MA
Mississippi river, MN
Lake Orono, MN
Passaic river, NJ
New York Bight
Newark Bay, NJ
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Location
description
Industrial
Industrial
Pristine
Urban
Industrial
Various
Urban
Urban
NR
Industrial
Industrial
Urban
Dump Site
Urban
NR
NR
NR
NR
NR
NR
NR
NR
Samp.
year
88
88
NR
NR
NR
86-89
84
84
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
Ref.
no.
19
19
5
5
5
20
4
4
6
8
8
10
10
10
12
12
12
12
12
12
12
12
Comments
samples composite 2-5 fish
composite 5 fish
composite 2-5 fish
composite 2-5 fish
composite 2-5 fish
samples composite 3-5 fish
Elk river power station
Elk river power station
former sewage sludge
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
B-53
-------
Table B-9. Environmental Levels of Dibenzofurans in Fish (ppt) (continued)
Chemical
HxCDFs (continued)
non-2,3,7,8-HxCDFs
Fish
species*
Various'
Barbel
Tissue
Mixed*
Fillets
Number
samples
34
1
Number
positive
samples
MR
1
Cone, range
ND-2.59
2.1
Cone.
mean
0.22
NA
Wt.
basis
Wet
Fat
Location*
Various, US
Neckar river, Germany
Location
description
Background"
Urban
Samp.
year
86-89
88
Ref.
no.
20
2
Comments
samples composite 3-5 fish
Heptachlorodibenzofurans(MW=409.31)
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
HpCDFs
Catp
Pike
Pike
Herring
Herring
Herring
Various'
Pike
Pike
Various'
Bream
Perch
Y. Perch
Carp
Carp
Blue Crab
Lobster
Str. Bass
Br. Trout
Whole
Muscle
Muscle
Whole
Whole
Whole
Mixed1
Muscle
Muscle
Mixed"
Whole
Whole
Whole
Meat
Meat
Fillets
Whole
2
16
2
1
2
2
314
8
1
314
13
2
1
3
3
2
2
2
1
2
1
0
0
2
1
170
0
0
13
13
2
0
1
0
2
0
2
0
3-4
ND(3-7)-17
ND(6-11)
ND(0.2)
0.8-1.2
ND-0.9
NR
ND(3-11)
ND(6)
NR
1.8-6.0
5.0-10.1
ND(S.O)
ND-14
ND(6.6)
2.8-3.5
ND(0.9)
1.3-2.4
ND(7)
3.5
3.41
NA
NA
1.0
0.5
1.91
NA
NA
1.24
3.6
7.6
NA
4.7
NA
3.15
NA
1.8
NA
NR
Fat
Fat
Fresh
Fresh
Fresh
Wet
Fat
Fat
Wet
Fresh
Fresh
NR
NR
NR
Wet
Wet
Wet
NR
Lake Huron
Lake Vanern, Sweden
Hedesunda Bay, Sweden
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Various, US
Lake Vanern, Sweden
Hedesunda Bay, Sweden
Various, US
Hamburg, Germany
Hamburg, Germany
Housatonic river, MA
Mississippi river, MN
Lake Orono, MN
Passaic river, NJ
New York Bight
Newark Bay, NJ
Lake Ontario
NR
Industrial
Industrial
Pristine
Urban
Industrial
Various
Industrial
Industrial
Various
Urban
Urban
NR
Industrial
Industrial
Urban
Dump Site
Urban
NR
NR
88
88
NR
NR
NR
86-89
88
88
86-89
84
84
NR
NR
NR
NR
NR
NR
NR
18
19
19
5
5
5
20
19
19
20
4
4
6
8
8
10
10
10
12
samples composite 3-5 fish
samples composite 2-5 fish
composite 5 fish
composite 2-5 fish
composite 2-5 fish
composite 2-5 fish
samples composite 3-5 fish
samples composite 2-5 fish
composite 5 fish
samples composite 3-5 fish
Elk river power station
Elk river power station
former sewage sludge
composite 3 samples
B-54
-------
Table B-9. Environmental Levels of Dibenzofurans in Fish (ppt) (continued)
Chemical
HpCDFs (continued)
Fish
species'
Br. Trout
Rb. Trout
Rb. Trout
Lake Trout
Lake Trout
Coho Salmon
Coho Salmon
Tissue
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Number
samples
1
1
1
1
1
1
1
Number
positive
samples
0
1
0
1
0
0
0
Cone, range
ND(7)
1
ND(7)
1
ND(7)
ND<7)
ND(7)
Cone.
mean
NA
NA
NA
NA
NA
NA
NA
Wt.
basis
NR
NR
NR
NR
NR
NR
NR
Location*
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Location
description
NR
NR
NR
NR
NR
NR
NR
Samp.
year
NR
NR
NR
NR
NR
NR
NR
Ret
no.
12
12
12
12
12
12
12
Comments
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
OctachIorodibenzofijrans(MW=:444.76)
1,2,3,4,6,7,8,9-OCDF
Bream
Perch
Herring
Herring
Herring
Salmon
Salmon
Perch
Pike
Pike
Carp
Y. Perch
Carp
Carp
Whole
Whole
Whole
Muscle
Muscle
NR
Muscle
Muscle
Whole
Whole
Whole
Whole
14
3
1
2
2
2
2
3
8
1
2
1
3
3
10
3
0
1
0
0
0
1
0
0
2
0
0
3
ND-3.1
1.1-8.3
ND(0.2)
ND-0.3
ND(0.2)
ND(2.0)
ND(0.5)
ND-1.7
ND(3-11)
ND(ll)
4-8
ND(5.0)
ND(6.6)
ND(6.6)
1.2
3.7
NA
0.2
NA
NA
NA
0.57
NA
NA
6
NA
NA
NA
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fresh
Fat
Fat
NR
NR
NR
NR
Hamburg, Germany
Hamburg, Germany
Atlantic Coast, Sweden
Baltic Sea, Sweden
Gulf of Bothnia, Sweden
Gulf of Bothnia, Sweden
Gulf of Bothnia, Sweden
Gulf of Bothnia, Sweden
Lake Vanern, Sweden
Hedesunda Bay, Sweden
Lake Huron
Housatonic river, MA
Mississippi river, MN
Lake Orono, MN
Urban
Urban
Pristine
Urban
Industrial
NR
NR
NR
Industrial
Industrial
NR
NR
Industrial
Industrial
84
84
NR
NR
NR
NR
NR
NR
88
88
NR
NR
NR
NR
4
4
5
5
5
5
5
5
19
19
18
6
8
8
composite 2-5 fish
composite 2-5 fish
composite 2-5 fish
wild salmon
hatched salmon
caught near pulp mill
samples composite 2-5 fish
composite 5 fish
samples composite 3-5 fish
Elk river power station
Elk river power station
B-55
-------
Table B-9. Environmental Levels of Dibenzofurans in Fish (ppt) (continued)
Chemical
1,2,3,4,6,7,8,9-OCDF
(continued)
Fish
species*
Blue Crab
Lobster
Str. Bass
Lake Trout
Lake Trout
Lake Trout
Walleye
Walleye
Lake Trout
Lake Trout
Br. Trout
Br. Trout
Rb. Trout
Rb. Trout
Lake Trout
Lake Trout
Coho Salmon
Coho Salmon
Cod
Haddock
P. Flounder
Plaice
Tissue
Meat
Meat
Fillets
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Fillets
Fillets
Fillets
Fillets
Number
samples
2
2
2
1
1
3
1
1
1
10
1
1
1
1
1
1
1
1
4
1
1
1
Number
positive
samples
0
0
0
1
1
3
1
1
1
10
0
0
0
0
1
0
0
0
4
1
1
1
Cone, range
ND(8 3)
ND(8.4)
ND(3.1)
0.4
0.1
0.3-1.0
0.9
0.4
0.4
0.1-3.3
ND(7)
ND(7)
ND(7)
ND(7)
2
ND(7)
ND(7)
ND(7)
3.4-9.6
4.3
4.7
41
Cone.
mean
NA
NA
NA
NA
NA
0.85
NA
NA
NA
0.65
NA
NA
NA
NA
NA
NA
NA
NA
6.3
NA
NA
NA
Wt.
basis
Wet
Wet
Wet
Wet
Wet
Wet
Wet
Wet
Wet
Wet
NR
NR
NR
NR
NR
NR
NR
NR
Fresh
Fresh
Fresh
Fresh
Location1'
New York Bight
Newark Bay, NJ
Lake Superior
Lake Huron
Lake Michigan
Lake Erie
Lake St. Clair
Lake Ontario
Lake Michigan
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
various, Sweden
various, Sweden
various, Sweden
various, Sweden
Location
description
Urban
Dump Site
Urban
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
Industrial
Industrial
Industrial
Industrial
Samp.
year
NR
NR
NR
84
84
84
84
84
84
84
NR
NR
NR
NR
NR
NR
NR
NR
88
88
88
88
Ref.
no.
in
10
10
11
11
11
11
11
11
11
12
12
12
12
12
12
12
12
17
17
17
17
Comments
former sewage sludge
mean 5 samples
mean 5 samples
range 3 sample sites
mean 5 samples
mean 5 samples
mean 5 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 3 samples
composite 10 samples
composite 10 samples
composite 10 samples
B-56
-------
Table B-9. Environmental Levels of Dibenzofurans in Fish (ppt) (continued)
Chemical
1,2,3,4,6,7,8,9-OCDF
(continued)
Fish
species*
Flounder
Eel
Mussel
Shrimp
Cod
Tissue
Fillets
Fillets
Muscle
Muscle
Fillets
Number
samples
1
4
3
2
6
Number
positive
samples
1
4
3
2
NR
Cone, range
5.6
31-581
13.4-933
2.3-41
ND-21
Cone.
mean
NA
205.2
339.1
21.6
NR
Wt.
basis
Fresh
Fresh
Fresh
Fresh
Fresh
Location1"
various, Sweden
various, Sweden
Grenlandsfjord,Sweden
Grenlandsfjord.Sweden
Frierfjord, Sweden
Location
description
Industrial
Industrial
Industrial
Industrial
Industrial
Samp.
year
88
88
87
88
87
Ref.
no.
17
17
17
17
17
Comments
composite 10 samples
only cone, range given
Footnote References
* Br. = Brown; Smk. = Smoked; Str. = Striped; Lg. M. = Large Mouth; Rb. = Rainbow; P. = Pole; Y. = Yellow.
b Various, Netherlands = samples taken from six locations around Ilsselmeer Lake; various, Sweden = samples taken from Grenlandsfjord and Frierfjord, US = samples taken from 314 sites across the US, including industrial
and background sites.
' Species were taken from both bottom feeders and open water feeders, and then composited.
* Whole fish samples and fillet samples were combined during analysis.
* The 34 background sites are a subset of the 314 US sites.
NOTES: Summary statistics provided in or derived from references; means from references included non-detects counted as 0.0; when reference did not compute mean, it was computed with same strategy;
NA = not applicable
ND = non-detected (limit of detection)
NR = not reported
Descriptions provided were those given by reference or surmised from study description when not given.
Sources: 1. Van den Berg, et al. (1987)
2. Frommberger(1991)
3. Rappe, et al. (1984)
4. Gotz, et al. (1990)
5. Rappe, et al. (1989)
8. Reed, et al. (1990)
9. Gardner and White (1990)
10. Rappe, et al. (1991)
11. Vault, etal. (1989)
12. Niimi and Oliver (1989a)
17. Oehme, et al. (1989)
18. Stalling, et al. (1983)
19. Kjeller, et al. (1990)
20. USEPA(1992)
B-57
-------
Table B-10. Environmental Levels of PCBs in Fish (ppt)
Chemical
Fish
species*
Tissue
Number
samples
Number
positive
samples
Cone, range
Conc»
mean
Wt.
basis
Location11
Location
description
Samp,
year
Ref.
no.
Tetrachloro-PCB (MW=291.99)
3,3',4,4'-TeCB
3,4,4',5-TeCB
BIk. Bullhead
Lg. M. Bass
Blk. Crappie
Wh. Sucker
Coho Salmon
Wh. Crappie
Y. Perch
Br. Trout
Br. Trout
Lake Trout
Lake Trout
Rb. Trout
Rb. Trout
Coho Salmon
Coho Salmon
Carp
Br. Trout
Br. Trout
Lake Trout
Lake Trout
Rb. Trout
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Whole
Whole
Fillets
Whole
Fillets
Whole
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
2
1
2
2
0
1
1
1
1
2
89,000
86,000
43,000
50,000
2,000
24,000
23,000
5,000
2,000
18,000
8,000
6000-11,000
ND-4,000
8000-10,000
3,000-5,000
ND(5,000)
24,000
10,000
90,000
38,000
13,000-30,000
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
8,500
2,000
9,000
4,000
NA
NA
NA
NA
NA
21,500
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
Waukegan Harbor, IL
Waukegan Harbor, IL
Waukegan Harbor, IL
Waukegan Harbor, IL
Waukegan Harbor, IL
Waukegan Harbor, IL
Waukegan Harbor, IL
Vineland, Lake Ontario
Vineland, Lake Ontario
Port Credit, Lake Ontario
Port Credit, Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Saginaw Bay
Vineland, Lake Ontario
Vineland, Lake Ontario
Port Credit, Lake Ontario
Port Credit, Lake Ontario
Lake Ontario
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
78
78
78
78
78
78
78
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
13
13
13
13
13
13
13
15
15
15
15
15
15
15
15
16
15
15
15
15
15
Comments
composite 6 samples
composite 3 samples
composite 6 samples
composite 5 samples
mean 10 samples
mean 10 samples
mean 10 samples
mean 10 samples
mean 10 samples
mean 10 samples
mean 10 samples
mean 10 samples
B-58
-------
Table B-10. Environmental Levels of PCBs in Fish (ppt) (continued)
Chemical
3,4,4',5-TeCB
(continued)
TeCBs
Fish
species*
Rb. Trout
Coho Salmon
Coho Salmon
Carp
Various'
Tissue
Fillets
Whole
Fillets
Whole
Mixed"
Number
samples
2
2
2
I
362
Number
positive
samples
1
2
2
1
263
Cone, range
ND-9,000
2,000-26,000
8,000-14,000
17,000
MR
Conc»
mean
4,500
14,000
11,000
NA
696,240
Wt.
basis
NR
NR
NR
NR
Wet
Location1'
Lake Ontario
Lake Ontario
Lake Ontario
Saginaw Bay
Various, US
Location
description
NR
NR
NR
NR
Various
Samp.
year
NR
NR
NR
NR
86-89
Ref.
no.
15
15
15
16
20
Comments
samples composite 3-5 fish
Pentachloro-PCB (MW= 326.44)
3,3',4,4',5-PeCB
2,3,3',4,4'-PeCB
Carp
Blk. Bullhead
Lg. M. Bass
Blk. Crappie
Wh. Sucker
Coho Salmon
Wh. Crappie
Y. Perch
Small Smelt
Large Smelt
Salmonids
Br. Trout
Br. Trout
Lake Trout
Lake Trout
Rb. Trout
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Whole
Fillets
Whole
Fillets
Whole
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
ND(5,000)
352,000
290,000
114,000
483,000
45,000
242,000
80,000
15,000
38,000
110,000
55,000
24,000
253,000
101,000
34,000-138,000
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
86,000
NR
NR
NR
NR
NR
NR
NR
NR
Wet
Wet
Wet
NR
NR
NR
NR
NR
Saginaw Bay
Waukegan Harbor, IL
Waukegan Harbor, IL
Waufcegan Harbor, IL
Waukegan Harbor, IL
Waukegan Harbor, tt.
Waukegan Harbor, IL
Waukegan Harbor, IL
Port Credit, Lake Ontario
Vineland, Lake Ontario
Lake Ontario
Vineland, Lake Ontario
Vineland, Lake Ontario
Port Credit, Lake Ontario
Port Credit, Lake Ontario
Lake Ontario
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
78
78
78
78
78
78
78
86
82
81-82
NR
NR
NR
NR
NR
16
13
13
13
13
13
13
13
14
14
14
15
15
15
15
15
composite 6 samples
composite 3 samples
composite 6 samples
composite 5 samples
composite 48 samples
composite 20 samples
composite 60 samples
mean 10 samples
mean 10 samples
mean 10 samples
mean 10 samples
B-59
-------
Table B-10. Environmental Levels of PCBs in Fish (ppt) (continued)
Chemical
2,3,4',4,4'-PeCB
(continued)
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
PeCBs
Fish
species1
Rb. Trout
Coho Salmon
Coho Salmon
Carp
Carp
Small Smelt
Large Smelt
Salmonids
Br. Trout
Br. Trout
Lake Trout
Lake Trout
Rb. Trout
Rb. Trout
Coho Salmon
Coho Salmon
Carp
Various'
Tissue
Fillets
Whole
Fillets
Whole
Whole
Whole
Whole
Whole
Whole
Fillets
Whole
Fillets
Whole
Fillets
Whole
Fillets
Whole
Mixed*
Number
samples
2
2
2
1
1
1
1
1
1
1
1
1
2
2
2
2
1
362
Number
positive
samples
2
2
2
1
1
1
1
1
1
1
1
1
2
2
2
2
1
314
Cone, range
6,000-50,000
48,000-121,000
19,000-56,000
427,000
57,000
37,000
87,000
250,000
133,000
60,000
634,000
242,000
80,000-310,000
16,000-115,000
100,000-271,000
39,000-136,000
1.35x \(f
NR
Cone,
mean
28,000
84,500
37,500
NA
NA
NA
NA
NA
NA
NA
NA
NA
195,000
65,500
185,500
87,500
NA
564,700
Wt.
basis
NR
NR
NR
NR
NR
Wet
Wet
Wet
NR
NR
NR
NR
NR
NR
NR
NR
NR
Wet
Location1"
Lake Ontario
Lake Ontario
Lake Ontario
Saginaw Bay
Saginaw Bay
Port Credit, Lake Ontario
Vineland, Lake Ontario
Lake Ontario
Vineland, Lake Ontario
Vineland, Lake Ontario
Port Credit, Lake Ontario
Port Credit, Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Lake Ontario
Saginaw Bay
Various, US
Location
description
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
Various
Samp.
year
NR
NR
NR
NR
NR
86
82
81-82
NR
NR
NR
NR
NR
NR
NR
NR
NR
86-89
Hexachloro-PCB (MW=360,88)
3,3',4,4',5,5'-HxCB
2,3,3',4,4',5-HxCB
Carp
Small Smelt
Large Smelt
Whole
Whole
Whole
1
1
1
0
1
1
ND(5,000)
2,700
6,100
NA
NA
NA
NR
Wet
Wet
Saginaw Bay
Port Credit, Lake Ontario
Vineland, Lake Ontario
NR
NR
NR
NR
86
82
Ref.
no.
15
15
15
16
16
14
14
14
15
15
15
15
15
15
15
15
16
20
Comments
composite 48 samples
composite 20 samples
composite 60 samples
mean 10 samples
mean 10 samples
mean 10 samples
mean 10 samples
samples composite 3-5 fish
16
14
14
composite 48 samples
composite 20 samples
B-60
-------
Table B-10. Environmental Levels of PCBs in Fish (ppt) (continued)
Chemical
2,3,3',4,4',5-HxCB
(continued)
2,3,3',4,4',5'-HxCB
2,3',4,4',S,5'-HxCB
HxCBs
Fish
species*
Salmonids
Carp
Carp
Carp
Various0
Tissue
Whole
Whole
Whole
Whole
Mixed"
Number
samples
1
1
1
1
362
Number
positive
samples
1
1
1
1
321
Cone, range
34,000
79,000
76,000
77,000
NR
Cone,
mean
NA
NA
NA
NA
355,930
Wt.
basis
Wet
NR
NR
NR
Wet
Location*
Lake Ontario
Saginaw Bay
Saginaw Bay
Saginaw Bay
Various, US
Location
description
NR
NR
NR
NR
Various
Samp.
year
81-82
NR
NR
NR
86-89
Ref.
no.
14
16
16
16
20
Comments
composite 60 samples
samples composite 3-5 fish
Heptachloro-PCB (MW=396.33)
2,3,3',4,4',5,5'-HpCB
HpCBs
Carp
Various'
Whole
Mixedd
1
362
1
250
29,000
NR
NA
96,700
NR
Wet
Saginaw Bay
Various, US
NR
Various
NR
86-89
16
20
samples composite 3-5 fish
Footnote References
Blk. = Black; Lg. M. = Large Mouth; Wh. = White; Y. = Yellow; Br. = Brown; Rb. = Rainbow
b US = samples taken from 362 sites across the US, including industrial and background sites.
° Species were taken from both bottom feeders and open water feeders, and then composited.
' Whole fish samples and fillet samples were combined for analysis.
NOTES: Summary statistics provided in or derived from references; means from references included non-detects counted as 0.0; when reference did not compute mean, it was computed with same strategy;
NA = not applicable
ND = non-detected (limit of detection)
NR = not reported
Descriptions provided were those given by reference or surmised from study description when not given.
Sources: 13. Huckins, et al. (1988)
14. Oliver and Niimi (1988)
15. Niimi and Oliver (1989b)
16. Smith, et al. (1990)
20. USEPA(1991)
B-61
-------
Table B-ll. Levels of Dioxins in Food Products (ppt)
Chemical
Number
samples
Number
positive
samples
Cone, range
Cone.
mean
Wt. basis
Location
Location
description
Sample
year
Ref,
no.
Comments
Tetrachlorodibenzo-p-dioxins(MW=321.98)
2,3,7,8-TCDD
3
1
1
2
4
2
5
3
8
25
16
8
7
4
5
3
3
0
1
1
0
3
2
5
3
5
NR
0
7
NR
0
0
0
0
ND
1.9
1.4
ND
ND-0.049
0.08-0.2
0.01-0.6
2.8-23
ND-0.33
ND-1.9
ND
ND-0.19
ND-0.013
ND
ND
ND
ND
-
-
—
—
0.027
0.14
0.23
10
0.2'
NC
—
0.12
0.009*
-
—
—
—
Fresh
Fresh
Fresh
Whole Milk
Whole Milk
Fat
Fat
Fat
Fat
Fat
Fat
Wet
Whole Milk
Wet
Wet
Wet
Lipid
Sweden
Sweden
Sweden
Switzerland
Switzerland
Germany
Germany
Germany
Germany
Germany
Germany
Norwich, UK
England/Wales
USSR
USSR
Vietnam
Vietnam
Background
Near Incinerators
Rural
NR
1988
1988
NR
NR
NR
NR
NR
NR
NR
NR
NR
1989
1988, 1989
1988, 1989
NR
NR
1
1
1
2
2
3
3
3
4
5
5
7
7
8
8
8
8
Food basket; LOD=0.1-0.4
Fish, cooked
Fish, raw
Milk; LOD = 0.012/0.013
Milk; LOD = 0.013
Dairy Products
Meat and Eggs
Fish
Milk; LOD Not reported
Dairy Products; LOD = 0.5
Meat; LOD = 0.5
Fish; LOD = 0.16
Milk; LOD = 0.004
Dairy Products; LOD=0.03-0.53
Meat; LOD=0.03-0.72
Meat; LOD=0.34-0.99
Meat; LOD = 1
HexachIorodibenzo-p-dioxins(MW = 390.87)
HxCDDs
26
3
3
13
3
3
ND-67
0.81-4.95
2.4-5.2
27
2.8
3.4
Fat
Wet
Lipid
Canada
Vietnam
Vietnam
1980
NR
NR
6
8
8
Chicken; LOD = 5-20
Meat; Total for 2,3,7,8 substitutes
Meat; Total for 2,3,7,8 substitutes
B-62
-------
Table B-ll. Levels of Dioxins in Food Products (ppt) (continued)
Chemical
Number
samples
Number
positive
samples
Cone, range
Cone.
mean
Wt. basis
Location
Location
description
Sample
year
Ret
no.
Comments
Heptachlorodibenzo-p-dioxins(MW = 425.31)
HpCDDs
26
16
ND-142
52
Fat
Canada
1980
6
Chicken; LOD = 5-20
Octachlorodibenzo-p-dioxin(MW=460.76)
OCDD
3
1
1
2
4
2
5
3
8
25
16
26
8
7
4
5
3
3
3
1
1
2
4
2
5
3
0
25
16
12
8
7
4
5
3
3
1.0-2.1
0.72
0.34
.
-------
Table B-ll. Levels of Dioxins in Food Products (ppt) (continued)
"For ND values 1/2 LOD was used in calculating the mean.
bFor ND values the detection limit was used in calculating the mean.
c Signal to noise ratio S/N _<_ 3 (values reported as " <* are used as real values in calculating mean).
d Signal to noise ratio S/N=5/1 (values reported as *£" are used as real values in calculating mean)
NC = value could not be calculated
NR = not reported
NA = not applicable
Sources: 1. de Wit et at. (1990)
2. Rappe et al. (1987)
3. Becketal. (1989)
4. Becketal. (1987)
5. Furstetal. (1990)
6. Ryan et al. (1985)
7. Startin et al. (1990)
8. Schecter et al. (1990)
B-64
-------
Table B-12. Levels of Dibenzofurans in Food Products (ppt)
Chemical
Number
Samples
Number
Positive
Samples
Cone. Range
Cone*
Mean
Wt. Basis
Location
Location
Description
Samp,
Year
Ref.
No.
Comments
Tetrachlorodibenzo&rans (MW=305 .98)
2,3,7,8-TCDF
3
1
1
2
4
2
5
3
8
25
16
8
7
4
5
3
3
3
1
1
2
4
2
5
3
8
MR
MR
8
NR
4
3
3
3
0.1-0.4
10
7.6
<0.021-j<0.028
.
-------
Table B-12. Levels of Dibenzofurans in Food Products (ppt) (continued)
Chemical
HxCDFs
(continued)
Number
Samples
3
Number
Positive
Samples
3
Cone. Range
1.1-1.5
Cone.
Mean
1.3
Wt. Basis
Lipid
Location
Vietnam
Location
Description
Samp.
Year
NR
Ref.
No.
8
Comments
Meat products; Total for 2,3,7,8-
substitutes
Octachlorodibenzoftirans (MW=444.76)
OCDF
3
1
1
2
4
2
5
3
8
25
16
8
7
4
5
3
3
0
0
0
1
1
2
5
3
0
NR
NR
7
7
0
1
2
2
ND
ND
ND
ND-^.0.20
ND-_<0.52
0.25-1
0.2-0.6
0.3-2.1
ND
ND-4.3
ND-5.0
ND-0.26
0.023-0.071
ND
ND-0.06
ND-0.57
ND-0.6
-
-
-
O.lc
0.13°
0.63
0.34
1.27
—
NC
NC
0.18
0.041
-
0.06
0.34
0.3
Fresh
Fresh
Fresh
Whole milk
Whole milk
Fat
Fat
Fat
Fat
Fat
Fat
Wet
Whole
Wet
Wet
Wet
Lipid
Sweden
Sweden
Sweden
Switzerland
Switzerland
Germany
Germany
Germany
Germany
Germany
Germany
Norwich, UK
England/Wales
USSR
USSR
Vietnam
Vietnam
Background
Near Incinerator
Rural
NR
1988
1988
NR
NR
NR
NR
NR
NR
NR
NR
NR
1989
1988, 1989
1988, 1989
NR
NR
1
1
1
2
2
3
3
3
4
5
5
7
7
8
8
8
8
Food basket; LOD= 0.4-2.1
Fish, cooked; LOD=0.07
Fish, raw; LOD=0.07
Milk; LOD=0.09
Milk; LOD=0.13-0.21
Dairy Products
Meat and Eggs
Fish
Milk; LOD=1.0
Dairy products; LOD=0.3
Meat; LOD=0.3
Fish; LOD=0.30
Milk
Dairy products; LOD=0.02-0.26
Meat; LOD=0.02-0.36
Meat; LOD=0.3
Meat; LOD=0.3
B-66
-------
Table B-12. Levels of Dibenzofurans in Food Products (ppt) (continued)
'For ND values 1/2 LOD was used in calculating the mean.
'For ND values the detection limit was used in calculating the mean.
'Signal to noise ratio S/N <3 (values reported as '_<* are used as real values in calculating mean).
'Signal to noise ratio S/N 5/1 (values reported as '_<" are used as real values in calculating mean).
NC = value culd not be calculated.
NR = not reported.
NA = not applicable.
Sources: 1. de Wit et al. (1990)
2. Rappe et al. (1987)
3. Beck et al. (1989)
4. Beck et al. (1987)
5. Furstetal. (1990)
6. Ryan et al. (1985)
7. Startin et al. (1990)
8. Schecter et al. (1990)
B-67
-------
Table B-13. Environmental Levels of Dioxins in Air (pg/in3)
Chemical
Number
samples
Number
positive
samples
Concentration
range
Cone.
mean
Location
Location
description
Sample
year
Ref.
no.
Comments
Tetrachlorodibenzo-p-dioxins(MW =321.98)
2,3,7,8-TCDD
TCDDs
3
3
NR
1
7
28
NR
NR
1
2
2
3
2
1
1
16
16
7
27
1
2
2
3
2
1
1
0
0
NR
0
2
2
NR
NR
1
2
2
0
0
0
0
3
12
6
20
1
2
2
1
0
0
0
ND(.01-.02)
ND(.01-.02)
ND(0.01)
ND(0.0095)
ND-0.05
ND-0.004
ND(0.02)
ND(0.03)
0.0004
0.02-0.06
0.02-0.08
ND(.012-0.2)
ND(0.24-0.82)
ND(0.15)
ND(0.058)
ND-0.18
ND-10.12
ND-0.54
ND-0.07
0.05
0.10-0.22
0.21-1.5
ND-0.18
ND(0.24-0.82)
ND(0.15)
ND(0.058)
NA
NA
NA
NA
0.01
0.002
NA
NA
NA
0.04
0.05
NA
NA
NA
NA
0.04
0.99
0.20
0.03
NA
0.16
0.86
0.12
NA
NA
NA
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra FaUs, NY
Bridgeport, CT
Wallingford, CT
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Urban
Rural
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Urban
Rural
86-87
86-87
NR
87
87-88
88
NR
NR
89
NR
NR
87
87
87
87
86-87
86-87
87-88
88
89
NR
NR
87
87
87
87
1
1
2
3
4
5
6
6
7
8
8
9
9
9
9
1
1
4
5
7
8
8
9
9
9
9
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
B-68
-------
Table B-13. Environmental Levels of Dioxins in Air (pg/m3) (continued)
Chemical
Number
samples
Number
positive
samples
Concentration
range
Cone,
mean
Location
Location
description
Sample
year
Ref,
no.
Comments
Pentachlorodibenzo-p-dioxifls(MW=356.42) ;
1,2,3,7,8-PeCDD
PeCDDs
3
3
NR
1
7
28
NR
NR
1
2
2
3
2
1
1
16
16
7
28
1
2
2
3
2
1
1
0
1
NR
0
4
9
NR
NR
1
1
2
0
0
1
0
1
11
7
21
1
2
2
1
0
0
0
ND(.01-.02)
ND-0.49
ND(0.02)
ND(0.039)
ND-0.07
ND-0.02
ND(0.03)
ND(0.01)
0.006
ND-0.28
0.22-0.60
ND(.034-0.27)
ND(.047-.06)
ND(0.082)
ND(0.033)
ND-0.05
ND-11.16
0.01-0.66
ND-0.15
0.11
0.07-1.3
2.4-5.0
ND-0.10
ND(0.47-.06)
ND(0.082)
ND(0.033)
NA
0.17
NA
NA
0.02
0.006
NA
NA
NA
0.14
0.41
NA
NA
NA
NA
0.02
1.04
0.24
0.05
NA
0.68
3.70
0.097
NA
NA
NA
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra Falls, NY
Bridgeport, CT
Wallingford, CT
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Urban
Rural
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Urban
Rural
86-87
86-87
NR
87
87-88
88
NR
NR
89
NR
NR
87
87
87
87
86-87
86-87
87-88
88
89
NR
NR
87
87
87
87
1
1
2
3
4
5
6
6
7
8
8
9
9
9
9
1
1
4
5
7
8
8
9
9
9
9
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
B-69
-------
Table B-13. Environmental Levels of Dioxins in Air (pg/m3) (continued)
Chemical
Number
samples
Number
positive
samples
Concentration
range
Cone.
mean
Location
Location
description
Sample
year
Ref.
no.
Comments
Hexachlorodibenzo-p-dioxins (MW =390 . 87)
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
3
3
NR
1
7
28
NR
NR
1
2
2
3
2
1
1
3
3
NR
1
7
28
NR
NR
1
2
2
3
0
3
NR
0
4
8
NR
NR
1
0
2
3
0
0
1
1
3
NR
0
5
22
NR
NR
1
2
2
3
ND(.01-.02)
.04-.64
NR
ND(0.076)
ND-0.08
ND-0.03
NR
NR
0.004
ND(0.08-0.17)
0.19-1.0
.032-.055
ND(.028-.039)
ND(0.032)
0.031
ND-.03
.05-1.06
NR
ND(0.083)
ND-0.13
ND-0.06
NR
NR
0.008
0.23-0.66
0.71-2.2
.052-.053
NA
0.24
0.01
NA
0.03
0.01
0.05
0.01
NA
NA
0.60
0.041
NA
NA
NA
0.02
0.39
0.02
NA
0.04
0.02
0.07
0.01
NA
0.44
1.46
0.053
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
• Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Urban
Rural
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
86-87
86-87
NR
87
87-88
88
NR
NR
89
NR
NR
87
87
87
87
86-87
86-87
NR
87
87-88
88
NR
NR
89
NR
NR
87
1
1
2
3
4
5
6
6
7
8
8
9
9
9
9
1
1
2
3
4
5
6
6
7
8
8
9
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
B-70
-------
Table B-13. Environmental Levels of Dioxins in Air (pg/m3) (continued)
Chemical
1,2,3,6,7,8-HxCDD
(continued)
1,2,3,7,8,9-HxCDD
HxCDDs
Number
samples
2
1
1
3
3
MR
1
7
28
NR
MR
1
2
2
3
2
1
1
16
16
7
28
1
2
2
3
2
1
Number
positive
samples
1
0
1
1
2
NR
0
6
18
NR
NR
0
0
2
3
1
0
1
11
12
6
27
1
2
2
3
2
1
Concentration
range
ND-0.078
ND(0.032)
0.025
ND-.03
ND-0.11
NR
ND(0.086)
ND-0.25
ND-0.07
NR
ND(0.01)
ND(0.001)
ND(0.08-0.17)
0.36-5.2
.017-.050
ND-.064
ND(0.032)
0.025
ND-0.23
ND-12.16
ND-2.17
ND-0.68
0.10
0.74-2.7
5.3-24
0.6-0.63
0.43-0.78
0.15
Cone,
mean
0.046
NA
NA
0.02
0.06
0.02
NA
0.08
0.03
0.05
NA
NA
NA
2.78
0.031
0.039
NA
NA
0.08
1.69
0.72
0.26
NA
1.72
14.6
0.62
0.60
NA
Location
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra Falls, NY
Bridgeport, CT
Wallingford, CT
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Location
description
Industrial
Urban
Rural
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Urban
Rural
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Urban
Sample
year
87
87
87
86-87
86-87
NR
87
87-88
88
NR
NR
89
NR
NR
87
87
87
87
86-87
86-87
87-88
88
89
NR
NR
87
87
87
Ref.
no.
9
9
9
1
1
2
3
4
5
6
6
7
8
8
9
9
9
9
1
1
4
5
7
8
8
9
9
9
Comments
near incinerators
next to interstate highway
background site
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
B-71
-------
Table B-13. Environmental Levels of Dioxins in Air (pg/m3) (continued)
Chemical
HxCDDs (continued)
Number
samples
1
Number
positive
samples
1
Concentration
range
0.33
Cone.
mean
NA
Location
Waldo, OH
Location
description
Rural
Sample
year
87
Ref.
no.
9
Comments
background site
Heptachlorodibenzo-p-dioxins (MW = 425 3V)
1,2,3,4,6,7,8-HpCDD
HpCDDs
3
3
MR
1
7
28
NR
NR
1
3
2
1
1
16
15
7
28
1
2
2
3
2
1
1
3
2
NR
1
7
23
NR
NR
1
3
2
1
1
14
15
7
26
1
2
2
3
2
1
1
0.34-0.51
ND-5.43
NR
0.25
0.02-1.07
ND-0.73
NR
NR
0.10
0.52-0.57
0.26-0.52
0.32
0.24
ND-0.86
0.24-9.78
0.02-2.19
ND-1.48
0.20
0.6-3.4
5.3-15
1.00-1.10
0.41-1.00
0.56
0.48
0.41
2.0
0.11
NA
0.48
0.29
0.41
0.04
NA
0.54
0.39
NA
NA
0.44
2.60
1.02
0.61
NA
2.0
10.2
1.07
0.70
NA
NA
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Stockholm, Sweden
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra Falls, NY
Bridgeport, CT
Wallingford, CT
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Urban
Rural
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Urban
Rural
86-87
86-87
NR
87
87-88
88
NR
NR
89
87
87
87
87
86-87
86-87
87-88
88
89
NR
NR
87
87
87
87
1
1
2
3
4
5
6
6
7
9
9
9
9
1
1
4
5
7
8
8
9
9
9
9
downwind industrial complex
near incinerators
near incinerators
next to interstate highway
background site
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
Octachlorodibenzo-p-dioxin(MW=460.76)
OCDD
16
14
NR
1
7
16
15
14
NR
1
7
15
ND-5.79
0.39-8.88
NR
1.9
0.17-5.55
ND-29.5
1.14
2.94
0.30
NA
2.10
5.53
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Urban
Industrial
Urban
Urban
Urban
Urban
86-87
86-87
NR
87
87-88
88
1
1
2
3
4
5
downwind industrial complex
B-72
-------
Table B-13. Environmental Levels of Dioxins in Air (pg/m*) (continued)
Chemical
OCDD
(continued)
Number
samples
NR
NR
1
2
2
3
2
1
1
Number
positive
samples
NR
NR
1
2
2
3
2
1
1
Concentration
range
NR
NR
0.23
0.37-6.4
7.4-40
1.00-1.20
0.51-1.10
0.96
0.50
Cone.
mean
1.10
0.13
NA
3.38
23.7
1.13
0.80
NA
NA
Location
Rutland, VT
Durham, NC
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Location
description
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Urban
Rural
Sample
year
NR
NR
89
NR
NR
87
87
87
87
Ref.
no.
6
6
7
8
8
9
9
9
9
Comments
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
Footnote References
NOTES: Summary statistics provided in or derived from references; means from references included non-detects counted as 0.0; when reference did not compute mean, it was computed with same strategy;
NA = Not applicable;
ND = Non-detect;
NR = Not reported.
Sources:
1. Smith, etal. (1989).
2. Harless, et al. (1990).
3. Maisel and Hunt (1990).
4. Hunt and Maisel (1990).
5.CDEP(1988).
6. Harless and Lewis (1991).
7. Naf, et al. (1990).
8. Rappe and Kjeller (1987).
9. Edgerton, et al. (1989).
B-73
-------
Table B-14. Environmental Levels of Dibenzofurans in Air (pg/ui3)
Chemical
Number
samples
Number
positive
samples
Concentration
range
Cone.
mean
Location
Location
description
Sample
year
Ref.
no.
Comments
Tetrachlorodibenzoftirans (MW=305 .98)
2,3,7,8-TCDF
TCDFs
3
3
NR
1
7
28
NR
NR
2
2
3
2
1
1
16
16
7
28
1
2
2
3
2
1
1
3
3
NR
1
6
22
NR
NR
2
2
3
2
0
1
10
16
6
26
1
2
2
3
2
0
1
0.04-0.14
.28-3.81
NR
0.02
ND-.20
ND-0.10
ND(0.1)
NR
0.04-0.72
0.18-0.38
0.19-0.20
0.32-0.49
ND(0.13)
0.13
ND-0.66
0.18-17.45
ND-2.29
ND-0.86
0.33
0.36-6.2
3.3-4.9
0.99-1.50
1.90-3.80
ND(0.13)
0.89
0.09
1.47
0.03
NA
0.08
0.04
NA
0.03
0.38
0.28
0.20
0.40
NA
NA
0.23
3.25
0.86
0.38
NA
3.28
4.1
1.23
2.85
NA
NA
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra Falls, NY
Bridgeport, CT
Wallingford, CT
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Industrial
Urban
Rural
Urban
Industrial
Urban
Urban
Urban
Urban
Industrial
Industrial
Industrial
Urban
Rural
86-87
86-87
NR
87
87-88
88
NR
NR
NR
NR
87
87
87
87
86-87
86-87
87-88
88
89
NR
NR
87
87
87
87
1
1
2
3
4
5
6
6
8
8
9
9
9
9
1
1
4
5
7
8
8
9
9
9
9
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
Pentachlorodibenzofiirans (MW=340.42)
1,2,3,7,8-PeCDF
3
0
ND(0.01)
NA
Niagra Falls, NY
Urban
86-87
1
B-74
-------
Table B-14. Environmental Levels of Dibenzofurans in Air (pg/m5) (continued)
Chemical
1,2,3,7,8-PeCDF
(continued)
2,3,4,7,8-PeCDF
PeCDFs
Number
samples
3
MR
1
7
28
NR
MR
3
2
1
1
3
3
NR
1
7
28
NR
NR
1
1
2
3
2
1
1
16
16
Number
positive
samples
3
NR
1
6
9
NR
NR
3
2
0
1
0
2
NR
1
6
16
NR
NR
1
1
2
3
1
0
0
8
15
Concentration
range
0.03-0.61
NR
0.08
ND-0.10
ND-0.02
NR
NR
.026- .033
.032-.057
ND(0.036)
0.021
ND(.01-.02)
ND-1.92
NR
0.08
ND-0.16
ND-0.04
NR
NR
0.02
0.04
0.43-1.2
.032-.042
ND-.089
ND(0.036)
ND(0.033)
ND-0.41
ND-12.4
Cone.
mean
0.25
0.05
NA
0.03
0.01
0.03
0.01
0.029
0.044
NA
NA
NA
0.68
0.04
NA
0.05
0.02
0.20
0.01
NA
NA
0.82
0.036
0.050
NA
NA
0.12
2.08
Location
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra Falls, NY
Location
description
Industrial
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Urban
Rural
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Industrial
Urban
Rural
Urban
Industrial
Sample
year
86-87
NR
87
87-88
88
NR
NR
87
87
87
87
86-87
86-87
NR
87
87-88
88
NR
NR
89
NR
NR
87
87
87
87
86-87
86-87
Ref.
no.
1
2
3
4
5
6
6
9
9
9
9
1
1
2
3
4
5
6
6
7
8
8
9
9
9
9
1
1
Comments
downwind industrial complex
near incinerators
near incinerators
next to interstate highway
background site
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
downwind industrial complex
B-75
-------
Table B-14. Environmental Levels of Dibenzofurans in Air (pg/m3) (continued)
Chemical
PeCDFs
(continued)
Number
samples
7
28
1
2
2
3
2
1
1
Number
positive
samples
6
28
1
2
2
3
2
0
1
Concentration
range
ND-1.77
0.04-0.71
0.17
0.51-4.1
5-10
0.53-0.66
0.69-1.30
ND(0.036)
0.50
Cone.
mean
0.57
0.29
NA
2.30
7.5
0.59
1.00
NA
NA
Location
Bridgeport, CT
Wallingford, CT
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Location
description
Urban
Urban
Urban
Urban
Industrial
Industrial
Industrial
Urban
Rural
Sample
year
87-88
88
89
NR
NR
87
87
87
87
Ref.
no.
4
5
7
8
8
9
9
9
9
Comments
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
Hexachlorodibenz6funms(MW=374.87)
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
3
3
MR
1
7
28
MR
NR
3
2
1
1
3
3
NR
1
7
28
1
2
NR
1
6
21
NR
NR
3
2
0
1
1
3
NR
1
6
17
ND-0.06
ND-0.22
NR
0.15
ND-0.41
ND-0.13
NR
NR
.053-0.10
.060-0.27
ND(0.034)
0.098
ND-0.02
0.05-1.17
NR
0.25
ND-0.15
ND-0.07
0.02
0.11
0.03
NA
0.11
0.05
0.04
0.02
0.083
0.16
NA
NA
0.01
0.45
0.03
NA
0.04
0.03
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Urban
Rural
Urban
Industrial
Urban
Urban
Urban
Urban
86-87
86-87
NR
87
87-88
88
NR
NR
87
87
87
87
86-87
86-87
NR
87
87-88
88
1
1
2
3
4
5
6
6
9
9
9
9
1
1
2
3
4
5
downwind industrial complex
near incinerators
near incinerators
next to interstate highway
background site
downwind industrial complex
B-76
-------
Table B-14. Environmental Levels of Dibenzofurans in Air (pg/m3) (continued)
Chemical
1,2,3,6,7,8-HxCDF
(continued)
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
Number
samples
MR
NR
1
2
2
3
2
1
1
3
3
NR
1
7
28
NR
NR
1
2
2
3
2
1
1
3
3
NR
1
Number
positive
samples
NR
NR
1
2
2
3
2
0
1
0
1
NR
0
2
1
NR
NR
1
0
1
3
2
0
1
1
2
NR
0
Concentration
range
NR
NR
0.008
0.03-0.15
0.24-1.4
.048-.092
.092-0.19
ND(0.034)
0.014
ND(.01-.02)
ND-0.1
NR
ND(0.08)
ND-0.02
ND-0.003
ND(0.03)
ND(0.01)
0.0008
ND(.Ol-.OS)
ND-0.33
.020-.039
.038-0.12
ND(0.034)
0.097
ND-0.04
ND-2.17
NR
ND(0.08)
Cone.
mean
0.03
0.02
NA
0.09
0.82
0.065
0.14
NA
NA
NA
0.04
0.01
NA
0.01
0.003
NA
NA
NA
NA
0.17
0.032
0.079
NA
NA
0.02
0.76
ND(.Ol)
NA
Location
Rutland, VT
Durham, NC
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Location
description
Urban
Urban
Urban
Urban
Industrial
Industrial
Industrial
Urban
Rural
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Industrial
Urban
Rural
Urban
Industrial
Urban
Urban
Sample
year
NR
NR
89
NR
NR
87
87
87
87
86-87
86-87
NR
87
87-88
88
NR
NR
89
NR
NR
87
87
87
87
86-87
86-87
NR
87
Ref,
no.
6
6
7
8
8
9
9
9
9
1
1
2
3
4
5
6
6
7
8
8
9
9
9
9
1
1
2
3
Comments
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
downwind industrial complex
B-77
-------
Table B-14. Environmental Levels of Dibenzofurans id Air (pg/m3) (continued)
Chemical
2,3,4,6,7,8-HxCDF
(continued)
HxCDFs
Number
samples
7
28
NR
MR
1
2
2
3
2
1
1
16
16
7
28
1
2
2
3
2
1
1
Number
positive
samples
6
19
NR
NR
1
1
2
0
0
0
0
11
15
6
28
1
2
2
3
2
1
1
Concentration
range
ND-0.30
ND-0.10
ND(0.03)
ND(0.01
0.005
ND-0.05
0.21-0.8
ND(.005-.036)
ND(.012-.028)
ND(0.034)
ND(.008)
ND-0.58
ND-10.23
ND-2.15
0.03-1.57
0.08
0.18-1.1
2.2-9.5
0.56-0.70
0.37-1.20
0.10
0.51
Cone.
mean
0.09
0.04
NA
NA
NA
0.03
0.50
NA
NA
NA
NA
0.14
1.96
0.58
0.49
NA
0.64
5.85
0.62
0.78
NA
NA
Location
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra Falls, NY
Bridgeport, CT
Wallingford, CT
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Location
description
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Industrial
Urban
Rural
Urban
Industrial
Urban
Urban
Urban
Urban
Industrial
Industrial
Industrial
Urban
Rural
Sample
year
87-88
88
NR
NR
89
NR
NR
87
87
87
87
86-87
86-87
87-88
88
89
NR
NR
87
87
87
87
Ref.
no.
4
5
6
6
7
8
8
9
9
9
9
1
1
4
5
7
8
8
9
9
9
9
Comments
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
Heptachlorodibenzofurans (MW=409.3 1)
1,2,3,4,6,7,8-HpCDF
3
3
NR
1
7
1
3
NR
0
6
ND-0.15
0.26-5.43
NR
ND(0.2)
ND-0.54
0.05
2.08
0.08
NA
0.21
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Urban
Industrial
Urban
Urban
Urban
86-87
86-87
NR
87
87-88
1
1
2
3
4
downwind industrial complex
B-78
-------
Table B-14. Eavironmental Levels of Dibenzofurans in Air (pg/m3) (continued)
Chemical
1,2,3,4,6,7,8-HpCDF
(continued)
1,2,3,4,7,8,9-HpCDF
HpCDFs
Number
samples
28
NR
MR
1
3
2
1
1
NR
1
7
28
NR
NR
1
3
2
1
1
16
16
7
28
1
2
2
3
2
Number
positive
samples
22
NR
NR
1
3
2
1
1
NR
0
6
12
NR
NR
0
1
0
0
1
12
12
6
26
1
2
2
3
2
Concentration
range
ND-0 80
NR
NR
0.09
0.22-0.25
0.20-0.47
0.087
0.22
NR
ND(0.02)
ND-0.07
ND-0.30
ND(0.01)
ND(0.01)
ND(0.001)
ND-0.031
ND(.015-.028)
ND(0.013)
0.019
ND-0.43
ND-8.76
ND-1.0
ND-1.58
0.11
0.1-1.2
2-5
0.37-0.39
0.26-0.64
Cone.
mean
0 26
0.12
0.02
NA
0.24
0.34
NA
NA
0.01
NA
0.03
0.03
NA
NA
NA
0.020
NA
NA
NA
0.13
1.67
0.37
0.47
NA
0.65
3.5
0.38
0.45
Location
Wallingford CT
Rutland, VT
Durham, NC
Stockholm, Sweden
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Stockholm, Sweden
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Niagra Falls, NY
Niagra Falls, NY
Bridgeport, CT
Wallingford, CT
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Location
description
TIrhan
Urban
Urban
Urban
Industrial
Industrial
Urban
Rural
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Urban
Rural
Urban
Industrial
Urban
Urban
Urban
Urban
Industrial
Industrial
Industrial
Sample
year
RX
NR
NR
89
87
87
87
87
NR
87
87-88
88
NR
NR
89
87
87
87
87
86-87
86-87
87-88
88
89
NR
NR
87
87
Ref.
no.
<
6
6
7
9
9
9
9
2
3
4
5
6
6
7
9
9
9
9
1
1
4
5
7
8
8
9
9
Comments
near incinerators
near incinerators
next to interstate highway
background site
near incinerators
near incinerators
next to interstate highway
background site
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
B-79
-------
Table B-14. Environmental Levels of Dibenzofurans in Air (pg/m3) (continued)
Chemical
HpCDFs
(continued)
Number
samples
1
1
Number
positive
samples
1
1
Concentration
range
0.15
0.29
Cone.
mean
NA
NA
Location
Columbus, OH
Waldo, OH
Location
description
Urban
Rural
Sample
year
87
87
Ref,
no.
9
9
Comments
next to interstate highway
background site
Octachlorodibenzofurans (MW =444.76)
OCDFs
16
IS
NR
1
7
28
NR
NR
1
2
2
3
2
1
1
11
8
NR
1
6
18
NR
NR
1
0
2
3
I
0
1
ND-0.22
ND-3.38
NR
0.06
ND-0.56
ND-0.70
NR
NR
0.02
ND(.ll-l.O)
0.78-7.0
0.17-0.19
ND-0.21
ND(0.16)
0.077
0.09
0.62
0.07
NA
0.21
0.21
0.14
0.03
NA
NA
3.89
0.18
0.18
NA
NA
Niagra Falls, NY
Niagra Falls, NY
Greenbay, WI
Los Angeles, CA
Bridgeport, CT
Wallingford, CT
Rutland, VT
Durham, NC
Stockholm, Sweden
Hamburg, Germany
Hamburg, Germany
Akron, OH
Columbus, OH
Columbus, OH
Waldo, OH
Urban
Industrial
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Urban
Industrial
Industrial
Industrial
Urban
Rural
86-87
86-87
NR
87
87-88
88
NR
NR
89
NR
NR
87
87
87
87
1
1
2
3
4
5
6
6
7
8
8
9
9
9
9
downwind industrial complex
urban air & inside traffic tunnel
downwind incinerator & industrial complex
near incinerators
near incinerators
next to interstate highway
background site
Footnote References
NOTES: Summary statistics provided in or derived from references; means from references included non-detects counted as 0.0; when reference did not compute mean, it was computed with same strategy;
NA = not available;
NR = not reported;
ND = Non-detect.
Sources:
1. Smith, etal. (1989).
2. Harless, et al. (1990).
3. Maisel and Hunt (1990).
4. Hunt and Maisel (1990).
5. CDEP(1988).
6. Harless and Lewis (1991).
7. Naf, et al. (1990).
8. Rappe and Kjeller (1987).
9. Edgerton, et al. (1989).
B-80
-------
Table B-15. Mean Background Environmental Levels of Dioxins in Soil (ppt)
Chemical
Number
samples
Number
positive
samples
Concentration
range
! Arithmetic
E mean*
Geometric
mean*
Location
Ref. no.
'- Tetrachlorodjbenzo-ivdioxins(MW=32l,98}
2,3,7,8-TCDD
TCDDs
135
43
92
155
65
90
19
7
12
13
0
13
ND-7
ND-5
ND-7
ND-120
ND(1.0-55)
ND-120
0.71
0.88
0.62
17.4
22.6
13.6
0.46
0.81
0.36
12.4
14.2
11.3
World Wide
North America
Europe
World Wide
North America
Europe
1,6,7,10,11
1,6
7,10,11
1,4,6,7,10,11
1,4,6
7,10,11
Pentachlorodibenzo-p-dioxins (MW=356.42)
1,2,3,7,8-PeCDD
PeCDDs
81
4
77
155
65
90
O.NR
0
NR
14
1
13
ND-2.4
ND(3.75)
ND-2.4
ND-50
ND-38
ND-50
0.33
1.88
0.25
14.5
23.1
8.36
0.28
1.88
0.25
10.3
15.8
7.62
World Wide
North America
Europe
World Wide
North America
Europe
6,11
6
11
1,4,6,7,10,11
1,4,6
7,10,11
Hexachlorodibenzo-p-dioxins (MW =390.87)
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
HxCDDs
4
4
ND
4
4
ND
4
4
ND
155
65
90
0
0
ND
1
1
ND
2
2
ND
18
5
13
ND(3.75)
ND(3.75)
ND
ND-14
ND-14
ND
ND-9.9
ND-9.9
ND
ND-165
ND-99
ND-165
1.88
1.88
ND
4
4
ND
9
9
ND
31.8
26.5
35.6
1.88
1.88
ND
4
4
ND
9
9
ND
28.9
22.5
34.7
World Wide
North America
Europe
World Wide
North America
Europe
World Wide
North America
Europe
World Wide
North America
Europe
6
6
ND
6
6
ND
6
6
ND
1,4,6,7,10,11
1,4,6
7,10,11
B-81
-------
Table B-15. Mean Background Environmental Levels of Dioxins in Soil (ppt) (continued)
Chemical
Number
samples
Number
positive
samples
Concentration
range
Arithmetic
mean"
Geometric
mean*
Location
Ref. no.
Heptachlorodibenzo-p-dioxins (MW=425.3 1)
1,2,3,4,6,7,8-HpCDD
HpCDDs
4
4
ND
187
95
92
4
4
ND
28
15
13
37-360
37-360
ND
ND-640
ND-640
ND-234
194
194
ND
51.4
39.2
63.9
194
194
ND
34.9
20.1
61.6
World Wide
North America
Europe
World Wide
North America
Europe
6
6
ND
1,4,6,7,10,11,13
1,4,6,13
7,10,11
Octachlorodibenzo-p-dioxin (MW=460.76)
OCDD
TEQs
for
Dibenzodioxins
185
95
90
42
29
13
ND-10,600
ND-10,600
ND-832
205
237
171
4.51
5.48
0.92
96.1
59.9
158
4.12
5.24
0.64
World Wide
North America
Europe
World Wide
North America
Europe
1,4,6,7,10,11,13
1,4,6,13
7,10,11
Footnote References
* Means were taken from pristine, reference, residential, rural, and agricultural sites.
Industrial, urban, and dump sites were not used because they were assumed to be contaminated.
NOTES: One-half the limit of detection was used in calculating the means where applicable.
ND = non-detected (limit of detection)
Sources:
1. EPA (1985)
4. Pearson, et al. (1990).
6. Reed, et al. (1990).
7. Stenhouse and Badsha (1990).
10. Rappe and Kjeller (1987).
11. Creaser, et al. (1989).
B-82
-------
Table B-16. Mean Background Environmental Levels of Dibenzofurans in Soil (ppt)
Chemical
Number
samples
Number
positive
samples
Cone, range
Arithmetic
mean*
Geometric
Mean'
Location
Ref. no(s).
Tetrachlorodibenzofurans (MW=305.98)
2,3,7,8-TCDF
TCDFs
24
12
12
150
58
92
14
2
12
17
2
15
ND-50
ND-6
3-50
ND-300
ND-280
ND-300
9.29
1.59
17
42.2
54.2
34.5
4.85
1.39
17
33.8
43.0
29.1
World Wide
North America
Europe
World Wide
North America
Europe
1,6,7
1,6
7
1,4,6,7,10,11
1,4,6
7,10,11
Pentachlorodibenzofurans (MW=340.42)
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
PeCDFs
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
HxCDF
16
4
12
16
4
12
154
62
92
4
4
4
4
ND
4
4
ND
4
4
157
65
12
0
12
12
0
12
18
3
15
ND-10
ND(3.75)
1-10
ND-5
ND(3.75)
1-5
ND-185
ND-45
ND-185
3.47
1.88
4
1.97
1.88
2
24.5
25.8
23.7
3.31
1.88
4
1.97
1.88
2
22.0
20.2
23.4
World Wide
North America
Europe
World Wide
North America
Europe
World Wide
North America
Europe
6,7
6
7
6,7
6
7
1,4,6,7,10,11
1,4,6
7,10,11
Hexachlorodibenzofurans (MW=374.87)
0
0
0
0
ND
0
0
ND
1
1
19
4
ND(3.75)
ND(3.75)
ND(3.75)
ND(3.75)
ND
ND(3.75)
ND(3.75)
ND
ND-7.1
ND-7.1
ND-212
ND-150
1.88
1.88
1.88
1.88
ND
1.88
1.88
ND
2
2
33.2
26.5
1.88
1.88
1.88
1.88
ND
1.88
1.88
ND
2
2
27.2
17.7
World Wide
North America
World Wide
North America
Europe
World Wide
North America
Europe
World Wide
North America
World Wide
North America
6
6
6
6
ND
6
6
ND
6
6
1,4,6,7,10,11
1,4,6
B-83
-------
Table B-16. Mean Background Environmental Levels of Dibenzofurans in Soil (ppt) (continued)
Chemical
Number
samples
92
Number
positive
samples
15
Cone, range
ND-212
Arithmetic
mean*
37.9
Geometric
Mean*
36.8
Location
Europe
Ref. no(s).
7,10,11
Heptachlorodibenzofurans (MW=409.31)
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
HpCDFs
4
4
ND
4
4
ND
157
65
92
4
4
ND
0
0
ND
20
5
15
11-80
11-80
ND
ND(3.75)
ND(3.75)
ND
ND-260
ND-260
ND-138
47
47
ND
1.88
1.88
ND
26.8
29.3
25.0
47
47
ND
1.88
1.88
ND
23.9
22.7
24.8
World Wide
North America
Europe
World Wide
North America
Europe
World Wide
North America
Europe
6
6
ND
6
6
ND
1,4,6,7,10,11
1,4,6
7,10,11
Octachlorodibenzofurans (MW=444.76)
OCDFs
TEQs
for
Dibenzofurans
155
65
90
17
4
13
ND-270
ND-270
ND-144
28.4
30.2
27.2
3.37
2.48
2.93
25.2
23.1
26.9
2.91
2.45
2.93
World Wide
North America
Europe
World Wide
North America
Europe
1,4,6,7,10,11
1,4,6
7,10,11
Footnote References
* Means were taken from pristine, reference, residential, rural, and agricultural sites.
Industrial, urban, and dump sites were not used because they were assumed to be contaminated.
NOTES: One-half the limit of detection was used in calculating the means where applicable.
ND = non-detected (limit of detection)
Sources:
1. EPA (1985)
4. Pearson, et al. (1990).
6. Reed, et al. (1990).
7. Stenhouse and Badsha (1990).
10. Rappe and Kjeller (1987)
11. Creaser, et al. (1989).
B-84
-------
Table B-17. Mean Background Levels of Dioxins in Water (ppq)
Chemical
Number
samples
Number
positive
samples
Cone, range
Arithmetic
mean*
Geometric
mean*
Location
Ref.
no(s).
Tetrachlorodibenzo-p-dioxins (MW=321 .98)
2,3,7,8-TCDD
TCDDs
1
1
208
208
0
0
2
2
ND(0.7)
ND(0.7)
ND-40
ND-40
0.35
0.35
0.37
0.37
0.35
0.35
0.28
0.37
World Wide
North America
World Wide
North America
2
2
1,2
1,2
PentachIorodibenzo-p-dioxins{MW== 356.42)
1,2,3,7,8-PeCDD
PeCDDs
1
1
23
23
0
0
0
0
ND(l.O)
ND(l.O)
ND(1. 0-7.4)
ND(1. 0-7.4)
0.5
0.5
4.13
4.13
0.5
0.5
3.92
3.92
World Wide
North America
World Wide
North America
2
2
2
2
Hexachlorodibenzo-p-dioxins (MW=390.87)
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
HxCDDs
1
1
1
1
1
1
23
23
0
0
0
0
0
0
0
0
ND(1.8)
ND(1.8)
ND(1.5)
ND(1.5)
ND(1.5)
ND(1.5)
ND(0.4-4.7)
ND(0.4-4.7)
0.9
0.9
0.75
0.75
0.75
0.75
2.47
2.47
0.9
0.9
0.75
0.75
0.75
0.75
2.55
2.55
World Wide
North America
World Wide
North America
World Wide
North America
World Wide
North America
2
2
2
2
2
2
2
2
B-85
-------
Table B-17. Mean Background Levels of Dioxins in Water (ppq) (continued)
Chemical
Number
samples
Number
positive
samples
Cone, range
Arithmetie
mean"
Geometric
mean"
Location
Ref.
no(s).
Heptachlorodibenzo-p-dioxlns (MW=425.31)
1,2,3,4,6,7,8-HpCDD
HpCDDs
1
1
23
23
0
0
0
0
ND(2.8)
ND(2.8)
ND(0.4-6.8)
ND(0.4-6.8)
1.4
1.4
3.50
3.5
1.4
1.4
3.46
3.46
World Wide
North America
World Wide
North America
2
2
2
2
Octachlorodibenzo-p-dioxin (MW=460.76)
1,2,3,4,6,7,8,9-OCDD
TEQs
for
Dibenzodioxins
399
399
36
36
ND-175
ND-175
4.26
4.26
0.86
0.86
2.10
2.10
0.86
0.86
World Wide
North America
World Wide
North America
1
1
Footnote References
* Means were taken from non-contaminated sites.
NOTES: One-half the limit of detection was used in calculating the means where applicable.
ND = non-detected (limit of detection)
Sources: 1. Jobb, et al. (1990)
2. Meyer, et al. (1989)
B-86
-------
Table B-18. Mean Background Levels of Dibenzofurans in Water (ppq)
Chemical
Number
samples
Number
positive
samples
Cone, range
Arithmetic
mean*
Geometric
mean"
Location
Ref. no(s).
Tetrachlorodibenzofurans(MW=305.98)
2,3,7,8-TCDF
TCDFs
23
23
23
23
1
1
2
2
ND-1.2
ND-1.2
ND-18
ND-18
0.06
0.06
0.9
0.9
0.05
0.05
0.15
0.15
World Wide
North America
World Wide
North America
2
2
2
2
Pentachlorodibenzofarans (MW=340.42)
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
PeCDFs
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1
1
1
1
23
23
1
1
0
0
1
1
2.0
2.0
ND(l.O)
ND(l.O)
ND-27
ND-27
2.0
2.0
0.5
0.5
3.18
3.18
2.0
2.0
0.5
0.5
2.35
2.35
World Wide
North America
World Wide
North America
World Wide
North America
2
2
2
2
2
2
Hexachlorodibenzofurans (MW=374.87)
1
1
1
1
1
1
1
1
1
1
1
1
0
0
0
0
39
39
9.2
9.2
ND(1.3)
ND(1.3)
ND(1.2)
ND(1.2)
39
39
9.2
9.2
0.65
0.65
0.6
0.6
39
39
9.2
9.2
0.65
0.65
0.6
0.6
World Wide
North America
World Wide
North America
World Wide
North America
World Wide
North America
2
2
2
2
2
2
2
2
B-87
-------
Table B-18. Mean Background Levels of Dibenzofurans in Water (ppq) (continued)
Chemical
HxCDFs
Number
Samples
23
23
Number
positive
samples
1
1
Cone, range
ND-85
ND-85
Arithmetic
mean*
5.94
5.94
Geometric
mean"
2.75
2.75
Location
World Wide
North America
Ref. no(s).
2
2
Heptachlorodibenzofurans (MW=409.31)
1,2,3,4,6,7,8-HpCDF
HpCDFs
1,2,3,4,6,7,8,9-OCDF
TEQs
lor
Dibenzofurans
1
1
23
23
1
1
1
1
210
210
ND-210
ND-210
210
210
12.67
12.67
210
210
4.41
4.41
World Wide
North America
World Wide
North America
2
2
2
2
Octachlorodibenzofurans (MW=444.76)
23
23
3
3
ND-230
ND-230
10.67
10.67
7.41
7.41
0.9
0.9
7.40
7.40
World Wide
North America
World Wide
North America
2
2
Footnote References
* Means were taken from non-contaminated sites.
NOTES: One-half the limit of detection was used in calculating the means where applicable.
ND = non-detected (limit of detection)
Sources: 2. Meyer, et al. (1989)
B-88
-------
Table B-19. Mean Background Environmental Levels of Dioxins in Sediments (ppt)
Chemical
Number
samples
Number
positive
samples
Concentration
range
Arithmetic
mean*
(PPt)
Geometric
mean*
(PPt)
Location
Ret. no(s).
Tetrachlorodibenzo-p-dioxins (MW=321.98)
2,3,7,8-TCDD
TCDDs
25
5
20
27
7
20
2
0
2
18
2
16
ND-35
ND(4.7-12.7)
ND-35
ND-1400
ND-26
ND-1400
25.9
3.16
31.6
252
6.81
338
16.2
2.88
25.0
87.4
2.95
286
World Wide
North America
Europe
World Wide
North America
Europe
1,4,5,13
4,5
1,13
1,3,5,13
3,5
1,13
Pentachlorodibenzo-p-dioxins (MW=356.42)
PeCDDs
9
7
2
4
2
2
ND-100
ND-12
52-100
20.5
4.66
76
5.52
2.61
76
World Wide
North America
Europe
3,5,13
3,5
13
Hexachlorodibenzo-p-dioxins (MW=390.87)
HxCDDs
9
7
2
6
4
2
ND-170
ND-14
120-170
36.6
5.57
145
8.56
3.81
145
World Wide
North America
Europe
3,5,13
3,5
13
Heptachlorodibenzo-p-dioxins (MW=425.31)
HpCDDs
6
4
2
6
4
2
7.3-210
7.3-110
79-210
95.7
71
145
90.1
71
145
World Wide
North America
Europe
5,13
5
13
Octachlorodibenzo-p-dioxin (MW=460.76)
OCDD
27
12
ND-600
131
22.5
World Wide
1,3,5,13
B-89
-------
Table B-19. Mean Background Environmental Levels of Dioxins in Sediments (ppt) (continued)
Chemical
OCDD (continued)
TEQs
for
Dibenzodioxins
Number
samples
7
20
Number
positive
samples
7
5
Concentration
range
54-600
ND-250
Arithmetic
mean*
(PPt)
439
22.5
26.0
3.60
31.6
Geometric
mean*
(PPt)
364
8.51
16.2
3.24
25.0
Location
North America
Europe
World Wide
North America
Europe
Ref. no(s).
3,5
1,13
Footnote References
* Means were taken from pristine, reference, residential, rural, and various location sites.
Industrial, urban, and dump sites were not used because they were assumed to be contaminated.
NOTES: One-half the limit of detection was used in calculating the means where applicable.
ND = non-detected (limit of detection)
ppt = Parts per trillion
Sources:
1. Koistinen et al. (1990)
3. Czuczwa et al. (1984)
4. Norwood et al. (1989)
5. Reed et al. (1990)
13. Rappe et al. (1989a)
B-90
-------
Table B-20. Mean Background Environmental Levels of Dibenzofurans in Sediments (ppt)
Chemical
Number
samples
Number
positive
samples
Cone.
range
Arithmetic
mean*
Geometric
mean*
Location
Ref. no(s).
Tetrachlorodibenzofurans (MW=305.98)
2,3,7,8-TCDF
TCDFs
25
5
20
9
7
2
4
2
2
6
4
2
ND-35
ND-15
ND-35
ND-130
ND-18
87-130
26.7
3.06
32.6
28.0
4.86
109
11.7
0.23
31.2
2.21
0.72
109
World Wide
North America
Europe
World Wide
North America
Europe
1,4,5,13
4,5
1,13
3,5,13
3,5
13
Pentachlorodibenzoiurans (MW= 340.42)
PeCDFs
9
7
2
6
4
2
ND-125
ND-25
66-125
25.4
5.26
96
7.25
3.47
96
World Wide
North America
Europe
3,5,13
3,5
13
Hexachlorodibenzofurans (MW=374.87)
HxCDFs
9
7
2
5
3
2
ND-150
ND-12
78-150
27.1
2.31
114
4.55
1.81
114
World Wide
North America
Europe
3,5,13
3,5
13
Heptachlorodibenzofurans (MW=409.31)
HpCDFs
6
4
2
5
3
2
ND-180
ND-30
79-180
54.0
16
130
32.2
16
130
World Wide
North America
Europe
5,13
5
13
Octachforodibenzofurans (MW=444,76)
OCDF
27
8
ND-160
10.6
8.72
World Wide
1,3,5,13
B-91
-------
Table B-20. Mean Background Environmental Levels of Dibenzofurans in Sediments (ppt) (continued)
Chemical
OCDF (continued)
TEQs
for
Dibenzofurans
Number
samples
7
20
Number
positive
samples
4
4
Cone,
range
ND-23
ND-160
Arithmetic
mean*
4.50
12.8
2.68
0.31
3.27
Geometric
mean*
3.98
11.5
1.18
0.03
3.13
Location
North America
Europe
World Wide
North America
Europe
Ref. no(s).
3,5
1,13
Footnote References
* Means were taken from pristine, reference, residential, rural, and various location sites.
Industrial, urban, and dump sites were not used because they were assumed to be contaminated.
NOTES: One-half the limit of detection was used in calculating the means where applicable.
ND = non-detected (limit of detection)
ppt = Parts per trillion
Sources:
1. Koistinenet al. (1990)
3. Czuczwa et al. (1984)
4. Norwood et al. (1989)
5. Reed et al. (1990)
13. Rappeet al. (1989a)
B-92
-------
Table B-21. Mean Background Environmental Levels of PCBs in Sediment (ppt)
IUPAC
number Chemical
77 3,3'4,4'-TeCB
81 3,4,4'5-TeCB
Number
samples
Number
positive
samples
Concentration
range
Arithmetic
mean*
Geometric
mean*
Location
Tetrachloro-PCB (MW=291.99)
19
NR
18
NR
NR
13
NR
13
NR
NR
ND-550
ND(500)
ND-550
ND(500)
ND(500)
144
250
137.7
250
250
142
250
137.7
250
250
World Wide
North America
Europe
World Wide
North America
Pentachloro-PCB (MW=326.44)
126 3,3',4,4',5-PeCB
105 2,3,3',4,4'-PeCB
114 2,3,4,4',5-PeCB
118 2,3,4,4',5-PeCB
19
NR
18
49
39
10
NR
NR
38
1
NR
1
10
NR
10
NR
NR
NR
ND-110
ND(500)
ND-110
ND-10,000
NR
52-120
NR
NR
11,000-15,000
11,000-15,000
18.9
250
6.1
7893
9892
96.4
1000
1000
14,897
14,897
7.42
250
6.1
3835
9861
96.4
1000
1000
14,881
14,881
World Wide
North America
Europe
World Wide
North America
Europe
World Wide
North America
World Wide
North America
Ref. no(s).
1,16
16
1
16
16
1,16
16
1
1,7,16
7,16
1
16
16
7,16
7,16
Hexachloro-PCB (MW=360.88)
156 2,3,3',4,4',5-HxCB
167 2,3',4,4',5,5'-HxCB
169 3,3',4,4',5,5'-HxCB
38
R
NR
18
NR
NR
NR
0
1700-2100
1700-2100
ND(500)
ND(500)
ND(36-500)
2090
2090
250
250
30.21
2089
2089
250
250
20.67
World Wide
North America
World Wide
North America
World Wide
7,16
V6
16
16
1,16
B-93
-------
Table B-21. Mean Background Environmental Levels of PCBs in Sediments (ppt) (continued)
IUPAC
number Chemical
169 3,3',4,4',5,5'-HxCB
(continued)
Number
samples
NR
18
Number
positive
samples
NR
0
Concentration
range
ND(500)
ND(36)
Arithmetic
mean*
250
18
Geometric
mean*
250
18
Location
North America
Europe
Ref, ao(s).
16
1
Heptachloro-PCB (MW=396.33)
189 2,3>3',4,4',5,5'-HpCB
NR
NR
NR
NR
ND(500)
ND(500)
250
250
250
250
World Wide
Europe
16
16
Footnote References
* Means were taken from various location sites. Industrial and urban sites were not used because they were assumed to be contaminated.
NOTES: One-half the limit of detection was used in calculating the means where applicable.
NR = Not Reported
ND = Not Detected (detection limit)
ppt = Parts per trillion
Sources:
1. Koistinen et al. (1990)
7. Oliver and Nilmi (1988)
9. Huckins et al. (1988)
16. Smith et al. (1990)
B-94
-------
Table B-22. Mean Background Environmental Levels of Dioxins in Finfish (ppt)
Chemical
Number
samples
Number
positive
samples
Cone, range
Arithmetic
mean*
Geometric
mean*
Location
Ref. no(s).
Tetrachlorodibenzo-p-dioxins (MW=321 .98)
2,3,7,8-TCDD
35
34
1
34
34
0
ND-2.26
0.06-2.26
ND(O.l)
0.54
0.56
0.025
0.51
0.56
0.025
World Wide
North America
Europe
5,20
20
5
Pentachlorodibenzb-p-dioxins (MW=356.42)
1,2,3,7,8-PeCDD
35
34
1
35
34
1
0.15-2.67
0.15-2.67
0.3
0.76
0.77
0.3
0.75
0.77
0.3
World Wide
North America
Europe
5,20
20
5
Hexachlorodibenzo-p-dioxins (MW= 390,87)
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
HxCDDs
1
1
1
1
1
1
34
34
0
0
0
0
0
0
NR
NR
ND(0.2)
ND(0.2)
ND(0.2)
ND(0.2)
ND(0.2)
ND(0.2)
ND-3.57
ND-3.57
0.05
0.05
0.05
0.05
0.05
0.05
0.39
0.39
0.05
0.05
0.05
0.05
0.05
0.05
0.39
0.39
World Wide
Europe
World Wide
Europe
World Wide
Europe
World Wide
North America
5
5
5
5
5
5
20
20
Heptachlorodibenzo-p-dioxins (MW=425.31)
1,2,3,4,6,7,8-HpCDD
1
1
0
0
ND(0.2)
ND(0.2)
0.05
0.05
0.05
0.05
World Wide
Europe
5
5
Octachlorodibenzo-p-dioxin (MW=460.76)
1,2,3,4,6,7,8,9-OCDD
1
1
1
1
0.55
0.55
0.55
0.55
0.55
0.55
World Wide
Europe
5
5
B-95
-------
Table B-22. Mean Background Environmental Levels of Dioxins in Finfish (ppt) (continued)
Chemical
TEQs
for
Dibenzodioxins
Number
samples
Number
positive
samples
Cone, range
Arithmetic
mean*
0.94
0.94
0.19
Geometric
mean*
0.90
0.94
0.19
Location
World Wide
North America
Europe
Ref, no(s).
Footnote References
' Whole fish concentrations were divided in half to obtain the mean concentrations (USEPA, 1990; Branson et al., 1985).
NOTES: One-half the limit of detection was used in calculating the means where applicable.
ND = non-detected (limit of detection);
Sources: 5. Rappe, et al. (1989)
20. USEPA (1992)
B-96
-------
Table B-23. Mean Background Environmental Levels of Dibenzofurans in Finfish (ppt)
Chemical
Number
samples
Number
positive
samples
Cone, range
Arithmetic
mean*
Geometric
mean*
Location
Ref. no(s).
Tetrachlorodibenzofurans (MW=305.98)
2,3,7,8-TCDF
35
34
1
35
34
1
0.1-13.73
0.1-13.73
0.85
1.59
1.61
0.85
1.58
1.61
0.85
World Wide
North America
Europe
5,20
20
5
Pentachlorodibenzofurans (MW=340.42)
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
35
34
1
35
34
1
35
34
1
35
34
1
0.1-1.9
0.1-1.9
0.2
0.1-1.5
0.1-1.39
1.5
0.42
0.43
0.2
0.53
0.50
1.5
0.42
0.43
0.2
0.52
0.50
1.5
World Wide
North America
Europe
World Wide
North America
Europe
5,20
20
5
5,20
20
5
Hexachlorodibenzofiirans (MW =374.87)
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
HxCDFs
1
1
1
1
1
1
34
34
1
1
1
1
0
0
NR
NR
0.1
0.1
0.05
0.05
ND(0.2)
ND(0.2)
ND-2.59
ND-2.59
0.1
0.1
0.05
0.05
0.05
0.05
0.22
0.22
0.1
0.1
0.05
0.05
0.05
0.05
0.22
0.22
World Wide
Europe
World Wide
Europe
World Wide
Europe
World Wide
North America
5
5
5
5
5
5
20
20
B-97
-------
Table B-23. Mean Background Environmental Levels of Dibenzofurans in Finfish (ppt) (continued)
Chemical
Number
samples
Number
positive
samples
Cone, range
Arithmetic
mean1
Geometric
mean*
Location
Ref. no(s).
Heptachlorodibenzofurans (MW=409,31)
1,2,3,4,6,7,8-HpCDF
1
1
0
0
ND(0.2)
ND(0.2)
0.05
0.05
0.05
0.05
World Wide
Europe
5
5
Octachlorodibenzofurans (MW=444.76)
1,2,3,4,6,7,8,9-OCDF
TEQs
for
Dibenzofurans
1
1
0
0
ND(0.2)
ND(0.2)
0.05
0.05
0.47
0.43
0.87
0.05
0.05
0.46
0.43
0.87
World Wide
Europe
World Wide
North America
Europe
5
5
Footnote References
• Whole fish concentrations were divided in half to obtain the mean concentrations (USEPA, 1990; Branson et al., 1985).
NOTES: One-half the limit of detection was used in calculating the means where applicable.
ND = non-detected (limit of detection)
Sources: 5. Rappe, et al. (1989)
20. USEPA (1992)
B-98
-------
Table B-24. Mean Background Levels of Dioxins in Food Products (ppt)
Chemical
Number
samples
Number
positive
samples
Cone, range
Arithmetic
mean*
Geometric
mean*
Location
Ref. no(s).
Comments
Tetrachlorodibenzo-p-dioxins(MW=32I.98)
2,3,7,8-TCDD
3
13
31
17
32
0
12
2
5
11
ND(0. 1-0.4)
ND-23
ND-1.9
ND-0.33
ND-5.2
0.125
2.64
0.92
0.099
0.77
0.125
0.50
0.80
0.040
0.44
World Wide
World Wide
World Wide
World Wide
World Wide
1
1,3,7
3,5,8
2,4,7
3,5,8
Food basket
Fish
Dairy products
Milk
Meat
Hexachlorodibenzo-p-dioxins(MW = 390.87)
HxCDDs
26
6
13
6
ND-67
0.81-5.2
27
3.1
27
3.09
Canada
World Wide
6
8
Chicken
Meat
Heptachlorodibenzo-p-dioxins(MW = 425.31)
HpCDDs
26
16
ND-142
52
52
Canada
6
Chicken
Octachlorodibenzo-p-dioxin(MW=460.76)
OCDD
TEQs
for
Dibenzodioxins
3
13
31
17
26
32
3
13
31
9
12
32
1.0-2.1
0.34-83
0.66-35
ND-0.323
ND-238
0.63-122
1.47
10.46
9.82
2.46
90
20.27
0.13
2.65
0.93
0.10
1.47
3.76
9.45
0.92
90
17.5
0.13
0.50
0.81
0.04
World Wide
World Wide
World Wide
World Wide
Canada
World Wide
World Wide
World Wide
World Wide
World Wide
1
1,3,7
3,5,8
2,4,7
6
3,5,8
Food basket
Fish
Dairy products
Milk
Chicken
Meat
Food basket
Fish
Dairy products
Milk
B-99
-------
Table B-24. Mean Background Levels of Dioxins in Food Products (ppt) (continued)
Chemical
TEQs for
Dibenzofurans
(continued)
Number
samples
Number
positive
samples
Conc» range
Arithmetic
mean*
0.09
0.79
Geometric
mean*
0.09
0.46
Location
World Wide
World Wide
Ref. no(s).
Comments
Chicken
Meat
Footnote References
* Means were taken from all sites except near incinerators.
World Wide locations do not include North America
Notes: One-half the limit of detection was used in calculating the means where applicable.
ND = not detected (limit of detection)
Sources: 1. de Wit et al. (1990)
2. Rappe et al. (1987)
3. Becketal. (1989)
4. Beck et al. (1987)
5. Furstetal. (1990)
6. Ryan et al. (1985)
7. Startin et al. (1990)
8. Schecteretal. (1990)
B-100
-------
Table B-25. Mean Background Levels of Dibenzofurans in Food Products (ppt)
Chemical
Number
Samples
Number
Positive
Samples
ConC. Range
Arithmetic mean
Geometric
mean*
Location
Ref, no(s).
Comments
Tetrachlorodibenzofurans (MW=305 .98)
2,3,7,8-TCDF
3
13
31
17
32
3
13
6
10
11
0.1-0.4
0.14-98
ND-10.0
ND-1.4
ND-4.0
2.3
20.1
4.24
0.34
1.45
2.3
4.29
2.83
0.075
1.03
World Wide
World Wide
World Wide
World Wide
World Wide
1
1,3,7
3,5,8
2,4,7
3,5,8
Food basket
Fish
Dairy products
Milk
Meat
Hexachlorodibenzofiirans (MW=*374.87)
HxCDFs
4
11
47
10
0.024-2.60
ND-1.5
1.06
0.75
1.06
0.60
World Wide
World Wide
8
8
Dairy Products
Meat
Octachlorodibenzofurans(MW=444.76)
1,2,3,4,6,7,8,9-OCDF
TEQs
for
Dibenzofurans
3
13
31
17
32
0
10
2
8
10
ND(4.50)
ND-2.1
NEM.3
ND-0.2
ND-5.0
2.25
0.41
1.91
0.26
1.45
0.23
2.01
0.43
0.03
0.15
2.25
0.23
1.47
0.15
0.72
0.23
0.43
0.43
0.008
0.10
World Wide
World Wide
World Wide
World Wide
World Wide
World Wide
World Wide
World Wide
World Wide
World Wide
1
1,3,7
3,5,8
2,4,7
3,5,8
Food basket
Fish
Dairy products
Milk
Meat
Food basket
Fish
Dairy products
Milk
Meat
B-101
-------
Table B-25. Mean Background Levels of Dibenzofurans in Food Products (ppt) (continued)
Footnote References
* Means were taken from all sites except near incinerators.
World Wide locations do not include North America
Notes: One-half the limit of detection was used in calculating the means where applicable.
ND = not detected (limit of detection)
Sources: 1. de Wit et al. (1990)
2. Rappe et al. (1987)
3. Becketal. (1989)
4. Beck et al. (1987)
S. Furst et al. (1990)
7. Startin et al. (1990)
8. Schecter et al. (1990)
B-102
-------
Table B-26. Mean Background Environmental Levels of Dioxins in Air (pg/m3)
Chemical
2,3,7,8-TCDD
TCDDs
1,2,3,7,8-PeCDD
PeCDDs
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
HxCDDs
Number
samples
41
40
1
52
51
1
Number
positive
samples
Concentration
range
Arithmetic
mean*
Geometric
mean*
Location
Tetrachlorodibenzo-p-dioxins (MW=32t.98)
5
4
1
30
29
1
ND-0.05
ND-0.05
0.0004
ND-0.54
ND-0.54
0.05
0.005
0.005
0.0004
0.06
0.06
0.05
0.003
0.003
0.0004
0.04
0.04
0.05
World Wide
North America
Europe
World Wide
North America
Europe
Ref. no.
1,3,4,5,7,9
1,3,4,5,9
7
1,4,5,7,9
1,4,5,9
7
Pentachlorodibenzo-p-dioxins (MW=356,42)
41
40
1
53
52
1
41
40
1
41
40
1
41
40
1
53
52
1
14
13
1
30
29
1
ND-0.07
ND-0.07
0.006
ND-0.66
ND-0.66
0.11
0.01
0.01
0.006
0.07
0.07
0.11
0.008
0.008
0.006
0.05
0.05
0.11
World Wide
North America
Europe
World Wide
North America
Europe
1,3,4,5,7,9
1,3,4,5,9
7
1,4,5,7,9
1,4,5,9
7
Hexachlorodibenzo-p-dioxins (MW =390.87)
14
13
1
30
29
1
26
26
0
46
45
1
ND-0.08
ND-0.08
0.004
ND-0.13
ND-0.13
0.008
ND-0.25
ND-0.25
ND(0.001)
ND-2.17
ND-2.17
0.1
0.01
0.02
0.004
0.02
0.02
0.008
0.04
0.04
0.0005
0.26
0.27
0.1
0.01
0.01
0.004
0.02
0.02
0.008
0.03
0.03
0.005
0.21
0.21
0.1
World Wide
North America
Europe
World Wide
North America
Europe
World Wide
North America
Europe
World Wide
North America
Europe
1,3,4,5,7,9
1,3,4,5,9
7
1,3,4,5,7,9
1,3,4,5,9
7
1,3,4,5,7,9
1,3,4,5,9
7
1,4,5,7,9
1,4,5,9
7
B-103
-------
Table B-26. Mean Background Environmental Levels of Dioxins in Air (pg/m3) (continued)
Chemical
Number
samples
Number
positive
samples
Concentration
range
Arithmetic
mean*
Geometric
mean*
Location
Ref. no.
Heptachlorodibenzo-p-dioxins (MW=425.31)
1,2,3,4,6,7,8-HpCDD
HpCDDs
41
40
1
53
52
1
36
35
1
49
48
1
ND-1.07
ND-1.07
0.1
ND-2.19
ND-2.19
0.2
0.32
0.33
0.1
0.60
0.61
0.2
0.31
0.32
0.1
0.58
0.59
0.2
World Wide
North America
Europe
World Wide
North America
Europe
1,3,4,5,7,9
1,3,4,5,9
7
1,4,5,7,9
1,4,5,9
7
Octachlorodibenzo-p-dioxin (MW=460.76)
OCDD
TEQs
for
Dibenzodioxins
42
41
1
40
39
1
ND-29.5
ND-29.5
0.23
2.95
3.02
0.23
0.023
0.024
0.0059
2.20
2.33
0.23
0.018
0.019
0.0059
World Wide
North America
Europe
World Wide
North America
Europe
1,3,4,5,7,9
1,3,4,5,9
7
Footnote References
* Means were taken from urban sites, and pristine sites.
Industrial, sites were not used because they were assumed to be contaminated.
NOTES: One-half the limit of detection was used in calculating the means where applicable.
ND = non-detected (limit of detection)
Sources:
1. Smith, etal. (1989).
3. Maisel and Hunt (1990).
4. Hunt and Maisel (1990).
5. CDEP (1988).
7. Naf, et al. (1990).
9. Edgerton, et al. (1989).
B-104
-------
Table B-27. Mean Background Environmental Levels of Dibenzofurans in Air (pg/m3)
Chemical
Number
samples
Number
positive
samples
Cone.
range
Arithmetic
mean*
Geometric
Mean*
Location
Ref. no(s).
Tetrachlorodibenzofurans (MW=305.98)
2,3,7,8-TCDF
TCDFs
40
40
53
52
1
33
33
44
43
1
ND-0.2
ND-0.2
ND-2.29
ND-2.29
0.33
0.05
0.05
0.41
0.41
0.33
0.05
0.05
0.37
0.37
0.33
World Wide
North America
World Wide
North America
Europe
1,3,4,5,9
1,3,4,5,9
1,4,5,7,9
1,4,5,9
7
Pentachlorodibenzofurans (MW=340.42)
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
PeCDFs
40
40
41
40
1
53
52
1
17
17
24
23
1
44
43
1
ND-0.10
ND-0.10
ND-0.16
ND-0.16
0.02
ND-1.77
ND-1.77
0.17
0.02
0.02
0.03
0.03
0.02
0.28
0.28
0.17
0.01
0.01
0.02
0.02
0.02
0.24
0.24
0.17
World Wide
North America
World Wide
North America
Europe
World Wide
North America
Europe
1,3,4,5,9
1,3,4,5,9
1,3,4,5,7,9
1,3,4,5,9
7
1,4,5,7,9
1,4,5,9
7
Hexachlorodibenzofiirans (MW=374.87)
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
HxCDFs
40
40
41
40
1
41
40
1
41
40
1
53
52
1
30
30
27
26
1
5
4
1
27
26
1
47
46
1
ND-0.41
ND-0.41
ND-0.25
ND-0.25
0.008
ND-0.097
ND-0.097
0.0008
ND-0.30
ND-0.30
0.005
ND-2.15
ND-2.15
0.08
0.06
0.06
0.03
0.03
0.008
0.008
0.009
0.0008
0.05
0.05
0.005
0.40
0.39
0.08
0.06
0.06
0.03
0.03
0.008
0.004
0.005
0.0008
0.04
0.04
0.005
0.33
0.34
0.08
World Wide
North America
World Wide
North America
Europe
World Wide
North America
Europe
World Wide
North America
Europe
World Wide
North America
Europe
1,3,4,5,9
1,3,4,5,9
1,3,4,5,7,9
1,3,4,5,9
7
1,3,4,5,7,9
1,3,4,5,9
7
1,3,4,5,7,9
1,3,4,5,9
7
1,4,5,7,9
1,4,5,9
7
B-105
-------
Table B-27. Mean Background Environmental Levels Dibenzofurans in Air (pg/m3) (continued)
Chemical
Number
samples
Number
positive
samples
Cone.
range
Arithmetic
mean*
Geometric
Mean*
Location
Ref. no(s).
Heptachlorodibenzofurans (MW=409.31)
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
HpCDFs
41
40
1
38
37
1
53
52
1
31
30
1
19
19
0
46
45
1
ND-0.80
ND-0.80
0.09
ND-0.30
ND-0.30
ND(O.OOl)
ND-1.58
ND-1.58
0.11
0.23
0.23
0.09
0.03
0.03
0.0005
0.34
0.35
0.11
0.21
0.22
0.09
0.03
0.03
0.0005
0.30
0.30
0.11
World Wide
North America
Europe
World Wide
North America
Europe
World Wide
North America
Europe
1,3,4,5,7,9
1,3,4,5,9
7
3,4,5,7,9
3,4,5,9
7
1,4,5,7,9
1,4,5,9
7
Octachlorodibenzofurans (MW=444.76)
OCDFs
TEQs
for
Dibenzofurans
54
53
1
38
37
1
ND-0.70
ND-0.70
0.02
0.17
0.17
0.02
0.039
0.039
0.012
0.15
0.16
0.02
0.031
0.032
0.012
World Wide
North America
Europe
World Wide
North America
Europe
1,3,4,5,7,9
1,3,4,5,9
7
Footnote References
* Means were taken from urban and pristine sites.
Industrial, sites were not used because they were assumed to be contaminated.
NOTES: One-half the limit of detection was used in calculating the means where applicable.
ND = non-detected (limit of detection)
Sources:
1. Smith, etal. (1989).
3. Maisel and Hunt (1990).
4. Hunt and Maisel (1990).
5. CDEP (1988).
7. Naf, et al. (1990).
9. Edgerton, et al. (1989).
B-106
-------
DRAFT-DO NOT QUOTE OR CITE
APPENDIX C. SPREADSHEET ANALYSIS AND MODEL PARAMETERS
All the results given for the example settings in Chapter 9, with the exception of
results associated with the Industrial Source Complex model (ISC model; see Chapter 6),
were generated using spreadsheets. The purpose of this appendix is to describe the
structure of the spreadsheets and to list all parameters and their values.
C.I. INTRODUCTION
Four separate spreadsheets were developed for this document. Each one
corresponds to one of four source categories that are described in Chapter 5. The four
source categories were termed on-site soil, off-site soil, incinerator stack emissions, and
incinerator ash disposal in a landfill. These categories roughly translate to beginning
points, or origins, of contamination. All spreadsheets estimate exposure, i.e., lifetime
average daily doses, LADD in mg/kg-day, for all exposure pathways considered in this
methodology. What is unique, therefore, for each spreadsheet are the methods used to
estimate exposure media concentrations. Each spreadsheet was used to model "central"
and "high end" exposure scenarios associated with each source category. This meant that
generating results for the high end scenario required editing several parameters after
generating results for the central scenario. All spreadsheets also had both a pond and a
stream as possible sources of drinking water and fish consumption.
Section C.2 describes the structure of the spreadsheets, and Section C.3 provides
tables which list all parameters, their values, and sections of the document where further
discussion of the parameters appear.
C.2. SPREADSHEET STRUCTURE
Each spreadsheet has three principal sections. The first section describes the
results of the assessment, which are the LADDs for each exposure pathway. The second
section lists all the parameters: those associated with estimating exposure media
concentrations, those associated with patterns of exposure, and those specific to
chemicals. The third section contains the equations which solve for exposure media
concentrations using methodologies given in Chapter 5. Example portions of these three
sections for the on-site source category are given in Figure C-1.
C-1 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
»»»»**•»»»**•»»»<
ON-SITE SOURCE CATEGORY
*»»»**»*»**»»»»**»»»»»*»»»
I. RESULTS
A. LADD in mg/kg-day
1. Soil ingestion
2. Soil dermal
Residence scenario
Farm scenario
3. Vegetable ingestion
Dioxin
8E-13
4E-14
4E-13
2E-14
Furan
8E-13
4E-14
4E-13
2E-14
PCB
8E-13
4E-14
4E-13
2E-14
PARAMETERS
A. FATE AND TRANSPORT PARAMETERS
1. Site characteristics
* Cs: Soil concentration, mg/kg
* ASC: Area of contamination, m2
1 .OOE-06 1 .OOE-06 1 .OOE-06
4.00E + 04 4.00E + 04 4.00E + 04
B. EXPOSURE AND RISK PARAMETERS
1. Durations and body weights
* EDa: Exposure duration, adult, yrs
2.00E + 01 2.00E + 01 2.00E + 01
C. CHEMICAL PARAMETERS
* Koc: Org. carbon part, coef., L/kg
2.69E + 06 5.13E + 06 3.163E + 07
EXPOSURE MEDIA CONCENTRATIONS
A. Vapor and paniculate air cone.
* DEA: Effective diffusivity, cm2/sec
FLUX: Average emission flux, g/cm2-sec
Ambient on-site concentration, //g/m3
3.97E-02 3.97E-02 3.97E-02
1.71E-21 6.86E-22 3.91E-21
4.38E-11 1.75E-11 9.96E-11
Figure C-1. Portions of the on-site source category spreadsheet.
C-2
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
What is also seen in Figure C-1 are features of the columnar structure of the
spreadsheet. The first column, column A, is wide and describes the contents in columns
B, C, and D. Columns B through D contain parameters and quantities specific to the three
example compounds of the assessment. Values in Column B not tied to a specific
chemical, such as the parameters associated with estimating exposure media
concentrations, the exposure parameters, and others, are also repeated in Columns C
and D. In this way, each of Columns B, C, and D contain the complete information to
uniquely estimate exposures for each of the example compounds. Each column could also
be used to develop distinct scenarios with this structural strategy.
Finally, Figure C-1 also shows the numeric format of the spreadsheets. All numbers
are in scientific notation. Dose results give only one significant figure while all other
quantities give three significant figures. Giving more than one significant figure for dose
estimates would give the inappropriate impression of accurracy.
C.3. PARAMETER VALUES
The second section of each spreadsheet contains all parameters needed to generate
exposure media concentrations and exposure estimates. That section is further subdivided
into three categories of parameters: exposure media concentration parameters, exposure
parameters, and chemical parameters. The exposure media parameters were identified and
discussed in Chapter 3. Table C-1 lists the parameter values for each parameter and
example setting. Chapter 3 describes the equations in which these parameter appears,
and also the justification for the selection of the values shown in this table. The final
column in Table C-1 lists the section in Chapter 3 where this information can be found.
Table C-2 lists the parameters for the three example compounds demonstrated in
Chapter 9. As discussed in that chapter, the example compounds include 2,3,7,8-TCDD,
2,3,4,7,8-PCDF, and 2,3,3',4,4',5,5'-PCB. Background information on the values selected
for these compounds can be found in Chapters 2, 5, 6, and 9. Specific sections in these
chapters are noted in Table C-2.
Table C-3 lists all the exposure-related parameters for all exposure pathways.
Information on these parameters can be found principally in Chapter 7, and sections in that
Chapter are noted on Table C-3. These parameters are independent of the example
scenarios, with one exception. The exposure duration of 20 years was specific to a
C-3 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
"farmer" who was the impacted individual in the high end scenarios. The "residence"
setting exposed individual was called an adult, and has an exposure duration of 9 years.
This assumption is noted on Table C-3.
C-4 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table C-1. Exposure media concentration parameters for the four source categories
Description13
Source category8
Chapter 5
sections
1. Contaminated site characteristics
ASC: area, m2
Psoil: particle bulk den, g/cm
OCsl: organic carbon fraction
ESLP: soil porosity
NA
NA
NA
NA
40000
2.65
0.01
0.50
NA
NA
NA
NA
270000
2.65
0.01
0.50
4.1, 6.1
3.2
3.1
3.2
2. Exposure site parameters
AES: area, "central", m2
AES: area, "high end", m2
Bsoil: soil bulk den, g/m3
Psoil: particle bulk den, g/cm3
dt: tillage mixing depth, m
dnot: no-till mixing depth, m
OCsl: organic carbon fraction
ESLP: soil porosity
Um: wind speed, m/sec
DL: distance to soil cont., m
4000
40000
NA
2.65
NA
NA
0.01
0.50
4.00
NA
4000
40000
1500
NA
0.20
0.01
0.01
NA
4.00
150
4000
40000
1500
NA
0.20
0.01
0.01
NA
NA
NA
4000
40000
1500
NA
0.20
0.01
0.01
NA
4.00
150
3.1,4.1.,
& Chap 9
4.1
3.2
4.1, 5.0
4.1, 5.0
3.1
3.2
3.2, 6.2
4.1
(continued on the following page)
C-5
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table C-1. (continued)
Source category8
Description13
3. Soil and Sediment Delivery Parameters
SLs: Soil loss, cont. site, kg/ha-yr
Cw: Watershed cont. cone., all chem, ppm
OCssed: Fraction org. car., sus. sed
OCsed: Fraction org. car., bot. sed
Aw: Watershed drainage area, ha:
SDw: Watershed sed. delivery ratio
SLw: Soil loss, watershed, kg/ha-yr
MXdep: Watershed mixing depth, m
TSS: Total suspended sediment, mg/L
DL: Distance, cont. site to water, m
4. Air modeling parameters
z: height of individual, m
V: fraction of veg. cover, central
V: fraction of veg. cover, high
Ut: threshold wind speed, m/sec
F(x): model-specific parameter
FREQ: frequency wind blows to site
1
1 1 8800
1 E(-6)
0.05
0.02
4000
0.15
11880
NA
10
150
2
0.90
0.50
6.50
1.05
NA
2
1 1 8800
0
0.05
0.02
4000
0.15
11880
NA
10
150
2
0.00
0.00
8.25
0.50
0.15
3
NA
NA
0.05
0.02
NA
NA
NA
0.10
10
NA
NA
NA
NA
NA
NA
NA
4
1 1 8800
0
0.05
0.02
4000
0.15
11880
NA
10
150
2
0.00
0.00
8.25
0.50
0.15
Chapter 5
sections
3.1
3.1, 4.0
3.1
3.1
3.1
3.1
3.1
3.1
3.1
3.1
3.2
3.3, 4.0
3.3, 4.0
3.3, 4.0
3.3, 4.0
4.2
(continued on the following page)
C-6
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table C-1. (continued)
Source category8
Description6
4. Air modeling parameters (continued)
kunp: particle size multiplier for unpaved roads
s: silt cont. of unpaved roads, %
W: mean vehicle wt., kg
VS: mean vehicle speed, km/hr
nw: number of wheels
Prec: ft of days/yr precip>.254 mm
Limp: length of impacted roadway, m
Nvp: number of vehicle passes/day
Nwd: number of working days/yr
CEFR: roadway control eff. factor
CDF: contaminant dilution factor
ASH: ash generated, kg/d
NDTL: daily truck loads
fe: time wind speed > 5.4 m/sec, %
TSA: truck surface area, m2
LOR: road length of truck emiss, m
Sa: silt content of ash, %
CEFT: truck control eff. factor
kunl: particle size multiplier
Ms: ash moisture content, %
1
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
2
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
4
0.36
13.4
27000
24
10
121
100
50
365
0.10
0.10
330000
8
100
40
600
6.7
0.10
0.35
0.25
Chapter 5
sections
6.2.1
6.2.1
6.2.1
6.2.1
6.2.1
6.2.1
6.2.1
6.2.1
6.2.1
6.2.1
6.2.1
6.2.2
6.2.2
6.2.2
6.2.2
6.2.2
6.2.2
6.2.2
6.2.3
6.2.3
(continued on the following page)
C-7
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table C-1. (continued)
Source category3
Description13 1
4. Air modeling parameters (continued)
ksc: particle size multiplier for NA
spreading/compacting
Dspr: depth of spread, m NA
TVL: truck volume, m3 NA
Chapter 5
234 sections
NA NA 0.21 6.2.4
NA NA 0.05 6.2.4
NA NA 40 6.2.4
5. Bioconcentration and biotransfer
parameters
f|ipid: fish lipid fraction
FDW: fresh to dry wt conversion
Vp: particle deposition vel, m/yr
Ygr: yield of grass, kg/m2 dry
INTG: grass intercept fraction
Yfod: cattle fodder yield, kg/m2
INTfod: fodder intercept fraction
Yveg: vegetable yield, kg/m2 dry
INTveg: vegetable intercept frac.
VGbg: below ground veg. correc. fact.
0.07
0.15
315360
0.15
0.35
0.45
0.73
1.17
0.48
0.01
0.07
0.15
315360
0.15
0.35
0.45
0.73
1.17
0.48
0.01
0.07
0.15
NA
0.15
0.35
0.45
0.73
1.17
0.48
0.01
0.07
0.15
315360
0.15
0.35
0.45
0.73
1.17
0.48
0.01
3.4.1
3.4.2
3.4.2
3.4.2
3.4.2
3.4.2
3.4.2
3.4.2
3.4.2
3.4.2
(continued on the following page)
C-8
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table C-1. (continued)
Source category8
Description13
5. Bioconcentration and biotransfer
parameters (cont'd)
VGveg: veg/fuit air-to-leaf corr.
VGgr: grass air-to-leaf correction
VGfod: fodder air-to-leaf corr.
SDFb: beef cattle soil diet frac.
SDFd: dairy cattle soil diet frac.
FDFb: beef cattle fodder diet frac.
FDFd: dairy cattle fodder diet frac.
GDFb: beef cattle grass diet frac.
GDFd: dairy cattle grass diet frac.
BCGRA: fraction beef cattle
grazing on contaminated land
DCGRA: fraction dairy cattle
grazing on contaminated land
BCFOD: fraction beef cattle
fodder grown on contaminated land
DCFOD: fraction dairy cattle
fodder grown on contaminated land
Bs: soil bioavailability of cont.
1
0.01
1.00
0.50
0.08
0.02
0.02
0.90
0.90
0.08
1.00
1.00
1.00
1.00
0.65
2
0.01
1.00
0.50
0.08
0.02
0.02
0.90
0.90
0.08
1.00
1.00
1.00
1.00
0.65
3
0.01
1.00
0.50
.08
0.02
0.02
0.90
0.90
0.08
1.00
1.00
1.00
1.00
0.65
4
0.01
1.00
0.50
0.08
0.02
0.02
0.90
0.90
0.08
1.00
1.00
1.00
1.00
0.65
Chapter 5
sections
3.4.2
3.4.2
3.4.2
3.4.3
3.4.3
3.4.3
3.4.3
3.4.3
3.4.3
3.4.3
3.4.3
3.4.3
3.4.3
3.4.3
(continued on the following page)
C-9
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table C-1. (continued)
aSource Categories: 1 =on-site soil, 2= off-site soil, 3= incinerator stack emissions, and 4 = incinerator ash disposed of in landfill.
bThe parameter names are the ones on the spreadsheets. Most often they match the names given in the referenced section in Chapter 5.
In some instances, however, they do not match exactly or the discussions in Chapter did not assign specific parameter names—for
example, the fraction of cattle grazing which occurs on contaminated land was not a named parameter in Chapter 5.
NA = not applicable; parameter was not required for noted spreadsheet.
C-10 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table C-2. Chemical parameters for the three example compounds for all settings
Description
2,3,7,8-
TCDD
2,3,4,7,8-
PCDF
2,3,3',4,4',5,5'-
HPCB
Chapter
section/table
I. Physical/chemical properties
MW: molecular weight, g/m
log Kowa: log octanol water partition coeff, L/kg
H: Henry's Constant, atm-m3/mole"5
Da: molecular diffusivity in air, cm2/s
322
6.64
1.65xlO-5
.05
340
6.92
4.99MO'6
.05
396
7.71
1.00*10'3
.05
Table A-1
2.2
2.2
5.3.2
II. Estimated parameters
Koc: organic carbon partition coeff., L/kg
Bva: Air-to-Leaf transfer factor
F: beef/milk bioconcentration factor
BSAF: lipid-based biota sediment ace. factor
k: dissipation rate constant for eroding
or depositing contaminants, yr"1
kw: first-order plant wash-off constant, yr"1
RCF: Root bioconcentration factor
2691500
640223
5.0
0.09
0.0693
18.02
3916
5128600
4033804
3.0
0.09
0.0693
18.02
6432
31622800
124112
0.5
2.00
0.0693
18.02
26103
5.3.1
5.3.4.2
5.3.4.3
5.3.4.1
5.4.1
5.3.4.2
5.3.4.2
"log Kow not strictly needed for any model estimations, but it was used in the estimation of several parameters, and it is
otherwise a critical compound characteristic.
(continued on the following page)
C-11
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table C-3. Summary of exposure pathway parameters
Pathway
description
Soil ingestion
Central
High end
Soil dermal contact
Central
High end
Dermal Absorption
Contact
rates
200 mg/d
800 mg/d
0.2 mg/cm2-event
5000 cm2, 40 events/yr
1 .0 mg/cm2-event
1000 cm2, 350 events/yr
0.03
Contact
fractions
1.0
1.0
1.0
1.0
Chapter
section
7.2.1
7.2.2
Vapor/dust inhalation
Central
High end
Water ingestion
Central
High end
Beef fat ingestion
Central
High end
Milk fat ingestion
Central
High end
Fish ingestion
Central
High end
20 m3/day
20 m3/day
1.4 L/day
1.4 L/day
NA
22 g/day
NA
10.5 g/day
1.2 g/day
4.1 g/day
0.75
0.90
0.75
0.90
NA
0.44
NA
0.40
1.00
1.00
7.2.3
7.2.4
7.2.5
7.2.5
7.2.6
(continued on the following page)
C-12
7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
Table C-3. (continued)
Pathway Contact Contact Chapter
description rates fractions section
Fruit ingestion
Central 0.20 7.2.7
Above Ground 88 g/day
Below Ground 0 g/day
High end 0.30
Above Ground 88 g/day
Below Ground 0 g/day
Vegetable ingestion
Central 0.25 7.2.7
Above Ground 76 g/day
Below Ground 28 g/day
High end 0.40
Above Ground 76 g/day
Below Ground 28 g/day
Exposure Duration: A duration of 9 years was assumed for central exposures, and a
duration of 20 years was selected for the high end exposures. The soil ingestion pathway
was demonstrated only for children and the exposure duration was 5 years. See Section
7.2
Body Weight/Lifetime: The standard assumptions of a 70 kg adult and 70 years lifetime
were assumed for all pathways, except that of soil ingestion. In that case, a child body
weight of 17 kg was used. See Section 7.2
C-13 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
APPENDIX D. MODELING THE IMPACT OF EFFLUENT DISCHARGES INTO SURFACE
WATERS
Dioxin-like compounds have been found in the effluent discharges from various
industries such as pulp and paper mills where chlorine bleaching processes are used. This
potential source for releasing dioxins to the environment was not considered earlier in this
document since it has been addressed in other recent Agency publications. However, in
response to review comments, a brief discussion is included here summarizing possible
approaches to evaluating this source.
In 1988, EPA and the paper mill industry cooperated on a study which examined
discharges from five pulp and paper mills (EPA, 1988). Among other findings, that study
concluded that: the principal dioxin-like compounds in effluent discharges were 2,3,7,8-
TCDF and 2,3,7,8-TCDD, that 2,3,7,8-TCDF concentrations exceeded those of 2,3,7,8-
TCDD by 2 to 18 times, and that once these compounds are formed, they are not altered
in further processing or wastewater treatment (EPA, 1988). A follow-up study (called the
"104-mill study") collected data on 2,3,7,8-TCDD and 2,3,7,8-TCDF effluent discharges
from 87 kraft pulp mills and 17 sulfite pulp mills in the United States. Further details on
this 104-mill EPA/Paper Industry Cooperative Dioxin Study can be found in EPA (1990a;
1990b).
EPA (1990b) used data from the 104-mill study to evaluate the impact of 2,3,7,8-
TCDD and 2,3,7,8-TCDF discharges to receiving water, fish, and to humans through
consumption of drinking water and fish. Two approaches were used to estimate the
impact to water and fish. One was to use a simple dilution model to estimate total
(soluble plus sorbed) water concentrations after the effluent mixes with the receiving
water (i.e., at some point near but not directly at the point of effluent discharge). These
total water concentrations were assumed to be bioavailable to fish, and also were used as
concentrations for drinking water exposures. This approach was identified as representing
the upper bound for bioaccumulation in fish, since the bioconcentration factors used were
based on dissolved contaminant concentrations, not total concentrations. The second was
to use the EXAMS II model which allowed for partitioning of these hydrophobic
contaminants between suspended and benthic sediments, and dissolved in water. The
same bioconcentration factors were used to estimate fish tissue concentration, but they
D-1 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
were applied to the dissolved phase concentration in the water column. EPA (1990b)
describes all other assumptions and results of this exercise.
EPA (1992) also used the EXAMS II model in a document which describes a
methodology to evaluate the impact of pulp and paper mill discharges on receiving water
bodies, fish, and humans that consume the fish. The use of the EXAMS II modeling in this
methodology document was different than the exercises described above. One key
objective of this methodology was to identify a "Geographic Area of Potential Population
Exposure", or GAE. Identification of a GAE begins with analysis of fish tissue data. This
can be existing data downstream from an effluent discharge, or data specifically collected
in order to evaluate the impact of the effluent discharges. Given fish tissue
concentrations, a BCF can be used to backcalculate a water concentration that
corresponds to this fish concentration. Then, the EXAMS II model can be used to estimate
how far from the discharge point these water concentrations would occur. This distance
is a principal point of definition of the GAE. This distance would vary depending on the
fish concentration used - it could be the mean, median, or minimum concentration found in
the fish tissue sampling program. Other site-specific characteristics of the water body in
question, such as the distance fisherman will travel to fish in the water body, further
define the GAE.
The approaches described above model dioxin transport in the water phase and
estimate fish tissue concentrations from water concentrations using a BCF. An alternative
method is to model the transport in the sediment phase and use a Bioavailability Index or
Bl to estimate fish concentrations.
Such models would require as input effluent discharge characterizations, water
body characterizations, and would predict downstream sediment impacts. The input data
pertaining to many specific water bodies is currently available through data bases such as
STORET, a water quality data base maintained by EPA; GAGE, a water quantity data base
supported by USGS and accessible through STORET; and the REACH file within GAGE,
which gives flow data for specific reaches within rivers. Effluent discharge data for
2,3,7,8-TCDD and 2,3,7,8-TCDF are summarized for the 104-mill study in Appendix C of
EPA (1990b). Although qualified as only a 5-day temporal sample from the 104 mills, this
does provide for all mills: total suspended sediment concentrations in effluent discharges
(mg/L), effluent flow rate (MGD), 2,3,7,8-TCDD and 2,3,7,8-TCDF total concentrations
D-2 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
(ppq total concentration) and total load (kg/hr), and receiving water body harmonic mean
flow and 7Q10 low flow (m3/hr).
This data could be used in a simplistic sediment dilution model similar to that used
in this assessment to estimate bottom and suspended sediment concentrations resulting
from erosion of contaminated and uncontaminated land within a watershed.
Specifically, the effluent data allows one to estimate the input of contaminated
sediment to the river by this effluent stream. River flows and information about the
watershed draining into the water body allow one to estimate the input of residue-free
eroded sediment, or sediment with a known or estimated concentration of 2,3,7,8-TCDD
or 2,3,7,8-TCDF. The following summarizes an example of a sediment dilution approach
in a four-step procedure. This example is based on the first entry in Table C.1 (Appendix
C of EPA (1990b)):
• Estimate the concentration of TCDD on effluent suspended sediment discharges:
First, a reasonable assumption to make is that all the discharged 2,3,7,8-TCDD is
associated with suspended sediment in the effluent discharge. The effluent is discharged
at a rate of 23 MGD, it has a total suspended sediments concentration of 248.9 mg/L, and
a total 2,3,7,8-TCDD concentration of 0.0068 pg/L (ppt, 6.8 ppq). The TCDD load to the
river is also listed, but can be calculated from the concentration and discharge rate, as
2.5x10~8 kg/hr. A concentration on that sediment can be estimated by first estimating the
rate of discharge of sediments, which for this example is 903 kg/hr, and then inverting and
multiplying by the 2,3,7,8-TCDD discharge rate. The resulting concentration, (2.5x10~8 kg
TCDD/hr)/(903 kg sus. sed/hr), equals 28 ng/kg (ppt).
• Estimate the total sediment load contributed by the effluent discharge: A total
sediment load to the river in this example equals 7.9x106 kg/yr (903 kg/hr * 24 hr/day *
365 day/yr). The simple sediment dilution model assumes that this will mix with all other
sediment inputs to the river which occur downstream of the effluent discharge point.
• Estimate the total sediment load contributed by other sources to the water body:
The example scenarios in Chapter 9 assumed a rural setting and that erosion from the
watershed contributed mixtures of residue-free and contaminated soils. Further, the
D-3 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
examples assumed that the contaminated land area was at the top of watershed; hence,
the contributions from the contaminated site to the river mixed with the contributions from
the watershed as a whole. The same assumption will be used for this example. Site-
specific information on the watershed size and erodibility can be obtained, but
assumptions will be made for this example. First, the size of the receiving water body can
be used to roughly estimate the watershed size, if that information is unavailable, as it is
for this example. The harmonic mean flow of the receiving water body is given as
184716 m3/hr, which converts to 1811 ft3/sec. An average runoff rate over a land area
can be translated to a stream flow rate. It was shown in Chapter 10, Section 10.2.4.2,
that an average runoff rate of 1 5 inches/year over a 10,000 acre watershed - the size of
the watershed in the example scenarios - translates to a stream flow of 17.2 ft3/sec. The
much higher flow rate of 1811 ft3/sec backcalculates to a watershed size just over
1,000,000 acres (assuming 15 in/year runoff). Next, an erosion rate is needed, and a rate
of 6 T/ac-yr will be assumed for this example - this was also the average watershed
erosion rate used in the example scenarios. Finally, a sediment delivery ratio will reduce
the total erosion to an amount which reaches the river. From Figure 5.1, a drainage area
of 1,000,000 (which equals just over 4000 km2) corresponds to a delivery ratio of about
0.04. The total sediment contributed by overland erosion from a 1,000,000 acre
watershed is: (1,000,000 ac) * (6 T/ac-yr) * 0.04 * 906 kg/T = 2.2 x 108 kg/yr.
• Determine the sediment concentrations using a simple dilution model: This
can be estimated using a dilution ratio approach similar to the approach given in Chapter 5,
Section 5.3.1:
'sed
Cw SEDW
SEDW
where:
Csed = bottom and suspended sediment concentration, ppt
Ceff = effluent sediment concentration, ppt
SEDeff = sediment contribution by effluent discharge, kg/yr
Cw = average concentration of dioxin-like compound in soil eroding into
D-4 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
water body, ppt
SEDW = sediment contribution from watershed, kg/yr
From the example discussed above, Ceff = 28 ppt 2,3,7,8-TCDD, SEDeff = 7.9x106
kg/yr, Cw can be assumed to be 0.0 which would estimate the increment of sediment
concentration due to the discharge, and SEDW = 2.2x108 kg/yr. Csed is solved for as
0.97, or 1 ppt. The dilution from pipe sediment concentrations to average river sediment
concentrations within the example watershed is 3.5% (1 ppt/28 ppt expressed in percent).
If a background watershed soil concentration were assumed to be 1 ppt, then the
sediment in the river is estimated as 2.0 ppt; i.e., the effluent discharge effectively
doubles the background sediment concentrations within the water body.
Conceptually this approach mirrors the one presented earlier in this document for
estimating the impact to a nearby water body from an area of contaminated land.
However, it could be argued that an effluent discharge, as a point source modeling
problem, is amenable to a more sophisticated approach as compared to nonpoint erosion
modeling. For effluent discharges, perfect mixing throughout a river system is clearly an
oversimplification - to a lesser extent, the same could be said for erosion from a relatively
small tract of contaminated land within a larger watershed. One empirical approach would
be to estimate the average concentration with this or a similar dilution approach, but then
assume a sediment concentration distribution around this average as a function of distance
from the pipe. Continuing the example above, instead of an average 3.5% dilution ratio
throughout the river, the dilution would be higher near the pipe discharge at 10 or 20%
and lower at great distances, say 1 % or less. The shape of such a distribution is unknown
at the current time.
Currently available models such as EXAMS II can be used to estimate bottom
sediment concentrations resulting from pipe discharges, although these equilibrium
partitioning models can also introduce uncertainties. Hanna, et al. (1992), using
relationships in Karichoff and Morris (1985), estimate that the time to equilibrium for
dioxin in a suspended sediment/water system is several months. Therefore, dioxin-like
compounds associated with suspended sediment in effluent discharges would likely reach
bottom sediments before they would equilibrate and partition partly into the water column
into a dissolved phase. The EXAMS II model assumes instantaneous equilibrium once
D-5 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
introduced into the water. It is not clear how much error is introduced with this
assumption of instantaneous equilibrium. Hanna, et al. (1992) also note that an important
parameter in an idealized "pipe-to-sediment" model is the size distribution of suspended
sediment particles in effluents. If they are predominantly large, for example, they may not
travel far from the point of discharge. Hanna, et al. (1992) further state that a generalized
pollutant fate and transport model "must include: convection/dispersion in receiving water;
particle sorption/desorption in sediment and water column; settling, resuspension, and
burial of particulates; diffusive exchange between water column, active sediments, and
deep sediments; pollutant sources via wet/dry deposition, point and nonpoint sources; and
sinks, e.g., volatilization, and chemical/biochemical reactions". Finally, Hanna, et al.
(1992) note that the fate and transport model could use a mixing zone analysis similar to
that of Holley and Jirka (1986). The problem with the analysis of Holley and Jirka (1986)
is that it is specific to totally soluble contaminants and would require empirical or
theoretical corrections for transport and fate of contaminants sorbed to sediments.
In summary, it appears that sophisticated "pipe-to-sediment" models have not yet
been well developed but may offer conceptual advantages over the classical water phase
based approaches.
REFERENCES
Hanna, L.M., E.J. Bouwer, and D.W. Smith. 1992. An Approach to Estimating Effluent
Limits for Hydrophobic Compounds Through Transport and Fate Considerations.
Summary of presentation made at a conference titled. National Conference on
Bioavailability of Dioxin, PCBs and Metals in Aquatic Ecosystems. Sponsored by BCM
Engineers, Inc. and Rifkin and Associates, May 14-15, Loews L'enfant Plaza Hotel,
Washington, D.C. Authors are from BCM Engineers, Inc., One Plymouth Meeting,
Plymouth Meeting, PA 19462.
Holley, E.R. and G.H. Jirka. 1986. "Mixing in Rivers". Technical Report E-86-11, US Army
Corp of Engineers, US Army Engineer Waterways Experiment Station, Vicksburg,
Mississippi.
Karickhoff, S.W. and K.R. Morris. 1985. Sorption Dynamics of Hydrophobic Pollutants in
Sediment Suspensions. Environmental Toxicology and Chemistry 4: 469-479.
U.S. EPA. 1988. U.S. EPA/Paper Industry Cooperative Dioxin Screening Study. Office of
Water Regulations and Standards, U.S. Environmental Protection Agency, Washington,
D.C. EPA-440/1-88-025. March, 1988.
D-6 7/31/92
-------
DRAFT-DO NOT QUOTE OR CITE
U.S. EPA. 1990a. Integrated Risk Assessment for Dioxins and Furans from Chloring
Bleaching in Pulp and Pepr Mills. Office of Toxic Substances, U.S. Environmental
Protection Agency, Washington, D.C. EPA 560/5-90-011.
U.S. EPA. 1990b. Risk Assessment for 2378-TCDD and 2378-TCDF Contaminated
Receiving Waters from U.S. Chlorine-Bleaching Pulp and Paper Mills. Prepared by Tetra
Tech, Inc., 10306 Eaton Place, Suite 340, Fairfax, VA 22030., Contract #68-C9-
0013.
U.S. EPA. 1992. A Methodology for Estimating Population Exposures from the Consumption
of Chemically Contaminated Fish. EPA/600/9-91/017 (Draft final as of 1/13/92).
U.S. GOVERNMENT PRINTING OFFICE: 1992
••1992-6 50 .938/50904
D-7 7/31/92
------- |